Your Roadmap is Bullshit: Build for the Wedge

Look at your product roadmap.

Go ahead. Open the slide.

Look at those beautiful blocks of color.

Quarter one: Feature A. Quarter two: Integration B. Year three: “The Global AI Operating System.”

It’s a fairy tale.

You’re sitting in your office, and you’re dreaming.

You think because you have a model and some compute that the world is your oyster.

The world is not your oyster. The world is a crowded, noisy, violent marketplace where forty other guys just like you are selling the exact same “Global AI Operating System.”

You are an early-stage B2B AI firm. You are pre-Series C. You are under pressure.

The VCs want growth. The market wants utility. And the big incumbents—the Microsofts, the Salesforces, the incumbents with the data and the distribution—they are coming for you. They don’t need to out-innovate you. They just need to out-wait you.

You want to tell me about your “Platform.”

“We’re a horizontal layer for enterprise intelligence.”
“We’re the connective tissue for the modern data stack.”

Listen to me.

A “platform” is something you earn. It is not something you start with.

Salesforce didn’t start as a platform. It started as a way to keep track of names and numbers without a server in the basement. It was a CRM. It was a wedge.

Once they owned the desk of every sales rep in the Valley, then they earned the right to be a platform.

In the AI world, the “Platform Trap” is even more dangerous.

The technology is moving too fast. If you build a broad, horizontal platform today, three months from now OpenAI or Anthropic will release an API update that turns your entire product into a free feature. They will expand the context window or add a native tool-calling function, and your “intelligence layer” vanishes.

Your imagination tells you to be everything to everyone. Your imagination wants the big TAM. Your imagination wants the headline in TechCrunch.

If you build for your imagination, you die. If you build for your vision, you go broke.

You must first build for the wedge.

What is the Wedge?

The wedge is the one specific, painful, high-frequency problem that you solve better than anyone else.

Not “five times better.” Ten times better. Twenty times better.

It is the problem that keeps your Panic Persona awake at 2:00 AM. It isn’t “efficiency.” It isn’t “productivity.” Those are words for people who have nothing to sell. Those are words that get deleted from an inbox in four seconds.

The wedge is narrow. It is boring. It is often ugly.

The wedge is: “Our legal team spends sixty hours a week manually redlining master service agreements.”

The wedge is: “Our supply chain managers can’t see the inventory levels in the Singapore warehouse in real-time because the legacy ERP doesn’t talk to the sensors.”

The wedge is: “Our SDRs are sending emails to dead leads because our database is a dumpster fire.”

You find the pain. You find the one enemy. If you’ve read my work at Win the Brand War, you know about the 7 Enemies.

You pick one. You attack it. You ignore everything else.

The Anatomy of a Winning B2B AI Wedge

Most founders fail here because they think the “AI” is the wedge. It isn’t. The AI is the engine. The wedge is the point of the spear.

To build a wedge that actually pierces the enterprise, it must meet three criteria:

1. High Gravity, High Frequency: Don’t solve a problem that happens once a year. Solve a problem that happens ten times a day. You want to be the tool they open before they even have their coffee. If you solve a high-gravity problem (it costs them a lot of money) with high frequency (they do it all the time), you become indispensable. You aren’t a “nice to have” line item. You are the infrastructure.

2. The “Only” Factor: If three other startups can claim to solve the same problem using the same LLM, you don’t have a wedge. You have a commodity. You need to find the specific “Only.” “We are the only firm that integrates with this specific, archaic document format used in maritime insurance.”
“We are the only agents that can operate inside a local environment without a cloud handshake.”

3. Immediate Time-to-Value (TTV): In the old days of SaaS, you could have a three-month implementation period. Those days are gone. In the AI war, if you can’t show value in fifteen minutes, you’ve lost the room. Your wedge should be so sharp that it cuts through the procurement red tape. It should be a no-brainer for a department head to put it on a corporate credit card.

The Imagination Roadmap vs. The Wedge Roadmap

The Imagination Roadmap is built on “Then.”

Then we will add the HR module.

Then we will expand into EMEA.

Then we will integrate with every ERP on the planet.

The Wedge Roadmap is built on “Now.”

Here is how you structure a roadmap that serves the wedge first and your imagination second:

Phase 1: Total Domination of the Specific (Month 0-6)

Your roadmap for the next six months should contain exactly zero “visionary” features. It should be a list of brutal, tactical improvements to your core wedge.

If your wedge is AI-driven contract analysis for mid-market logistics firms, then every single update should make that analysis faster, more accurate, or easier to export.

