springload

Jun 03 2026

Tech Startups are hard: Fundraising is the hardest part

By Eric Picard

I’ve been getting a lot of outreach lately. Brilliant people, world-class engineers and research scientists, some of the smartest humans I’ve ever met, are reaching out about AI startups they want to build. Most of these folks have made real money in their careers. They’re self-funding the early work. And almost all of them are about to get a very rude awakening.

These folks are coming to me because they know I’ve done four startups, and that I’ve raised money before. Bluestreak, my first startup, raised $28MM between 1997 and 2001. I will tell you for certain that the seed funding was the hardest part. We raised $150,000 from local angels, and then $1.3MM from some professional angels. Then a Series A, Series B, and Series C. Every one of those raises was incredibly hard. Rare Crowds, my second startup, raised $750K and it was harder than any of the fundraising we did for Bluestreak.

My engineering and scientist friends know that building the technology is hard. They’re right. But they assume that’s the hard mountain to climb. It isn’t. Building the technology is the mountain you already know how to climb. Raising the money is a different mountain entirely, and brilliance in engineering and science does not transfer natively to climbing that mountain, it’s a whole new skillset to learn.

It isn’t about talent or ability. These are people who have never failed at anything in their professional lives. They excelled academically, landed elite research roles or senior engineering positions, built things other people couldn’t. That track record makes it very hard to see what’s coming. Startups are harder than anything they’ve done before, and fundraising is a completely different skillset than building technology. If I had to bet on the one thing that will most surprise them, it’s that.

Raising money is selling a piece of your company

Fundraising is selling a stake in your company to a stranger who has no reason to trust you, based on a pitch about an unproven future. You are asking them to take a risk, and your job is to mitigate that risk in their mind with three things: a genuinely innovative idea, a technology problem that is hard enough to defend (what investors call a moat), and evidence that your team is the one that will win.

That last one matters more than most first-time founders expect. Investors invest in teams as much as they invest in ideas. Most companies pivot multiple times. The idea you’re pitching today may not be the business you’re running in three years. Experienced investors know this, and they’re betting on whether you and your team can navigate the chaos of a startup, not just whether the technology works as advertised. And it isn’t just one member of the team, they’re looking for a strong team that works well together.

So you’re not just pitching a product. You’re selling yourself, your team, your judgment, and your ability to execute under conditions that are nothing like the well-resourced environments most technical founders have thrived in. Brilliant engineers trying to raise money are not expert at fundraising until they have a lot of experience doing it. And most of them are starting from zero.

This also goes for product managers reading this. You’re a hybrid, part business and part technology, and it’s tempting to assume that gives you an edge in the room. Maybe. Maybe not. Don’t walk in overconfident. I’ve seen product people get just as humbled as their engineering co-founders, because pitching investors is its own craft, and it doesn’t care about your functional background. It’s much closer to sales than engineering.

The numbers, and why they should reset your expectations

This has been studied carefully, and the findings are sobering.

DocSend, working with Harvard Business School, analyzed hundreds of startups across multiple years. The headline finding for a successful seed round: founders contacted an average of 58 investors and held around 30 to 40 meetings before closing, over roughly 12.5 weeks. By 2023, the average was 66 investors contacted with about 38 meetings set. Series A is different. By then your track record filters the funnel, and founders contacted an average of 26 investors rather than 58.

Nathan Beckord of Foundersuite, who has run many raises, puts the conversion rate plainly: he pitched more than 200 investors and landed one seed fund and 10 angels, a roughly 5% conversion rate, and calls that “pretty common.” The 2026 benchmarks from Founder Institute tell founders to prepare to pitch 100 to 200 investors.

Here is the actual funnel, using the data as reported:

You contact roughly 58 to 66 investors to run a seed round.
About half convert to meetings, roughly 30 to 38.
Roughly 3 to 5 write a check, around 5% of the people you contacted.

Two ratios to internalize: about half your outreach should turn into meetings, and about 1 in 10 of those meetings turns into money.

The founder who takes 8 or 10 meetings and concludes the market is closed has not run the funnel. They haven’t finished yet.

Meetings matter more than outreach volume

The insight buried in the DocSend data that changes how you should think about this: it’s not the number of investors you contact that predicts success. It’s the number of meetings you get.

In one cohort, founders who didn’t get funded actually contacted slightly more investors than those who did, 77 versus 70, but landed far fewer meetings, averaging just 15. The funded founders got roughly twice as many meetings from roughly the same outreach.

This means meetings are the signal to optimize for, and not just because more meetings mathematically means more chances at a check. Each meeting is a rep. You get feedback, you adjust the pitch, you sharpen the story. Later pitches with a more refined narrative, and with more evidence of traction as you build, are meaningfully more likely to succeed than your first few. The pitch you give in month one is not the pitch you give in month three, and it shouldn’t be.

Not all ideas are fundable as standalone companies. Some things that are technically impressive are better as a feature inside a large platform, or as a services business, and experienced investors recognize that immediately. If you’re pitching a fundable idea and getting meetings but not money, your pitch is the problem. If you can’t even get the meetings, the targeting or the narrative is the problem. Either way, the fix is more reps and more feedback, not just more outreach.

Know who you’re pitching

This is where a lot of first-time founders waste months and burn their best opportunities.

Venture capital firms, the branded ones you’ve heard of, primarily invest in later seed rounds and Series A, and some focus on later stage (Series B and C) rounds. If you’re connected, some will take your meeting, especially if the technology is interesting. But the most likely outcome is a genuine, friendly conversation that ends with, “This is exciting, come back when you’re further along.” That is not a rejection. That is an investor telling you what milestone will get them in. Keep the relationship warm. It has real long-term value. But it does not put money in the bank this quarter.

For pre-seed and early seed funding, the capital you need to build something worth pitching to a VC, your target market is angel investors. Specifically angels who invest at your stage and in your sector. Research them. Find out what they’ve funded, what stage they enter at, what check sizes they write. This is sales 101 applied to fundraising: know your customer before you walk in the door.

One other misunderstanding that’s worth calling out: If they didn’t say yes, they really said no. A polite, super interested conversation isn’t a yes. A promise to bring your idea to their next review meeting with their partners is a win. But it’s not a yes, and it relies on that partner being able to articulate your pitch (note: your leave-behind materials are important.) So individual investors are easier, they can make the decision on their own without convincing anyone else. You’re better off getting a fast “no” instead of a slow “no.” Don’t be discouraged by the fast “no”, be happy it was fast. Ask for clarity on why they said “no”, and hope to glean input that can help hone your pitch for the next investor.

The problem of overfitting on success stories

Exceptional people are more susceptible to this trap than most, because their own lives keep providing the evidence.

Jim Carrey famously told graduates to take risks and follow their dreams. He’s living proof that it worked. What he can’t tell you about is the thousands of genuinely talented comedians and actors who took the same risks, worked just as hard, and didn’t make it, because we’ve never heard of them. Talent matters. Jim Carrey is exceptionally talented. He was also very lucky, and luck doesn’t scale to everyone who deserves it.

Here’s the piece that gets underappreciated: talented actors trying to get cast in roles are not expert at auditioning until they’ve done it hundreds of times. The audition is a separate skill from the craft. Talented engineers trying to raise money are not expert at fundraising until they have a lot of experience doing it. Experience tips the scale. Pitch a lot.

The standout AI startup raises you’re reading about, $20M at a $200M valuation pre-revenue, are the Jim Carrey story. They are not the base rate. They are the outcomes you read about because the others didn’t make the news.

This matters for how you plan. Your talent is real. Your odds are better than average because of it. They are still hard.

Product-market fit is not a buzzword

You’ll hear “product-market fit” constantly. There’s a reason. Tuning what you’re building to what the market actually wants, or can be convinced it wants, is one of the hardest and most time-consuming problems in building a company. Most early startups are wrong about this at least once before they figure it out. Most are wrong more than once. Success isn’t about betting once, it’s about betting, learning, and betting smarter the next time.

Investors who get in ahead of product-market fit are betting almost entirely on the team. If you can show early evidence of fit, even a handful of paying customers, even a waiting list with real signal, you dramatically increase your chances. Every proof point you can put in front of an investor reduces the risk they’re taking on, and risk reduction is the whole job of the pitch.

It takes a team with all the right skills

I want to make a case for something that brilliant technical founders are often the most resistant to hearing.

The best salespeople and business people are not less intelligent than engineers and scientists. They’re differently intelligent. Building relationships, reading a room, navigating objections, understanding what someone needs to hear before they’ll commit, these require a kind of emotional intelligence that is just as rare and just as valuable as the ability to solve hard technical problems. High EQ is a real capability, and in a fundraise it may matter more than high IQ. The two are not in competition. They’re complements.

The great ones are comfortable with rejection in a way that most technical people aren’t wired to be. They think in pipeline. They treat objection-handling as a craft. They genuinely enjoy the human dynamics of closing a deal. Finding someone great at this is as hard as finding a great engineer. They’re just hard in different ways.

Building a founding team is not a hiring problem. It’s the most important set of relationships your company will ever have. But if your founding team is all technical depth and no sales or business instinct, you are going into the hardest sales process of your life, raising money, without anyone on the team who has ever done it before.

Think hard about that before you start pitching.

Let’s talk about the roles and the real job description from the lens of a startup. I’m going to use C-level titles, but you can extrapolate. And remember, a 5 person company with 5 C-level titles isn’t necessarily a good idea or a bad idea.

