How to Tell If You Have a Marketing Problem or an ICP Problem

"We need more leads."

It's the most common request we hear from software companies — and after working with 80+ of them, we can tell you it's almost never the actual problem. It's the symptom.

Here's the uncomfortable pattern: a company invests in content, runs ads, maybe hires an agency. The metrics move a little, then flatten. So the conclusion becomes "our marketing isn't working," and the response is to do more marketing. More posts, more budget, another agency.

But if the real problem sits one level deeper — in who all of that marketing is for — then more of it doesn't fix anything. It just makes the wrong message louder.

This article gives you a practical way to tell the difference between a genuine marketing problem and an ICP problem, and a five-step process to fix the second one. No theory you can't use by Friday.

What an ICP problem looks like (and why it hides so well)

Your ICP — ideal customer profile — is the specific type of company your product or service creates the most value for. Not who could buy from you. Who you're built for.

An ICP problem hides well because its symptoms look exactly like marketing failures:

Your content gets no traction — because it's written for "tech decision makers" in general, which means it resonates with no one in particular. Your ads don't convert — because a message built for everyone gives no single reader a reason to click. Your pipeline runs almost entirely on referrals — because referrals carry context and trust with them, doing the job your positioning should be doing on its own.

Notice what these have in common: each one looks like an execution problem. Better copy, better targeting, better creative. And sometimes it is. But when you fix the execution and the needle still doesn't move, you're looking at something the execution can't reach.

A dev shop that says "we build software for everyone" competes with ten thousand identical companies — on price, because price is the only visible difference. No budget fixes that. It's not a marketing problem. It's a decision that hasn't been made.

Four signals it's your ICP, not your marketing

Run these four checks. You don't need analytics access for any of them, just honesty.

1. Ask three people on your team who your ideal client is. If you get three different answers, there's your diagnosis. When the ICP lives in the founder's head instead of on paper, every person in the company markets and sells to a slightly different customer.

2. Look at your last ten proposals. Did they go to ten wildly different companies, different sizes, industries, budgets, problems? A pipeline with no pattern means your targeting doesn't have one either.

3. Count the price objections. If most sales conversations turn into negotiations, your prospects can't see specific value — so they default to comparing numbers. People rarely negotiate hard on something they clearly need.

4. Read your case studies as if a competitor wrote them. If any competitor could plausibly claim the same story, your case studies aren't proof of anything specific. They're filler.

Two or more of these signals? Keep reading. Zero? You may genuinely have an execution problem — and that's a different article.

How to fix it: five steps

This is the process we run with clients before touching any campaign. It requires no tools you don't already have.

Step 1: Start with revenue, not personas

Don't start by imagining your ideal customer. Start with the ones you already have. Pull your last fifteen clients and score each from 1 to 5 on four things: margin (not revenue; margin), how easy they were to work with, results you can actually prove, and the clone test: would you take ten more exactly like them tomorrow?

Personas built in a workshop describe who you wish bought from you. Your client list describes who actually does, and at what cost.

Step 2: Find the pattern in your top five

Take your five highest-scoring clients and look at what they shared before they hired you. Industry and company size, yes... but more importantly, the trigger: what was happening inside their business when they decided to buy? A funding round, a failed launch, a new competitor, a sales team with nothing to sell against?

The trigger matters more than the demographics. Two companies can look identical on paper; the one living through the trigger is your buyer.

Step 3: Write the exclusion list

This is the step everyone resists, and the one that changes everything. Write down who you don't serve: "We don't work with X, Y, or Z." On paper, agreed on, visible to the team.

Your ICP only becomes real when the exclusion list exists. Until then, it's a wish, and the first tempting off-profile deal will erase it.

Step 4: Run the 5-second test

Show your homepage to someone who fits the profile you just defined. Can they say "this is for me" within five seconds? If they hesitate, your ICP lives in your head, not in your message — and everything downstream of the homepage inherits that vagueness.

Step 5: Rewrite one asset at a time

Now — only now — does the marketing work start, and in this order: homepage headline first, sales deck second, content third. All of it written for that one reader.

Resist redoing everything at once. The headline alone changes what your pipeline looks like, and it tells you quickly whether the new ICP definition holds before you invest in the rest.

What happens when it's done right

One of our clients, a development company, went through exactly this process. Same team, same budget — the difference was a sharper definition of who they were for, and the discipline to let every asset follow from it. Over the next twelve months, their LinkedIn engagement grew 3.7×, and their team grew from 20 to 120 people. Not because they posted more. Because the right people finally recognized themselves in what they published.

The one-sentence test

Before you spend another dollar on marketing, answer this: who is your product NOT for?

If you can answer in one sentence, your marketing problems are probably real marketing problems — fixable with better execution. If you can't, start with Step 1. The fifteen-client spreadsheet takes an afternoon, and it will tell you more than a quarter of campaign data.

And if you'd rather have an outside set of eyes on the diagnosis — that's literally what our discovery calls are for. Thirty minutes, no pitch, just the question most teams can't ask themselves from the inside.