You Know Which Segment to Be In. You Just Won't Bet on It.
At Altium, I had a campaign idea. One line.
Why would you buy PCB tools from an IC company?
Management killed it. The logic: if Cadence decided to care about PCB, they could build the best PCB tool in the world. Deep pockets, deep engineering bench, deep relationships with the same customers we were chasing. Don't poke the bear.
So we didn't.
Years later I went to Cadence. Spent enough time there to know: they don't give a damn about PCB. They serve PCB designers because they can, not because they chose to. PCB is a line item in a portfolio organized around IC design. The tools exist. The attention doesn't.
The incumbent doesn't own your segment because they chose it. They own it because nobody else committed to it completely. Their indifference isn't an oversight they'll correct when they notice you. It's structural — it's how a $3B company allocates attention across seventeen product lines. It won't change because you got funded. It won't change because you're growing. It's a permanent feature of how large companies work.
That gap is your opening.
The startup that blows its own opening by running the same multi-segment play as the incumbent — just with fewer resources, less runway, and no existing customer relationships — closes the one door the incumbent left open.
Before the Diagnosis, the Tool
Not every segment is worth committing to first. The right first segment passes three tests. All three. Not two.
The buyer has already tried to solve this problem and failed. Not mild dissatisfaction. Not a vague preference for something better. Active failure — they ran a project, bought a tool, assigned headcount, and it still isn't working. Present tense. The cost of that failure is sitting on someone's desk right now. A buyer with an active failed solution is worth ten buyers with a vague sense that things could be improved.
Your technology is genuinely capable of serving it. Not "could be adapted with six months of additional engineering." Capable now, or close enough that the gap is execution, not invention.
Real TAM with prestige. Winning here means something. The segment has prestige when its buyers talk to each other — at conferences, in forums, in the peer networks that exist inside every technical community. The test: does a win at Company A generate an unprompted call from Company B? Not because you marketed to Company B. Because the VP at Company A mentioned it to a peer. If that mechanism exists in your segment, your wins travel and you compound. If buyers don't compare notes, don't attend the same events, don't have the same watering holes — your wins stay local. You're starting from zero with every deal.
Find the segment that passes all three. Then stop looking at the other columns.
Most founders know which segment it is. They've known for months. The problem isn't the framework. The problem is trust — specifically, the willingness to bet on one answer when staying in five segments feels like prudence.
Here's what commitment looks like on Monday morning. You rewrite the homepage for one segment. You decline the next meeting request from outside that segment. When someone asks what market you're in, you give one answer — not "primarily X but also seeing interest in Y and Z."
Compare these two:
"We serve all engineering teams across industries."
"We serve RF IC design teams working on advanced wireless systems."
If you can't make that choice in one sentence without hedging, you haven't committed yet.
What You Think You're Doing
You didn't build the shotgun because you were undisciplined.
You built it because you were being responsible. The automotive company emailed. Real company, real budget, real timeline. Saying no felt irresponsible. Founders are supposed to follow the market. Founders are supposed to stay open to signals.
So you said yes. You had six conversations. Nothing moved.
Or: you've been in semiconductor for a year. A customer — your best customer, the one you've invested the most in, the reference you use in every late-stage deal — asks you to support an adjacent use case. They're not a prospect from the wrong segment. They're the relationship you've built most carefully. Saying not yet to them feels genuinely insane. These aren't automotive strangers. This is the account your company's early story is built around. If they're also 40% of your ARR, not yet feels like a business decision, not a focus decision.
It's still a focus decision.
Both of those moments feel like the right call. Responsive to market signals. Loyal to customers. Good founder behavior.
Say not yet anyway.
Every segment you stay present in is a bet you're hedging. Hedged bets don't compound. They dilute.
What It's Costing Your Message
Target enterprise engineering teams across multiple verticals.
That's a job description I've seen. Probably written by a founder who knew exactly what they were doing and had no idea what they were doing.
Show me the homepage. It reads like a paragraph written by committee — hedged, abstract, designed to offend nobody, which means it connects with nobody. You can't write a great positioning sentence for a company that serves everyone. The discipline of deciding who you're writing for is the same discipline as deciding which segment you're in. One follows the other.
In 1997, Apple had 40 products. Jobs cut it to four. People called it destruction. It was the condition under which any of the four could become excellent. Forty products means forty stories, forty customer types, forty reasons to buy — none of them said with enough conviction to land anywhere.
The segment decision isn't just a messaging decision. It's every decision.
You cannot brief a sales rep without a segment. You cannot write a job description for a BDR that means anything. You cannot onboard a head of marketing into a campaign that doesn't have a reader.
The segment decision is the load-bearing wall. Pull it out and the pitch falls down, the homepage falls down, the onboarding falls down, the hiring falls down. You cannot fix any of those things without fixing the decision underneath them.