The real job descriptions

CEO: Your job is to make sure the company has enough money to survive. Period. You’re on the hook to make payroll. To ensure that the AWS or GCP, Anthropic or OpenAI, BigQuery or Snowflake bills are paid. Your job is to make sure the company lives through the startup experience. This means you own the investor relationship, and you need to actually own it. It’s one of your many full-time jobs. Even if you’re a technical CEO, or if you’re a product CEO – your primary job is making sure the company survives to continue – which means you own the investor relationships.

CTO: Build the thing. Innovate enough to differentiate. But build the thing. Fast. Take the right shortcuts, you’re not at a F500 company. Amazon, Microsoft, Google – they all have early-days stories about technical risks and shortcuts they took. Often laughed about, as in, “I can’t believe we had $5M in sales running on a server under Jim’s desk. Remember the time the cleaning person unplugged it?” Make sure your shortcuts are the right shortcuts, not the wrong ones. Make sure they’re survivable.

Chief Product Officer: Figure out the thing that will find product-market fit. Talk to the market. Develop a hypothesis. Test and retest it. Apply learnings. Convince the CTO to make the changes. Package the idea so it can be sold. Startups overfit against invention and building, and underfit against “Go-to-Market.” Don’t forget that great products fail all the time, and weaker products win all the time because they were brought to market better. Listen to your sales inputs on how the pitch was received, and tune both the product and the pitch!

Chief Client Officer: Make any customers happy at almost any cost. Winning a customer is super hard. Keeping that customer happy is also super hard. Keep them happy, churn is not something an early stage startup can weather as easily as a late stage startup. Make up for product shortcomings with human effort, but make sure it’s not permanent.

CRO: Build a pipeline, and convert. Do it with investors. Do it with customers. Hone the GTM pitch on both decks, learn and adapt. But close at all costs. Revenue does two things: It funds the company to offset burn rate, and it proves to investors that you’re on a path to product-market fit and ultimately to success. Be methodical, analytical, and get to 10% week-over-week and month-over month success metrics (or better!)

Why the CRO matters more than technical founders think: The right founding team for most successful technical startups includes someone who lives for the close, isn’t wounded by the no, and wakes up thinking about how to move people from interested to committed. That person needs the engineer as much as the engineer needs them. The technology without the ability to sell it is a project. The ability to sell without something genuinely hard and defensible behind it is a dead end. Together, you have a company.

Fundraising is a grind. It takes months, hundreds of contacts, dozens of meetings, and more patience than most people who’ve never struggled professionally have had to develop. The technology problem is hard. The fundraising problem is harder in a completely different way. Go in knowing that, prepare for it like the discipline it is, and find the people around you who are as talented at selling as you are at building.

You’re going to need each other.

Tagged business, entrepreneurship, startup, Startups, technology

May 19 2026

Leave a comment

Uncategorized

The AI bill is coming due

By Eric Picard

A story is making the rounds in business circles. AI is being sold to us at investor subsidized prices because the model labs are losing money on every prompt. When the venture capital subsidy ends, prices will jump several-fold, and any company that bet big on AI will find itself holding an existential bag. Adopt extensively, the argument goes, and you put your business at risk.

It’s not entirely wrong. But it points at the wrong problem, and it misses where the real exposure sits. Let’s work through what’s happening.

The cost curve has been collapsing

The cost to run a model with GPT-4-level intelligence has dropped roughly 50x in three years. At GPT-4’s launch in March 2023, that capability cost about $20 per million tokens. Today it costs $0.40 or less. Pull back further and the picture is even more dramatic: for GPT-3-level intelligence, the cost has fallen roughly 1,000x since 2021, going from around $60 per million tokens to about $0.06. Either way you slice it, this is one of the fastest cost declines in the history of computing — faster than anything Moore’s Law ever produced.

That decline came from four compounding sources, each contributing across multiple model generations: better hardware, better inference software, better model architectures like Mixture of Experts, and quantization techniques that run models at lower numerical precision. Each of these has delivered roughly 2-4x improvements per generation, and they stack — across several generations of each, you get the orders of magnitude we’ve actually seen.

Worth noticing what’s not on that list: Moore’s Law and quantum computing. Quantum has no demonstrated path to running AI inference cheaper than GPUs in this decade, and probably not the next. If your optimism is anchored on quantum, you’re betting on a horse that hasn’t entered the race. The horses running today — inference-optimized silicon, sparse architectures, aggressive quantization, semantic caching, model routing — are real, shipping, and doing the work.

The labs are figuring it out faster than people realize

OpenAI generated about $20 billion in revenue in 2025 and is projected to burn $17 billion in cash in 2026. Their gross margin sits in the 33-48% range depending on whose leak you trust. Those numbers feed the doomer story. But they describe one company’s strategy, not the technology’s economics.

The more informative number is Anthropic’s gross margin trajectory: negative 94% in 2024, roughly 40% in 2025 (revised down from an original 50% projection when inference costs came in 23% higher than expected), and a target of 77% by 2028. Anthropic figured out early they’d rather sell to enterprises than subsidize a billion free consumer users. Roughly 80% of their revenue now comes from business customers, and that mix is what makes the path to 77% margins plausible at all. OpenAI is in a stickier spot because they built a consumer brand they can’t easily walk back from. You can’t tell 850 million free users “actually, that’ll be $30 a month now” without detonating the moat that scale was supposed to create.

When people say “AI is unprofitable,” what they usually mean is “OpenAI’s consumer business is unprofitable.” Different statement.

The real exposure is somewhere else entirely

If frontier model prices double or triple at the top tier, it doesn’t matter much for most businesses. Roughly 90% of real-world AI tasks — summarization, classification, drafting, retrieval, basic analysis — don’t require the smartest model available. They run fine on tier-two cloud models or open-weight models that now lag the frontier by about three months on standard benchmarks. That gap is shrinking. A sensible enterprise with disciplined model routing finds a price hike on Opus or GPT-5 annoying, not existential.

The exposure sits in usage patterns, not pricing.

Cost per token has dropped 50x at the frontier and 1,000x at older tiers. But agentic workflows — autonomous systems that reason in loops, call tools, and self-correct — consume 5 to 30 times more tokens per task than a single chatbot query. Cheaper inference doesn’t mean lower bills when usage is exploding. This is Jevons paradox in real time: a 19th century observation by economist William Stanley Jevons that improving steam engine fuel efficiency didn’t reduce coal consumption — it increased it, because cheaper coal opened up uses that hadn’t been economical before. When something gets cheaper, we use vastly more of it, and total spend goes up.

The companies blowing past their AI budgets aren’t doing so because the labs ended a subsidy. They’re doing so because they wired Claude Opus or GPT-5 into every step of every agentic workflow without asking whether each step required a frontier model. They got it working with the smartest model available, shipped it, and never went back to optimize.

What this means for business leaders

The right question isn’t “what’s our spend on AI?” It’s “what’s our cost per resolved task, and how does it compare to the human-equivalent cost?”

If you’re spending $4 per AI-resolved customer service ticket and a human agent could resolve it for $3, you have a zombie agent draining your quarterly budget. The wrong response is to panic about AI cost. The right response is to route 80% of those tickets to a small fast model that handles them for ten cents each, and reserve the frontier model for the 20% of cases that genuinely require it.

This discipline has a name: FinOps for AI. It’s the same discipline mature cloud-native companies built ten years ago for AWS, GCP, and Azure spend. Tag every model call. Make the cost visible to the developer making the architectural choice. Build dashboards that show cost per task type, escalation rate to the expensive model, and trend over time.

When developers can see their choices reflected in cost, they make better choices. When they can’t, they don’t.

The architectural piece matters too. Don’t build a system where every call goes to a frontier model just because frontier was easiest to wire up during the prototype. Build tiered systems. Small fast models for routine work, mid-tier for the harder stuff, frontier only for what genuinely requires it. Add semantic caching so you’re not paying twice to generate the same answer for slightly differently worded queries.

This isn’t speculative. The companies running AI profitably already operate this way.

The cost curve is real and compounding fast. The labs are figuring out unit economics in front of us, not collapsing under them. We don’t need quantum or Moore’s Law to keep delivering — we need MoE, quantization, and inference ASICs to keep doing what they’re already doing.

The companies that will get hurt aren’t the ones that adopted AI extensively. They’re the ones that adopted it naively — without the discipline to route workloads, cache aggressively, and meter every model call. The companies that will thrive are treating AI compute the way mature companies have treated cloud infrastructure for a decade: as a real cost center that needs real engineering discipline, not a magic productivity tool that’s free until further notice.

The AI bill is coming due. It’s coming due for the people who didn’t read it.

Tagged AI, Artificial Intelligence, chatgpt, llm, technology

Oct 06 2025

Leave a comment

Uncategorized

Not getting left behind: AI business transformation

By Eric Picard

Here’s something that people are missing about the changes coming from AI. You can’t just treat AI as a productivity tool. You can’t just use it to speed up document creation, you have to turn it into a thinking tool, a tool to make your ideation and decision-making better.

If the CEO and the rest of the leadership team are not treating AI as a core capability across all aspects of the organization, their company is going to be standing still while competitors run by. So what do I mean by “core capability across all aspects of the organization?”

AI used correctly creates a workbench of hyper-intelligent advisors who can augment the intelligence of the leadership team – especially the CEO – so that they can make better decisions in less time. It needs to come from the top down, and it needs to be mandated. And not delegated to the IT team, whose job is to get the best price with the highest uptime with the lowest number of support tickets (great for many things but bad when trying to do business transformation.) Companies need to keep AI adoption at very senior levels abstracted from the IT team. And must be willing to live with the friction this causes.