That’s it.

I’ve seen founders get distracted by a “big fish” customer who says, “We’ll sign a $500k contract if you just add this one payroll feature.”

Do not take that money.

That money is a trap. It turns you into a dev shop for one customer. It kills your ability to scale the wedge. You tell them no. You tell them to go buy Gusto. You are not Gusto. You are the anti-Gusto.

You are building a brand. And a brand is a promise. If you promise everything, you are promising nothing. When you own the wedge, you own the category.

Phase 2: The Adjacent Expansion (Month 6-18)

Once you have 80% market share in your tiny, narrow, boring wedge, you look left, and you look right.

What is the very next problem the same user has?

If you fixed their contracts, maybe now you help them manage the vendors inside those contracts. You don’t jump to a new department. You don’t jump to a new ICP.

You stay in the foxhole with the person who already trusts you.

This is where your roadmap starts to look like a product, not just a tool. But every feature must be a logical extension of the trust you’ve already built.

Your imagination wants you to jump to the next shiny thing. The wedge roadmap demands that you stay focused until the job is done.

Phase 3: The Platform Emergence (Year 2+)

Only now do you start talking about being a platform.

By this point, you have the data. You have the trust. You have the integration points.

Your “imagination” is no longer a guess; it is a response to the reality of your market position.

Most AI firms try to do Phase 3 in Month 1. That is why they fail. They try to be the “Operating System” before they’ve even learned how to be a single useful app.

Why the “Big Vision” is a Sales Liability

You think your big vision is helping you sell. You think the VP of Operations cares about your “2030 Roadmap.”

They don’t.

They are under-resourced. They are being told to “implement AI” by a board that doesn’t understand what AI is. They are scared. They don’t want a “visionary partner.” They want someone who can make a specific problem go away so they can go home at 5:00 PM.

When you lead with your imagination, you create friction. You make the buyer think about the “future state,” which requires change management, budget reallocations, and meetings with IT.

When you lead with the wedge, you solve a problem. Solving a problem is easy to buy. “Vision” is hard to buy.

The Threat of Commoditization

The reason you need a wedge is that AI is becoming a commodity.

The models are getting cheaper. The capabilities are getting more uniform. If your value proposition is “We use GPT-4 to do X,” you are already dead. You just haven’t fallen over yet.

The only way to beat commoditization is through brand and positioning. You have to be the “only.”

A wedge gives you a moat. A “vision” gives you a target on your back.

If you are building a horizontal intelligence platform, Microsoft will eventually build a better one and

give it away for free with Office 365.

But Microsoft will never build a hyper-specific tool for managing the liability of chemical spills in the Ohio River Valley. That market is too small for them.

And that is why that market is perfect for you.

How to Pivot Your Roadmap Today

If you are a founder and you are reading this, I want you to do something.

Take your current roadmap. Print it out. (Or put it on a big screen).

Take a red pen. Cross out every feature that doesn’t directly serve your entry-point wedge.

Does this feature help the user solve the primary pain in under five minutes? No? Cross it out.

Does this feature require us to enter a category where we don’t have a unique USP? Yes? Cross it out.

Is this feature here because an investor said “It would be cool if you did X”? Cross it out.

What you are left with is your real roadmap. It will look smaller. It will look less impressive. It might even look scary.

But it’s real.

And in a world of AI hallucinations, “real” is the only thing that sells.

The War for the Category

You are in a war.

You’re fighting for the mindshare of a buyer who is being bombarded by a thousand different messages every single day.

If you show up and say, “I have a platform that will revolutionize your entire business,” they will ignore you. Their brain is trained to filter out that noise. It’s too big. It’s too vague. It’s too much work.

If you show up and say, “I know you hate this one specific thing. I have the one specific tool that kills it. It takes ten minutes to set up. Do you want to see it?”

This is how you win.

Your imagination is a beautiful thing. It’s why you started the company. But your imagination is for late-night drinks with your co-founders.

When it comes to your roadmap, serve the wedge.

Build the entry point. Dominate the niche. Create the category.

Then, and only then, can you go back to dreaming.

Are you struggling to find your wedge?

I help early-stage B2B AI firms find their “only,” kill their generic positioning, and win the brand war.

If you’re under pressure to scale but your GTM feels like throwing spaghetti at a wall, we should talk.

I’m currently taking on a limited number of consulting projects for Pre-Series C firms. I also help VCs fix the positioning of their struggling portfolio companies.

Let’s build a roadmap that actually works.

Work with me at Win the Brand War.

Connect with me on LinkedIn.