And the buyer who can't find their industry in your homepage doesn't call. They move on.
What's Actually Happening to Your Buyers
Your buyer is not evaluating your product. They are managing their career risk.
The question they're asking isn't "is this the best tool for our use case?" It's "is this a safe choice?" Safe means someone like them already bought it. Safe means if this goes wrong, they can explain why they chose it. Safe means a name that arrives before the first sales call, before the first demo, before the first proposal.
You are not that name yet. In five segments simultaneously, you are not that name anywhere.
Here's a scene you've been in. You're twenty minutes into a sales call. The buyer asks, almost casually, "have you worked with anyone else in our space?" You have. The reference you're about to give is semiconductor. The buyer is aerospace. You answer confidently. Technically accurate. The buyer nods. The conversation continues.
But something shifted. You answered the wrong question with the right information. The buyer's internal question — who is everyone like me buying? — landed on silence. The reference didn't connect. It said: this vendor knows people in engineering. It didn't say: this vendor is the obvious choice for people exactly like you.
Now the opposite scene.
You walk into a sales call. The buyer says, almost as an aside: "yeah, we heard about you from Sarah at Northrop." You didn't set that up. You didn't ask Sarah to make that introduction. It happened because people in the segment are talking about you in rooms you're not in. The buyer's internal question already has an answer before you open your mouth.
That's what owning a segment feels like from the inside. Before that scene happens, you're still earning.
The Board Made It Worse
You were already running the shotgun when you walked into the Series A meeting.
But then someone asked about defense.
You said yes, we're also looking at defense — because the question was asked, because the TAM needed to be bigger, because saying "we're only in semiconductor" felt like leaving the table small. One sentence. The board wrote it down. Now there's a slide.
The board's incentive and your company's incentive are not the same thing at this stage. The board needs a market size story for the deck. You need a win. "We serve the entire engineering software market" closes no sales calls and drives no strategy. "We've won 40% of RF engineering in semiconductor and here's our next bowling pin" closes board meetings and enterprise deals both.
The founder who walks out of that board meeting and removes the defense column from the deck is the one who closes deals.
And the board will ask again next quarter. And the quarter after. The pressure is not a one-time event. It's the quarterly gravity of people who don't have to execute what they're asking you to consider.
The Referrals Are the Tell
Three columns. Semiconductor, defense electronics, industrial IoT. Opportunities in each. Conversations in each.
Not wins. Not trust. Conversations.
The distinction matters. You can have a full pipeline across five segments and zero of the specific thing that actually moves enterprise decisions: the feeling — on the buyer's side — that everyone like them is already evaluating you. That the conversation they're in is one of many identical conversations you've had with people at companies exactly like theirs.
I was a startup that some people in three industries had heard of. And none of them trusted me.
Year two. Deleted two columns. Stopped answering the automotive emails. Stayed in semiconductor.
Three of the next five deals came from referrals inside that one segment. People I hadn't called. Companies that reached out because a peer had mentioned it. The community had started working for me instead of against me.
The referrals are the tell. They don't happen because you have the best product. They happen when a community decides you're theirs — when being associated with you says something good about the buyer, when recommending you is a favor and not a risk.
Communities don't form around companies that also serve other communities. They form around companies that chose them. Specifically. Visibly. At the cost of saying no to other things.
Remember the scene where the reference didn't land — the buyer who nodded and moved on? Here's what the opposite looks like.
Earning the Right to Expand
This is not an argument for staying in one segment forever.
It's an argument for earning the right to leave it.
The test isn't a revenue milestone. It isn't a number of deals closed. It's behavioral. Deals get shorter without you changing the pitch. Objections that took an hour in your first ten enterprise conversations take five minutes now — because someone in the room already heard the answer from a peer before you arrived. The procurement team tells you they've already reviewed your security questionnaire. You walk into a call and the buyer says they've spoken to two of your customers already. You didn't set that up.
None of that happens because you got better at selling. It happens because the segment started working for you instead of against you.
If you're asking yourself whether you've earned the right to expand, you haven't. The signal is unmistakable when it's there. The absence of certainty is the answer.
When you have it, the first segment becomes the story you tell in the second. The wins travel. The references travel. The shortcut you built in segment one opens the door into segment two.
Before that: more commitment, not more segments.
The founders who build durable companies in this space have one thing in common that isn't in any playbook: the capacity to say "not yet."
The automotive email arrives. Real company. Real budget. Real timeline. You say: not right now. We're focused elsewhere.
That feels insane. Like leaving money on the table. Like arrogance dressed up as discipline.
It's the only move that makes your primary segment feel what they need to feel: that you chose them. Not because they were the only option. Because they were the right one.
The shotgun doesn't diversify risk.
It is the risk.