In my practice I’m seeing that when organizationally, the leadership team starts using AI tools in their daily workflow, ingraining them into their habits, the whole org starts moving faster. It’s critical that AI is used not only for document creation (boy we’ve really sped up writing – what a productivity boost!) , but for ideation (our ideas are now battle tested and pushed around and stronger before we start executing!) that things really change. It’s critical that we’re clear – this is not abdicating their ideation to the AI, but bouncing ideas off the AI, getting feedback through a variety of points of view.

If you don’t understand what I’m talking about, try this simple experiment:

Create several specific personas – Here are some basic prompts that will help you try this out:

World Class CEO: You are a world class CEO with 25 years of leadership experience in the _______ industry. You are now retired, but mentoring other CEOs and leadership teams. You are cranky – you don’t glad-hand your mentees, you don’t tell them what they want to hear. You are inherently kind, you don’t argue for the sake of arguing, but you don’t hold back when you see mistakes being made. Your point of view is that _______________ and ____________ are great leaders who you try to emulate. And ____________ and _________ are overrated.

World Class CFO: You are a world class CFO with 25 years of experience at a combination of large publicly traded companies and startups that went through rapid growth and went public under your leadership. You understand the needs of a growing business, as well as the concerns of a large publicly traded company with regulatory oversight in the following industries _______________, _______________, ________________.

Innovative Entrepreneur: You are a massively successful entrepreneur in the ________________ and ____________ industries. You’ve broken each of these industries previously by disrupting the status quo and driving incredible success doing things that nobody expected to be successful. You are seasoned enough that you understand the issue of inertia, and you are not afraid of experimenting and failing. You’ve been through 3 acquisitions, where you had to transition to roles within large companies and you understand how big companies differ from startups, and were surprisingly successful navigating these organizations. You’re somewhat brash, and you tell the truth regardless of who you’re talking to.

Finish these prompts off by putting in the industry and names (the names have to be well known figures) and copy/paste these into your AI tool of choice before starting to run ideas by them. You’ll be shocked at the outcome, and the difference of opinion you get from each of these.

Now imagine if you had a stable of 20 or 30 or 100 of these predefined prompts that you could pull out whenever you needed them. Imagine if you debated all your big decisions with a large group of experts with strong points of view. Do you think your ideas would come out the other side stronger or weaker?

Having done this a lot – I can tell you my experience. I often disagreed with the input I was getting, which was half the value. It helped me become clearer on my own point of view. I got feedback I didn’t like, but sometimes it was what I needed to hear. When I pulled in advice from “experts” with different skillsets than my own, I got really valuable expansions of my thinking. For instance – when I was CPO at Bark, I knew very little about supply chain when I started, and became pretty knowledgeable over the time I was there. Part of the way I did that was to create a supply-chain expert prompt who I could run my ideas by. Note – I also became pretty close with our leadership on supply chain and had weekly meetings with various folks to get smarter faster. But I could ask really dumb questions of my AI expert, as often as I wanted, and then take my refined understanding to those other meetings.

It’s like having a superpower. And anyone not doing this in 2 years is going to be left behind.

I happen to use an AI workbench called CharmIQ that makes this much easier. You create “charms” in this tool, which are saved prompts like the ones I describe above. You can assign these prompts to any LLM – CharmIQ comes with baked in access to all of them for one monthly subscription fee. It makes things much easier.

If you want to try it out, feel free to use this link, it gives you a discount and puts some change in my pocket. I will tell you I use this tool literally hours every day. It’s a game changer.

Click Here to Try CharmIQ.

Tagged AI, Artificial Intelligence, chatgpt, technology, Writing

Aug 21 2025

Leave a comment

Uncategorized

How AI is going to change software development

by Eric Picard

As someone who’s spent decades watching technology waves crash over the software industry, I find myself constantly recalibrating my predictions about AI’s impact on how we build software. Two years ago, I wrote about the potential for AI to create massive productivity gains and fundamentally alter development practices. Now, with some genuinely surprising developments happening right under our noses, I can see we’re heading somewhere much more interesting than I initially thought.

The productivity gains from AI tools like GitHub Copilot, ChatGPT, Cursor, and Replit are real, but what’s caught my attention is how we’re seeing two distinct paths emerge. Professional developers are using AI as incredibly powerful assistants, while non-developers are essentially managing teams of AI-powered junior developers through conversational interfaces. Both approaches are working, just for different types of projects and different scales of ambition.

Two Worlds of AI-Powered Development

Professional developers have largely embraced AI as sophisticated tooling that makes them dramatically more effective. They’re using AI for code generation, debugging assistance, refactoring, and rapid prototyping, but they’re doing it within established architectural patterns and development methodologies. The productivity gains are genuine – I’m consistently hearing reports of 30-50% improvements in specific tasks – but these developers maintain architectural oversight and code comprehension. I’ve talked to several developers who are often referred to as “10x” developers, meaning they’re ten times more effective than others on their teams. These particular developers were 10x devs at big tech companies, so certainly more than 10 times most developers. Each of them has told me that they are 100X-ing themselves using AI. So this is not a small change.

Then there’s what Andrej Karpathy coined “vibe coding” – developers describing what they want in plain English and accepting AI-generated code without necessarily understanding every line. This “fully giving into the vibes” approach emphasizes rapid experimentation over careful architectural planning. Y Combinator reported that 25% of their Winter 2025 batch had codebases that were 95% AI-generated, which represents a fundamentally different relationship with code creation.

The key insight is that these aren’t competing approaches – they’re serving different needs. Professional development teams use AI to accelerate work within proven frameworks, while vibe coding enables non-developers or small teams to build functional applications that would have been impossible for them to create just a few years ago.

Replit: DevOps Intelligence for the AI Era

Replit’s explosive growth – from $10M to $100M ARR in less than six months – illustrates something important about where this is heading. Replit is actually the intelligent evolution of DevOps principles, bringing continuous integration, automated testing, and deployment automation to AI-driven development.

Traditional DevOps emerged because managing infrastructure, deployment pipelines, and scaling manually was becoming impossible at scale. Smart development teams adopted CI/CD, automated testing, infrastructure as code, and monitoring because these practices made complex systems manageable and reliable.

Replit takes these same principles and makes them accessible through conversational interfaces. When their AI Agent selects a tech stack, generates code, sets up databases, and handles deployment, it’s not eliminating DevOps – it’s automating DevOps intelligence so that non-developers can benefit from these practices without needing to understand them deeply.

This matters because it’s expanding who can build functional software. You don’t need to understand Kubernetes, Docker, CI/CD pipelines, or infrastructure configuration to get the benefits of modern deployment practices. The AI handles the complexity while applying proven DevOps principles under the hood.

The Custom Application Renaissance

What excites me most about this trend is that we’re finally approaching the custom application future that the internet promised but never quite delivered. For twenty years, we’ve talked about having rich, customized web applications for internal business processes, but the development overhead made it impractical for most organizations.

Now we’re entering an era where custom web applications for fairly complex business tasks can be built quickly and cost-effectively. The “intranet” that organizations have wished for – dynamic, task-specific applications that actually solve their particular workflow problems – is becoming achievable. Just this week I built two very powerful internal apps for one of my clients inside of CharmIQ. These apps automate extremely intensive processes that were bottlenecks for my client. And three weeks ago, I had no idea how to do this. I’d have hired someone to build them.

I’m hearing about custom applications for everything from inventory management to customer onboarding to internal reporting, applications that would have required months of development and significant ongoing maintenance. These aren’t replacing enterprise software entirely, but they’re filling the gaps where off-the-shelf solutions don’t quite fit.

The Architecture Challenge Ahead

As these AI-powered development approaches mature, we’re approaching a fundamental architectural question. Current approaches work well for their respective use cases, but we need architectural patterns optimized for AI collaboration rather than just AI assistance. Right now AI developers are about as talented as junior developers – with maybe 2-3 years of experience. They break things a lot, the code isn’t efficient, they’re not always architecting things properly. But that’s a short-term problem – we’re only a few years away from AI developing software as well as any human, or better. What happens then?

The traditional monolithic applications or coarse-grained microservices that work well for human development teams may not be optimal for AI-powered development environments. Some teams experiment with treating code as completely disposable, letting AI regenerate implementations for each iteration. This works for prototypes and simple applications, but it breaks down for complex systems where you lose accumulated knowledge and performance optimizations.

Components: The Architecture for AI Collaboration

I think the future lies in component-based architectures that provide the right granularity for AI systems to work effectively. This draws inspiration from earlier component models like Microsoft’s COM objects, adapted for modern cloud environments and AI capabilities.

Applications would be built from well-defined components with stable interfaces and clear functional boundaries. Each component handles specific capabilities – user authentication, payment processing, data transformation, content generation – with explicit contracts for inputs and outputs. The critical insight is that while these interfaces remain stable, AI systems can continuously optimize, refactor, or completely reimplement the internal logic of individual components.

This architecture offers several advantages for AI-powered development. Components provide bounded problem spaces where AI systems can operate effectively without breaking broader system functionality. The stable interfaces enable comprehensive testing and debugging, while internal flexibility allows for continuous optimization based on performance data and changing requirements.

What This Looks Like in Practice

Over the next five to ten years, I expect we’ll see component registries emerge that catalog available functionality with detailed specifications. AI systems will continuously monitor component performance and generate optimized versions for testing and gradual deployment.

Applications will become more dynamic, automatically reconfiguring by swapping component implementations based on load patterns, user behavior, or resource availability. Unlike current microservices managed by human teams, these components would be maintained by AI systems operating within architectural guidelines defined by human engineers. And eventually, we may even let the AI take that over too.

The development process shifts toward interface-first design, where human architects focus on defining component boundaries and interactions, while AI systems handle implementation details. This division of labor plays to respective strengths: humans excel at architectural thinking and business requirements, while AI systems optimize implementations and handle routine coding tasks. And as the AI gets better and better at architecture and business requirements development, we may see a whole new world emerge.

The Transition Path Forward

This transformation is happening gradually but accelerating quickly. Professional developers are becoming more effective through AI assistance while maintaining architectural oversight. Non-developers are building functional applications through conversational interfaces that would have been impossible for them to create previously.

Current service-oriented architectures provide a foundation that can evolve toward component models as AI capabilities mature. Organizations with good interface design practices, comprehensive testing strategies, and strong observability will be best positioned for this transition.

The engineers who thrive will be those who can think architecturally about system design while effectively directing AI systems. Product managers become even more critical because rapid prototyping capabilities make clear product vision, competitive intelligence, customer-centric approaches and market understanding the primary competitive differentiators.

The Strategic Reality

As we move toward this future, competition shifts in important ways. Technical barriers to building certain types of software continue falling, but success increasingly depends on architectural excellence and product strategy rather than implementation speed alone.

Organizations that can design effective component architectures and orchestrate AI development systems will gain significant advantages in both development velocity and system reliability. The ability to continuously optimize software systems without traditional refactoring risks could become a major competitive edge.

However, this also presents new challenges around managing dynamic system complexity, ensuring security across AI-generated code, and maintaining coherent user experiences across rapidly evolving implementations.

The transformation isn’t about replacing human engineers – it’s about creating new collaboration models between human architectural thinking and AI implementation capabilities. The future belongs to organizations that can effectively combine these strengths while maintaining clear product vision and strategic focus.

We’re witnessing a shift from static implementations toward dynamic, continuously optimizing systems. While full realization is still years away, the foundation is being built through current experiments with AI-assisted development, vibe coding platforms, and component-based architectures. Replit’s growth numbers suggest this isn’t theoretical anymore – it’s happening faster than most of us expected, and the organizations preparing now will be best positioned to capitalize on the opportunities it creates.

Tagged AI, Artificial Intelligence, chatgpt, llm, technology

Mar 05 2025

Leave a comment

Uncategorized

AI Agents via CharmIQ have Supercharged my work

by Eric Picard

Today I’m exploring a tool that stands out in its ability to harness AI potential in ways I haven’t seen before—CharmIQ. This article discusses the capabilities of CharmIQ and is itself crafted using the platform. I’ve spent over twenty-five years building products and leading teams. I’ve witnessed technology evolve from multiple perspectives, helping guide both startups and large corporations through innovation challenges. This experience gives me a nuanced view on tools that genuinely transform how we work.

The artificial intelligence landscape, especially regarding large language models (LLMs), continues to expand rapidly. Each model offers distinct strengths and weaknesses. Navigating these options can overwhelm even experienced practitioners. CharmIQ addresses this challenge with a document-based approach rather than the chat interfaces that dominate the market. This represents more than a superficial interface change. It fundamentally changes how we interact with AI, allowing these capabilities to integrate more naturally into existing workflows.

CharmIQ’s power comes from its ability to create specialized AI agents called Charms. These agents function as virtual team members, each bringing unique perspectives to problem-solving. Every Charm possesses specialized knowledge and capabilities, enabling me to handle tasks that traditionally required teams of experts. These Charms act as cognitive partners, working alongside you whether generating strategic solutions, refining methodologies, or creating content.

Creating a Charm takes minimal effort. You simply describe the type of expert you need, and the system uses another Charm to write the definition for your new agent. The process typically takes 2-3 minutes to create an effective Charm. The output from different Charms varies significantly. Asking two different Charms the same question produces distinctly different answers.

CharmIQ has integrated with virtually every commercially available and open-source LLM. These integrations enable power users to leverage specific strengths of different models. Anthropic’s models excel at writing content. ChatGPT leads in creativity and reasoning. CharmIQ lets users switch between these models seamlessly, ensuring the right tool for each task.

This capability allows small teams to operate with the capacity and expertise of much larger organizations. By enabling AI-enhanced collaboration among human team members and AI-based agents, CharmIQ democratizes access to advanced capabilities. This matters significantly in today’s competitive landscape where agility drives success.

Consider a practical scenario: a product manager developing a go-to-market strategy. Traditionally, this involves coordinating across departments, synthesizing input from market research, sales, and engineering, and iterating through numerous drafts. With CharmIQ, the product manager can use specialized Charms for each step. One Charm analyzes market trends and customer insights, while another drafts a comprehensive go-to-market plan. This approach saves time and enhances quality by incorporating diverse perspectives.

The document-centric approach means all interactions, feedback, and iterations exist within a cohesive workspace. This eliminates friction associated with switching between tools and interfaces. The result? A streamlined workflow that lets teams focus on innovation and value delivery.

CharmIQ’s collaborative features extend beyond AI-human interaction to include real-time collaboration among team members. This proves invaluable in today’s remote and hybrid work environments. Team members work together within the dynamic workspace, sharing insights, providing feedback, and iterating on ideas without the constraints of traditional interfaces.

CharmIQ stands out not just for technical capabilities but for its strategic vision. By allowing users to customize and deploy multiple AI personas, it fosters creativity and experimentation. Users aren’t limited to predefined workflows. They can explore new approaches, test hypotheses, and iterate on solutions in real-time.

This flexibility benefits organizations trying to stay ahead in rapidly evolving markets. CharmIQ provides tools to adapt quickly to changing conditions, identify emerging opportunities, and respond with agility. For entrepreneurs and startups, this capability can determine success or failure.

As someone immersed in product management, I recognize the importance of aligning technology with business objectives. CharmIQ accomplishes this by providing a platform that enhances productivity, fosters collaboration, and drives innovation. It lets users focus on strategic thinking and high-quality work while reducing time spent on repetitive tasks.

By shifting from chat-based to document-centric interactions, CharmIQ redefines how we work with AI and each other. Its ability to integrate multiple AI models and create specialized agents offers remarkable flexibility and power. For teams of all sizes, CharmIQ enables AI-driven collaboration that unlocks new productivity and creativity levels.

I use this tool for hours daily. It saves me at least 1-2 days of work every week when generating work product. Incorporating CharmIQ into my workflow has boosted my productivity by 10-30x. If I primarily created documents or wrote code, I believe it would approach a 100x multiplier.

I’ve advised their team and CEO since they released their first internal beta. As they expanded access, they quickly discovered its broad appeal. Everyone who uses it becomes an avid power user within days.

If you want to try it, they offer a free trial. However, to fully understand its capabilities, I recommend signing up for a paid plan that unlocks all features. Use it for a few consecutive days, and you’ll likely find it transforming how you work.

Some personal and professional use cases:

I’ve introduced CharmIQ to all my teams and watched those who adopt it transform their workflow and approach. This transformation spans product managers, marketing teams, writers, and software developers. The software architect Charms I’ve created have educated teams on best practices and streamlined product launches.

Professional Example:

While leading Technology at Bark, we launched a new mobile app in just months using React Native. We accomplished this with two full-time developers (neither had used React Native before), plus a fractional team of one product manager and one QA specialist. From kick-off to launching both iOS and Android apps took three months. We used CharmIQ to create Charms that amplified each team member’s work: Market Research, Competitive Analysis, Product Strategy, Go-to-Market Plans, Architecture Design, Software Development Environment Configuration, and code writing and testing.

Personal Example – Writing:

As an author, I had spent three years writing a novel but got stuck with about 100 pages remaining. I couldn’t organize the final chapters and remained completely blocked for almost a year. I pasted my novel into CharmIQ and created Charms based on my favorite authors. I included my outline and what I had written so far, then asked these author-Charms (Neal Stephenson, Neil Gaiman, C.J. Cherryh, Frank Miller, Stephen King, Cormac McCarthy, Joan Didion, and Ian McEwan) for detailed feedback and help refining the remaining outline.

Their initial feedback proved harsh but broke my creative block. I finalized the outline in one night and wrote the remaining 100 pages as a first draft in about two weeks. After gathering human feedback, I made major revisions that continue today. My writing process now deeply incorporates CharmIQ.

Personal Example – Health:

I’m in my 50s with several long-term managed health issues, I’ve created a repository for all my medical data including test results, visit summaries, scans, and reports. I’ve created Charms representing my doctors: Primary Care Physician, Neurologist, Cardiologist, and Vascular Surgeon. The feedback from these Charms mirrors what my actual doctors tell me. I can also have them collaborate with each other, something nearly impossible in real medical practice.

I’ve created additional Charms for recipes, mixology, veterinary advice, restaurant recommendations, vacation planning, plumbing, HVAC, electrical work, home automation, and more.

Give Charm a try using this affiliate link and I’ll get a small financial bonus (anyone can sign up to be an affiliate – I’m just beta testing this for them.) They’re a great team of wonderful humans, and they’ve built a product that has changed everything for me. I’d cry if it went away.

https://www.charmiq.ai/landing?linkId=lp_024743&sourceId=eric-picard&tenantId=partner-charmiq-ai

Tagged AI, Artificial Intelligence, chatgpt, technology, Writing

Dec 17 2024

Leave a comment

Uncategorized

Programmatic Curation Surges Ahead: What is it, and why is it so hot?

by Eric Picard

As programmatic advertising continues to evolve, the concept of curation has become a critical focus. My article from a few years ago, The Fifth Wave of ad tech highlighted the rise of “privileged programmatic”, but today’s discussions center around the nuanced roles of curation. To truly understand its implications, we need to dig into the distinctions between manual and smart curation, and clarify how these approaches differ from traditional ad networks and the proto-curation methods established by programmatic buyers using PMPs.

Ad Networks: The Initial Curators

In the early days of programmatic advertising, ad networks acted as intermediaries, facilitating transactions between buyers and sellers. However, their model was fundamentally flawed. By exploiting market inefficiencies, ad networks engaged in arbitrage—buying low and selling high—without significantly enhancing transactional value. This approach, while initially convenient, soon revealed its limitations as advertisers grew wary of inflated costs and minimal value addition.The recognition of ad networks’ inefficiencies spurred a shift towards more transparent and efficient transaction models. Although some ad networks persist, their model is increasingly viewed as outdated and incompatible with the demands of a sophisticated programmatic ecosystem.

Proto-Curation: Traders using PMPs

For over a decade, advanced programmatic buyers have employed a strategy that could be termed proto-curation. This involves negotiating with publishers for privileged access to inventory, resulting in the manual creation and management of a vast array of Private Marketplace (PMP) deals within DSPs. These PMPs offer buyers curated inventory aligned with their specific needs, but managing thousands of such deals is labor-intensive and complex, and negotiating for privileged inventory access has varied results. This proto-curation is distinct from the curation models we see emerging today, and is the answer to the question of why curation is different from the approach trading desks have taken with PMPs for the last decade.

The Evolution of Curation: Standard and Smart Approaches

In recent years, curation has evolved into two distinct forms: Standard Curation and Smart Curation. Both approaches build upon the foundations laid by proto-curation, yet they offer unique methodologies and benefits.

Standard Curation involves human intervention to select inventory based on specific buyer criteria. This approach is akin to proto-curation but is more focused and refined, and often done on behalf of the publisher. Manual curators negotiate inventory access with publishers, ensuring that DSPs receive bid opportunities that meet predefined criteria. This method provides a critical layer of control and efficiency that open exchanges cannot offer, making it indispensable for buyers seeking to optimize their programmatic strategies. This curation is happening inside of platforms designed to improve and streamline the work buyers have been doing for the last decade by providing strong workflow and tools to streamline the process of curating inventory through PMPs.

Another piece of the puzzle is that curation is done typically on the sell-side of the ecosystem. It’s in the publisher’s best interest to curate inventory from their end and to ensure that any privileged access to inventory is coming through curation platforms, so they can preserve prices and margins. Sometimes the manual curation is done by the publisher’s sales team, sometimes it’s done by a third party on behalf of the publisher.

Frequently these platforms bring together audience data as a differentiator, sometimes they act as the means for an advertiser to bring their own first party data to the media environment. Publishers typically put PMPs from their curation partners into higher privileged positions in the ad server than those done for advertisers and agencies directly – because it’s in the publisher’s interest to increase curated inventory’s value. Examples of these curation platforms include Permutive and Audigent.

Smart Curation, on the other hand, leverages advanced technology to enhance the curation process. By utilizing proprietary algorithms, signals, and data, smart curation refines inventory selection, aligning buying decisions with advertisers’ strategic goals. Unlike manual curation, smart curation minimizes human intervention, relying on advanced technology to streamline processes and maximize efficiency. Examples of smart curation include Yieldmo and OneTag.

Note – for all forms of curation, every vendor in the ecosystem is developing some curation product that proposes to be the way that curation should be done. Nearly every SSP/Exchange has a curation tool or marketplace, lots of the older data companies are getting into the curation game, and there are several standalone curation platforms on the market now. The goal of this article is to get you up to speed on what everyone’s talking about, and go a bit deeper into why it matters.

Curation isn’t just about Curated Audiences

While this is a significant use-case, curated inventory against audiences defined in advertiser first-party data, potentially with lookalike audiences, it’s not the only use case. Many curation engines are not using user data or targeting audiences. Many are curating using contextual data, some with other performance signals. This is an important distinction because there have been several movements to rebrand curation against the concept of Curated Audiences, which in my mind are a subset of curation overall.

Dispelling Misconceptions: Curation vs. Ad Networks

Beware anyone telling you that curation is merely a rebranded version of ad networks. This simply isn’t true, and is often thrown out by very experienced people in the industry as a way to diminish the value of curation – but saying it sounds smart while truly missing the point of what curation is. While both models involve intermediaries, their methodologies and value propositions are fundamentally different.

Ad networks thrived on market inefficiencies, engaging in arbitrage without adding significant value beyond transactional convenience. Conversely, curation—whether manual or smart—focuses on optimizing inventory selection without engaging in arbitrage. Curation grants inventory access ahead of buyers coming in directly through their DSP through the open exchange. Curation provides DSPs with refined bid opportunities at higher levels of privilege in the auction to improve results. There are no hidden costs or markups; instead, curation aims to maximize advertisers’ investments by aligning inventory with campaign goals.

Navigating the Programmatic Ecosystem with Curation

To fully appreciate curation’s value, it’s important to understand the programmatic ecosystem’s complexity. From Supply-Side Platforms (SSPs) to Pre-Bid Frameworks and Ad Server prioritization rules, numerous factors influence buyer-seller relationships. Advertisers lacking privileged access risk losing valuable impressions, a challenge that curation effectively addresses by refining bid opportunities.

The impact of this kind of privileged inventory access:

Imagine two bids on the same root impression that is sent to the exchange – even from the same DSP. Bid 1 could be $5.00 against the open exchange. Bid 2 could be $5.00 against a curated PMP. Bid 2 will always win because the publisher is going to favor (give privilege to) the PMP that they are curating for that advertiser. To make it even more complicated, some publishers may give enough privilege to curated PMPs that cost doesn’t even matter. If Bid 1 through the open exchange was $50, and Bid 2 through the curated PMP was $5 – Bid 2 would always win.

DSPs evaluate each bid opportunity provided by exchanges and SSPs, valuing them based on campaign objectives. While comprehensive, this approach is inefficient, as most bid opportunities hold little value for a specific campaign. And DSP bidder algorithms are valuing every bid opportunity – when not every bid opportunity even warrants any scrutiny. Buying a bad piece of inventory just because the cost is low enough doesn’t really help lead to good outcomes. Consequently, the shift towards PMPs and curated inventory has become a strategic necessity to screen out inventory that shouldn’t be in consideration.

Standard curation continues to provide value, especially when curators negotiate priority access with publishers. Meanwhile, smart curation utilizes technology to either streamline the process, or to find powerful new ways to define and value inventory altogether. Smart curation is not the evolution of curation, it’s a subset of curation that defines and values inventory differently based on proprietary data and advanced algorithms and data to make informed decisions earlier in the bid stream than the DSP. Both approaches have value in enhanced access to inventory, increases in performance, increases in efficiency and offering significant value beyond what DSPs alone can achieve.

Strategic Implications and Future Directions

As programmatic advertising advances, the strategic implications of curation are profound. Advertisers must discern which platforms and technologies offer genuine value, distinguishing between superficial buzzwords and solutions delivering tangible results.Publishers, too, should embrace the transparency and efficiency that curation offers. By collaborating with advertisers and technology providers, they can increase yield, preserve pricing, and unlock new revenue streams that enhance their competitive edge in a rapidly evolving market.

The lessons from previous ad tech waves remain relevant. Balancing innovation with value creation is critical, and success hinges on our ability to adapt and evolve. Curation represents not only a technological advancement but a strategic shift poised to redefine programmatic advertising’s future. By navigating this new terrain thoughtfully, advertisers and publishers can unlock new opportunities and drive meaningful results in an increasingly competitive market.

Tagged advertising, Buy-Side, Curation, Data, digital-marketing, marketing, Programmatic, programmatic-advertising, Sell-Side, Targeting, technology

Oct 30 2024

Leave a comment

Uncategorized

From Small to Large: Scaling Product Management

By Eric Picard

In my career as a Chief Product Officer, I’ve had the opportunity to witness firsthand the evolution of product management roles in both small and large companies. This journey has given me a unique perspective on the challenges and opportunities that product managers face as they navigate these different environments. Today, I want to share some insights on how product management scales from small organizations to larger ones and a little story I like to call “The Parable of the Rocks.”

“HEY! Give me back my rocks!”

In smaller organizations, product managers are like Swiss Army knives. They juggle an array of roles, from product marketing to technical product management. This requires a versatile skill set and the ability to adapt quickly to shifting demands. In these environments, the scope of each role is broad, and resources are often limited. The challenge lies in effectively balancing these diverse responsibilities. The ability to switch contexts seamlessly and maintain organization is not just a helpful trait; it becomes a superpower.

For instance, a product manager in a small company might start their day aligning the product roadmap with the engineering team, spend the afternoon crafting go-to-market strategies, and end the day troubleshooting technical issues. They are the glue that holds disparate functions together, ensuring that the product not only meets market needs but also stays on track with the company’s overarching goals.This breadth of responsibility fosters a deep understanding of the product and its ecosystem. However, it also means that product managers in smaller companies often feel like they are carrying the weight of the world—or at least the product—on their shoulders. It’s a fast-paced and demanding role, but it also provides an unparalleled learning experience.

There comes a time in every company or team when the work becomes too much for one person, and that’s where things get very interesting. Eventually the work gets split across multiple product managers. Sometimes that individual contributor becomes a manager and has to divvy out their work. Sometimes a product leader is hired into the company as well. And as organizations grow, the product management role becomes more specialized, breaking into a variety of focused positions that allow for deeper expertise and efficiency.

Product Marketers focus on the Go-to-Market strategy, developing value propositions, creating sales materials, and assisting marketing and sales teams in targeting prospects. They decide whether to roll out by geography, market segment, or industry vertical, and prioritize efforts accordingly.
Product Strategists spend their time analyzing market opportunities, engaging with analysts and customers, crafting Market Requirements Documents, and conducting competitive analysis. Their role is to understand where the product fits in the market and how it can best meet customer needs.
Product Analysts or Product Operations Specialists ensure that products are properly instrumented for capturing user activity, enabling path analysis and financial performance evaluation. They provide invaluable insights into how the product is used and where improvements can be made.
Product Designers are responsible for the product’s look and feel, focusing on usability and user feedback. They conduct both qualitative and quantitative analyses, ensuring that the user interface is intuitive and effective.
Technical Project Managers coordinate the various teams and deliverables, ensuring that deadlines are met and resources are allocated efficiently. They play a critical role in keeping projects on track.

This specialization allows Technical Product Managers to concentrate on a more focused yet still pivotal role. They “own” the product, defining what will be built, prioritizing features on the roadmap, and writing the specifications that engineers use to develop the product. They still need to talk to customers, and they still need to stay on top of the market, but now they have help from partners. The Product Manager role now requires even more synthesizing input from various stakeholders, convincing the organization that their vision is the best way to tackle business challenges. They need a strong ego to hold firm opinions backed by data, yet remain open to ideas coming from anywhere. This all sounds wonderful, but the organizational transition and the personal transition that these previous superstar unicorns have to go through can be daunting.

This brings me to a story I often share when discussing this transition, which I call “The Parable of the Rocks.” Imagine being a product manager in a small team. Your day is spent picking up rocks—tasks, feature areas, responsibilities, and challenges—and putting them in your backpack. As the product develops and matures, you accumulate more and more rocks, and your backpack grows heavier. Eventually, it’s breaking your back. You’re walking hunched over, struggling to move forward, your chin is almost scraping the ground.

Finally, the company recognizes the need for help and hires a new product manager or even a leader for the PM organization, or splits the responsibilities out into some of these specializations mentioned above. This new person walks in, sees you bent double under the weight of all those rocks, and says, “Oh my god, let’s get some of that weight off.” They take some rocks out of your backpack, and either put them in their own backpack, or they hand them off to other PMs or teams. If that person is a new product leader, they might decide, we shouldn’t be doing some of these things, and they might throw those rocks back on the ground.

At first, the product manager feels relief. They stand up straight, stretch, crack their back, and take a few steps forward. But then they notice those rocks on the ground, or see others carrying them, and doing things with them differently than they’d have done, and they say, “Hey, those are my rocks! Give me back my rocks!” This parable illustrates a common pitfall in transitioning from small to large teams. It’s natural to feel a sense of ownership over tasks you’ve been managing, but it’s critical to embrace the shift.

Letting go of certain responsibilities allows you to focus on strategic priorities and leverage the strengths of a larger team. It can be very hard to let go, because the new person who owns that rock might see it very differently, might change the very nature of a feature and how it solves the customer problem, or might deprioritize that feature altogether. The transition from small to large companies can be a transformative experience for product managers. It requires a willingness to adapt and a readiness to embrace new challenges and opportunities. Here are some strategies to navigate this transition successfully:

Develop a Growth Mindset: Be open to learning and adapting to new ways of working. Embrace the opportunity to deepen your expertise in specific areas and collaborate with specialized teams.
Cultivate Strong Communication Skills: In larger organizations, the ability to effectively communicate your vision and align cross-functional teams is paramount. Become a great data-driven storyteller. Inspire your teammates, inspire your customers. Foster relationships with stakeholders and build a network of allies.
Focus on Strategic Impact: Learn to balance bot the day-to-day tasks with long-term strategic goals. Leverage the resources available to you in larger organizations to drive meaningful impact. Don’t feel like you need to own all the rocks.
Let Go of the Rocks: Recognize the value in delegating responsibilities and sharing the load with your team. Trust in the capabilities of others and focus on the bigger picture.
Embrace Change: Change is inevitable in the transition from small to large companies. Embrace it as an opportunity for growth and innovation.

Scaling product management from small to large organizations involves a shift in mindset and approach. It’s a journey that offers both challenges and rewards, and one that can ultimately lead to greater strategic impact and career fulfillment. Embrace the shift, learn to love to give your rocks away, but ensure the new people have all the context they need to value them appropriately. Learn to tell great, inspirational, fact-based and data-driven stories. It’s only by convincing others that what you believe should be done or built that you’ll win – both as a company and you personally as part of your career development.

Tagged business, entrepreneurship, leadership, product-management, technology

Sep 30 2024

Leave a comment

Ad Technology, Online Media, Software Development, Startups

What is Product Management, anyway?

Defining Product Management as a discipline can be hard, here’s everything you ever needed to know!

by Eric Picard

Product Management resides within the “builder” organization, the team responsible for creating products, services, or media offerings to be sold to customers. These offerings are typically developed by engineers, powered by scientists, and designed by product designers. The entire process is orchestrated by Product Managers, who are ultimately accountable for ensuring products are properly defined, developed efficiently, launched effectively, and operate profitably.

The Spectrum of Product Management Work

Product Management spans a wide range, covering everything from what a sales team sells to what an engineering team builds. The work in between encompasses various roles, and in different companies, these roles may be divided differently. Below is a breakdown of the interrelated disciplines, illustrating why Product Management is challenging to define and execute. Note that while each discipline is described separately, Product Managers in many companies may be responsible for multiple (or even all) of these areas in their day-to-day work.

If we move from sales towards engineering, the breakdown sort of looks like this (note that the path isn’t truly linear, so it’s just a loose framework.)

1. Product Marketing / Go-to-Market

Product Marketing involves preparing products for market consumption, ensuring they are presented in a consumable form that sales and marketing teams can utilize to reach customers. Product Marketing is a distinct discipline closely aligned with the builder organizations, although sometimes it reports into Marketing. Product Marketing is not Marketing Communications, and is much more closely aligned to Product Management. Responsibilities include: Developing compelling product positioning that clearly defines the product’s value proposition. Crafting messaging that highlights key differentiators and resonates with target customers. Managing the end-to-end launch process, including internal training, external communications, and post-launch analysis. Sales enablement functions such as creating sales materials, presentations, and demos. A critical deliverable is the Product Positioning document, which includes messages or claims about the product that can be used broadly across the organization and easily transformed into marketing or sales materials. For example, a product positioning statement might be, “Renovate your home with confidence using our easy-to-use device that scans walls for wiring before you begin a project.”

2. Product Strategy / Product Planning

This discipline focuses on determining what to build, requiring extensive research to understand market needs. It includes competitive analysis to understand the market landscape, customer research to gather insights from current, potential, and former customers, and market requirements analysis. The long-term vision, ideally spanning three years, often resides here. A key deliverable is the Market Requirements Document, distinct from Product Requirements Documents, which emphasizes customer needs without detailing implementation specifics. For instance, a market requirement could be, “Customers need a way to find wires inside walls to avoid accidentally cutting them during renovations.”

3. Product Operations

Product Operations might fall under the Product Manager’s purview or exist as a separate discipline, especially in larger companies. This role manages the business of existing products in the market, ensuring product health, profitability and alignment across teams. It involves defining and measuring metrics to ensure product success, optimizing to increase profitability, and collaborating across disciplines to maintain clear communication and alignment. Product Operations is often responsible for pricing, performance measurement, and ensuring products meet KPIs and business goals. Product Operations Specialists work closely with Product Management, Engineering, Marketing, Sales, and Customer Support to align all aspects of the product effectively. A primary task is defining and implementing product dashboards to measure performance against KPIs.

4. Technical Product Management

Often considered the “core” of Product Management, Technical Product Management is responsible for creating and managing a product roadmap. This includes developing a Product Requirements Document that spans multiple releases, breaking it into individual features, and writing specifications. It involves collaborating with Product Design for user-facing features to develop prototypes, design systems, and screen layouts. Technical Product Managers work closely with Engineering to ensure requirements are clear, fostering debate and improvement within a healthy organization. They also collaborate with Sales and other customer-facing teams to ensure feature prioritization aligns with business needs and work with Marketing to ensure demand creation. They often manage daily coordination efforts like standups or scrums, or other mechanisms for coordinating the day-to-day work related to building and releasing software (unless a Technical Project Manager is assigned to the project, in which case they do those things). A primary deliverable of a Technical Product Manager is the Product Requirements Document, which details necessary features to solve customer problems while avoiding implementation specifics, which are handled by Engineering. Another important deliverable is the Product Roadmap, which can take a variety of forms. Ultimately PMs are responsible for defining the product and getting the right product built, and writing requirements is a core function. A good high level requirement could be, “Create a small handheld device that clearly alerts a customer that active or inactive electrical wires are inside of a given wall.”

5. Technical Project Management

Technical Project Management focuses on coordinating engineering and other resources across a complex organization with varying timelines and levels of effort. The purpose is to keep projects on track and ensure complete transparency and coordination across teams. A key responsibility is maintaining a unified Gantt Chart, detailing team deliverables and timelines. Unlike other roles, Project Managers aren’t tasked with deciding what to build or communicating product details; they ensure that all teams understand their responsibilities, deadlines, and resource allocation, optimizing the efficiency of the overall process.

6. Product Design

Product Designers, sometimes called User Experience Designers, have a unique discipline within the world of design. Very often Product Designers are misunderstood by people outside of their day-to-day working chain as Graphic Designers. While Product Designers do sometimes create work that feels a bit like Graphic Design, their work is fundamentally different. The discipline ranges broadly from Research to Usability. User Experience Research includes deliverables like User Journeys, User Personas, Competitive Analysis Reports (from a design perspective), and Usability Research. They own the look and feel of the product by implementing design system that define how all aspects of the product look, and what the user interaction model of the product will be. This includes colors, fonts, and various elements of the styling of the product (buttons look like this, form fields look like that, there should be x many pixels buffer around images, etc…). Execution oriented deliverables include Information Architectures, proof of concept Wireframes, Pixel-Perfect Wireframes, and clickable prototypes. Product Designers, especially those building digital products or software, establish a design system for all aspects of the user experience. Product Designers are also responsible for usability testing and talking to customers about the products that they use to understand what works, and what doesn’t work. They sometimes do this with quantitative methods, and sometimes qualitative methods. Often they will also use automated tools that provide heatmaps and click-tracking analysis to show what users are looking at on the page, and what paths they are taking through the product or site.

A Complex Discipline

As you can see, the discipline of Product Management is complex and comprehensive. The work extends across a lot of discrete skills, and requires many different capabilities. Finding a product person who excels in all of these various disciplines is hard, we usually refer to those people as “Unicorns.” At early stage startups, the product managers are generally responsible for several, if not all, of these disciplines. As companies mature, the work breaks down into specific roles and responsibilities. When I was at Microsoft each of these were very clearly defined and separate roles, albeit sometimes with different names. Unicorns do exist, you can find people with the ability to effectively do all of this work, and to do it well. But they are extremely rare. My advice to most companies is to figure out what your teams do well, and what they need help with. Then find a logical breakpoint between the disciplines and establish the work in a way that everyone involved celebrates, rather than feeling like the work that they are good at and enjoy is being ‘taken away’ from them.

To show you how these teams interact, see the chart below. We’re looking at the product development life cycle from Left to Right, and the interaction between Product Marketing, Product Strategy and Product Management. You can see that Product Strategy starts the process off, and then hands off to Product Management. As the life cycle progresses, different roles are more or less engaged.

This is a product development life cycle map from a very mature company with a global footprint. Smaller companies will have less complexity in their map, and frankly, the roles will condense down to one or two rather than all three.

This chart doesn’t include software developers, scientists, researchers, technical operations, release management, product designers, or product operations. You can imagine how complex things might get in a fully resourced organization – and why a team might need a Technical Project Manager to coordinate it all.

I hope this gives you some sense of what Product Management as a discipline is all about. It takes a lot of work to build a product. Whether you’re a 3 person team or a 300,000 person organization, the work is the work, is the work. It all needs to be done in order to deliver a great product. It’s just a question of how large the offering is, and how streamlined of a process the team can use in order to deliver the product in a reasonable amount of time.

Tagged agile, business, commercialization, leadership, Product Lifecycle, product-design, product-management, product-marketing, product-strategy, program-management, project-management, Software Development, technology

Sep 17 2024

Leave a comment

Ad Technology, Artificial Intelligence, Marketers, Media Agencies, Online Media, Publishers

How AI is actively transforming advertising

by Eric Picard

Artificial Intelligence (AI) and its subset, Machine Learning (ML), have been integral to the advertising industry for decades. These technologies have transformed how businesses connect with consumers, optimizing many of the dollars spent and ensuring targeted engagement. But as AI evolves, particularly with the advent of Large Language Models (LLMs) and generative AI, we’re entering a new era where smart automation is a reality. Let’s explore how AI, ML, and LLMs are reshaping advertising, untangling the complexities of these technologies, and understanding their distinct roles and applications.

To begin with, it’s important to differentiate AI and ML. AI is the broader concept of machines performing tasks that typically require human intelligence, such as decision-making and pattern recognition.

Machine Learning (ML) is a subset of AI, and has been the engine that powers many of the smart decisions in today’s advertising landscape. At its core, ML is about teaching computers to learn from data, much like how we learn from experience. There are several approaches to ML, each with their own unique applications and strengths in advertising.

One of the most prevalent techniques is supervised learning. This approach involves training an ML model on a dataset that includes both inputs and the correct outputs—sort of like coaching a sports team with the playbook in hand. In advertising, supervised learning is often used for predictive targeting. By analyzing historical data, these models can forecast which segments of an audience are most likely to respond to a specific ad campaign. This allows advertisers to allocate resources more effectively and maximize return on investment.

Unsupervised learning, on the other hand, is a bit like sending a detective into a room full of clues without any instructions. The model explores the data, finding patterns and relationships on its own. This technique is ideal for audience segmentation, helping advertisers discover new and potentially valuable consumer groups based on shared behaviors or characteristics. It’s akin to discovering hidden subcultures within a larger community, providing insights that can drive more personalized marketing strategies.

Reinforcement learning is another fascinating ML approach, where models learn by trial and error—similar to training a pet with rewards and consequences. In the dynamic world of real-time bidding, reinforcement learning algorithms adjust bidding strategies on the fly, learning which actions yield the best results. This adaptability is crucial in environments where market conditions and consumer behavior can change rapidly.

It’s also worth mentioning neural networks, a type of ML model inspired by the human brain’s structure and function. These networks are particularly powerful in tasks involving complex pattern recognition, such as image and speech recognition. In advertising, neural networks can enhance programmatic buying by evaluating vast datasets to identify subtle patterns in consumer behavior, enabling more precise targeting and personalization.

While these examples illustrate some of the common ML techniques used in advertising, the field is vast and continually evolving. Each method brings its own set of tools to the table, contributing to a more nuanced and sophisticated advertising ecosystem. As ML technology advances, its role in crafting more targeted, efficient, and engaging advertising experiences will only grow, pushing the boundaries of what’s possible in the digital marketing space.

Now, let’s address the role of LLMs in advertising. LLMs, such as GPT-4o, are advanced AI models that excel in understanding and generating human-like text. They are not primarily designed for data analysis or real-time decision-making—which are traditional ML use cases—but rather excel in tasks that involve language processing and text-based interactions. LLMs are being leveraged to automate processes that require a nuanced understanding of language and context, such as drafting personalized ad copy, facilitating customer service interactions, and enhancing chatbots.

In media buying and selling, LLMs are being applied to automate complex processes that traditionally required human intervention. By programming LLMs to think with a specific viewpoint and set of instructions, they can streamline tasks like scheduling and orchestrating campaigns, crafting and refining ad messages, and even generating comprehensive reports. These models act more as strategic partners, assisting human teams in managing and executing processes efficiently.

The application of ML in advertising remains robust, focusing on data-driven decision-making. ML algorithms excel in predictive targeting, analyzing vast datasets to identify optimal audiences, and optimizing bids in real-time to maximize ad spend efficiency. These capabilities are essential for real-time bidding environments and dynamic pricing models, where decisions must be made swiftly and accurately based on ever-changing data inputs.

Generative AI, closely related to LLMs but with distinct applications, is making significant impacts in creative advertising. While LLMs are adept at processing language, generative AI models are designed to create new content—be it text, images, or even video. In advertising, generative AI can automate the creation of ad visuals or video content, generating variations tailored to different audience segments. This capability not only enhances creativity but also accelerates the production process, allowing for rapid experimentation and iteration.

The distinction between generative AI and LLMs is important. While both can be used in the creative process, LLMs focus on language and dialogue, whereas generative AI extends to producing varied forms of media content. Together, they offer a comprehensive toolkit for advertisers looking to innovate and engage audiences more effectively.

As AI continues to advance, ethical considerations around data privacy and algorithmic bias abound. Transparency in how AI systems operate and make decisions is essential to maintain consumer trust. Balancing automation with human creativity and insight is also crucial. While AI can handle data-driven processes, human expertise is irreplaceable in crafting compelling narratives and understanding the subtleties of consumer emotions.

Looking ahead, the fusion of AI technologies promises even more sophisticated advertising solutions. We can anticipate AI models that predict consumer needs with remarkable precision, integrating seamlessly into the consumer journey. This future of advertising is not just about efficiency—it’s about creating meaningful, anticipatory experiences that resonate with consumers on a personal level.

By automating routine tasks and augmenting human capabilities, AI is enabling advertisers to deliver more personalized, effective, and efficient campaigns. The key to success will be embracing AI as a partner, leveraging its strengths while preserving the creativity and empathy that only humans can provide.

Tagged AI, Artificial Intelligence, chatgpt, llm, technology

Sep 16 2024

Leave a comment

Ad Ecosystem, Ad Serving, Ad Technology, Marketers, Media Agencies, Online Media, Online Privacy, Programmatic, Publishers

Programmatic Ads in 2024

A privacy-centric new world

By Eric Picard

I’ve been writing about advertising technology and digital advertising since 1999. Every five to seven years we go through a new wave of transformation. The current wave is the Privacy wave, and it’s transforming the way that the infrastructure of the ad technology space functions. Previously we were reliant on digital identifiers, generally enabled by cookies on the web, and AdIDs in the mobile app space, but today these mechanisms have been deprecated or are in the process of being deprecated. This has decimated a huge swath of the industry that relied on these IDs to make all sorts of decisions, including linking together multiple tracking methodologies – to power 3rd party data. That industry has shrunk significantly, since users are becoming impossible to track that way.

This transformation is largely driven by increasing consumer demands for privacy and stringent regulatory requirements. Today, the focus has shifted towards privacy-centric methodologies, with first-party data taking center stage. In this comprehensive overview, we’ll explore the current state of programmatic advertising, delve into the innovative strategies employed to maintain effectiveness while upholding privacy, and highlight the strategic implications for advertisers.

Understanding Programmatic Advertising Today

At its core, programmatic advertising is the automated buying and selling of online ad space, allowing brands to target specific audiences at scale. This process involves real-time bidding (RTB), where ad inventory is bought and sold on a per-impression basis in a few hundred milliseconds. Programmatic platforms use sophisticated algorithms to analyze vast amounts of data, enabling advertisers to reach their desired audiences with precision.

Without cookies and AdIDs, the whole industry is in the process of retooling.

The Role of First-Party Data

First-party data refers to information collected directly from consumers either through brand-owned channels such as websites, apps, and loyalty programs on the advertiser side, or through the publisher’s relationship with the consumer. This data is gathered with explicit user consent, making it both reliable and compliant with privacy regulations like the General Data Protection Regulation (GDPR) and the California Consumer Privacy Act (CCPA).

The importance of first-party data cannot be overstated. It provides a comprehensive view of consumer behavior, preferences, and demographics, enabling advertisers to create personalized and relevant ad experiences. By leveraging first-party data, brands and publishers can build direct relationships with their customers. This data can be used in many valuable ways, but importantly can be used to deliver the correct advertising to consumers, and to measure effectiveness of advertising.

Privacy-Centric Targeting Methodologies

In response to the growing emphasis on privacy, advertisers have adopted several innovative targeting methodologies that prioritize user rights while maintaining effectiveness:

Contextual Targeting: In a world where user data is increasingly protected, contextual targeting offers a privacy-friendly alternative. This method involves placing ads based on the content of the web page rather than individual user data. For example, an ad for hiking gear might appear on an article about nature trails. By focusing on the context rather than the person, advertisers can maintain relevance without compromising privacy. Contextual has been around forever, but new approaches to contextual are being used in much more sophisticated ways. What’s old is new again.

Cohort-Based Targeting: Google’s Privacy Sandbox initiative popularized the concept of privacy-first approaches to targeting users, their first broadly discussed version was called Federated Learning of Cohorts (FLoC), which groups users into cohorts based on similar browsing patterns. Google has reiterated and revised their approach several times, and hasn’t really gotten traction. But that’s mostly because they’re Google, and the industry isn’t sure they trust the approach they have come up with. While Google’s approach isn’t gaining traction, the overall approach of building privacy-safe cohorts of users for anonymous targeting is sound, and may well get cracked at the industry level sooner or later. This approach allows advertisers to target clusters of users with shared interests, removing the need for individual tracking.

Identity Solutions: With the decline of third-party cookies, identity solutions have emerged as a viable alternative. These solutions use hashed identifiers like email addresses or phone numbers to create a persistent identity across different platforms. The success of these solutions depends on user consent and robust data protection measures, offering a way to recognize users while respecting their privacy. While these approaches work technically, getting to scaled anonymous identity is a real challenge that still is being overcome.

Data Clean Rooms: These secure environments allow brands and publishers to collaborate and combine their first-party data without exposing user identities. Data clean rooms enable advertisers to gain insights and enhance targeting precision while adhering to privacy regulations. By facilitating data collaboration in a controlled setting, clean rooms offer a solution to the challenges posed by privacy laws. Note that while this approach overcomes the legal issues, it’s still not quite clear that consumers will be accepting of the approach as it becomes normalized and written about in the press.

Beyond Basic Targeting: Curated Audiences and Inventory

Publishers lost a lot of ground in the first wave of programmatic advertising, which pushed all the power to the media buyer. This created a long period of data asymmetry where publishers didn’t know why an advertiser was buying any given impression – and the knowledge of their own audience was ignored by the buyer. Things have changed with the loss of 3rd party data, cookies and IDs.

The industry has responded by rallying around the sell-side of the market for the first time in many years. The outcome is what is being called Curated Audiences and Curated Inventory. Effectively the sell-side of the market has access to an immense amount of information on their side of the fence, about the consumers visiting their sites, and about the behavior of those consumers. This all is Publisher First Party Data, and is able to be blended with other data sources by the Publisher to create large pools of inventory that are curated to the needs of the buyer. Vendors and the publishers themselves have found ways to build high scale and highly effective targeting and optimization technologies based on these approaches, and package the inventory into programmatic deals (PMPs):

Curated Audiences: Publishers can create pre-defined audience segments based on aggregated data insights and consumer behaviors. These segments are crafted using a combination of consented user data (the consumer said the publisher can use it) and non-personal data, ensuring compliance with privacy regulations. By curating audiences, advertisers can buy advertising at scale that allows them to reach relevant consumers without relying on intrusive data collection methods.

Curated Inventory: Publishers can package their ad inventory using their own data and intelligence, often with the support of sophisticated external data providers. This approach allows advertisers to target ads to appropriate audiences without using personal data. For instance, sophisticated geo-targeting can be employed to deliver regionally relevant ads, enhancing user engagement without compromising privacy. Some vendors use decades old approaches in new ways, using census and other forms of data that allow demographic, psychographic, and other types of targeting by overlaying the location of the user when they receive the ad against known information about that location. For instance, assuming someone on a golf course at 3PM is likely a golfer.

These curated solutions provide advertisers with powerful tools to reach their desired audiences while navigating the complexities of privacy regulations.

Strategic Implications for Advertisers

The shift towards privacy-centric methodologies is not just a technical adjustment; it’s a strategic imperative. Advertisers must align their programmatic strategies with these approaches to maintain consumer trust and comply with legal requirements. This involves investing in data infrastructure, nurturing direct customer relationships, and staying informed about regulatory changes.

Moreover, as privacy becomes a selling point, brands that demonstrate a commitment to safeguarding user data can differentiate themselves in a crowded marketplace. Transparent data practices and clear communication about how consumer data is used can build loyalty and encourage engagement.

Overcoming Challenges in the Programmatic Space

While the transition to privacy-centric advertising offers numerous benefits, it also presents challenges. Advertisers must adapt to new technologies and methodologies, requiring investment in training and infrastructure. Additionally, the loss of third-party cookies necessitates a reevaluation of measurement and attribution models, as traditional methods may no longer apply.

To overcome these challenges, collaboration within the industry is crucial. Brands, agencies, publishers and technology providers must work together to develop standards and best practices that prioritize privacy while delivering effective results. By fostering an ecosystem of transparency and cooperation, the industry can navigate the complexities of programmatic advertising in 2024 and beyond. The IAB Tech Lab is a great example of an industry organization working to drive this kind of collaborative adoption and roll out.

The Future of Programmatic Advertising

As we look to the future, programmatic advertising will continue to evolve in response to technological advancements and changing consumer expectations. The integration of artificial intelligence (AI) and machine learning will enhance targeting capabilities, allowing for even greater personalization while respecting privacy. Additionally, the rise of connected devices and the Internet of Things (IoT) will present new opportunities for advertisers to engage with consumers in innovative ways.

Ultimately, the key to success in programmatic advertising lies in embracing privacy as a core value. By doing so, advertisers not only comply with regulations but also foster trust and loyalty among their audiences. In 2024 and beyond, the challenge will be to innovate continuously while keeping consumer privacy at the forefront of every decision. As the landscape continues to evolve, staying informed and adaptable will be essential for advertisers seeking to thrive in this dynamic environment.

The current state of programmatic advertising is characterized by a delicate balance between effective targeting and stringent privacy requirements. First-party data has emerged as the linchpin of modern advertising strategies, offering a path to personalization that respects user privacy. By adopting innovative targeting methodologies and maintaining a strong commitment to privacy, advertisers can successfully navigate the programmatic landscape, building lasting relationships with their audiences in the process.

Tagged advertising, business, digital-marketing, marketing, programmatic-advertising

A Product Strategy Consultancy focusing on Ad Technology and Online Media consultancy

Raising money is selling a piece of your company

The numbers, and why they should reset your expectations

Meetings matter more than outreach volume

Know who you’re pitching

The problem of overfitting on success stories

Product-market fit is not a buzzword

It takes a team with all the right skills

The real job descriptions

Ad Networks: The Initial Curators

Proto-Curation: Traders using PMPs

The Evolution of Curation: Standard and Smart Approaches

Curation isn’t just about Curated Audiences

Dispelling Misconceptions: Curation vs. Ad Networks

Navigating the Programmatic Ecosystem with Curation

Strategic Implications and Future Directions

The Spectrum of Product Management Work

1. Product Marketing / Go-to-Market

2. Product Strategy / Product Planning

3. Product Operations

4. Technical Product Management

5. Technical Project Management

6. Product Design

A Complex Discipline

Understanding Programmatic Advertising Today

The Role of First-Party Data

Privacy-Centric Targeting Methodologies

Beyond Basic Targeting: Curated Audiences and Inventory

Strategic Implications for Advertisers

Overcoming Challenges in the Programmatic Space

The Future of Programmatic Advertising