How do you know your positioning is wrong?

How do you know your positioning is wrong?

You know it from the questions you get. When buyers repeatedly ask what you do after reading your homepage, when deals die comparing you to products you do not consider competitors, and when good prospects say the price feels high for what it is, that is positioning talking, not copywriting.

Positioning problems hide well because every symptom looks like a different department's fault. Traffic looks fine, so it must be the sales team. The sales team says the leads are bad, so it must be marketing. Marketing says the product is hard to explain, so it must be product.

Usually nobody is failing. The product is simply being introduced as the wrong kind of thing, and every downstream conversation inherits that confusion.

What is positioning, in the only sense that matters to a buyer?

Positioning is the answer to what kind of thing this is and who it is for. Not your tagline, not your feature list, not your brand. A buyer assigns you to a category in the first few seconds, compares you to whatever else is in that category, and judges your price against that comparison.

That assignment is the whole game, because it decides your competition for you. If a buyer files you next to a Google Sheet or an Excel file, your price is compared to a Google Sheet. If the same product gets filed next to a specialist platform, the same price reads as cheap. Nothing about the software changed.

This is why positioning is not a messaging exercise. Messaging is what you say once the buyer has already decided what you are. Positioning decides which shelf you are standing on, and you are standing on one whether you chose it or not.

Which symptoms point at positioning rather than at marketing?

Four symptoms point upstream. Buyers describe you back to you incorrectly. You keep losing to alternatives you did not plan for, including doing nothing. Your best customers look nothing like your stated audience. And your team cannot finish the sentence about who should not buy.

The first is the cheapest to spot. Read the notes from your last ten calls and look for the moment the buyer restates your product. If three different people restated it three different ways, the market has not agreed on what you are, and it will not agree on its own.

The second symptom shows up in the lost deal reasons sitting in your CRM, whether that is HubSpot, Pipedrive, or Salesforce. Losing to a named competitor is normal. Losing to a spreadsheet, an internal build on Airtable or Notion, or silence usually means the buyer never accepted the category you claimed, so your differentiators never got evaluated at all.

The third is the most uncomfortable. If your happiest accounts sit in one segment while your website speaks to another, your market has already told you where you fit and you are still arguing with it.

Why do sales calls tell you more than analytics here?

Because analytics measures behaviour on the page you wrote, not the frame in the buyer's head. A bounce tells you someone left. It does not tell you that they left because they thought you were a reporting tool. Call recordings from Gong, Fathom, or plain Zoom, plus your support tickets, carry the language that analytics cannot see.

When I take on a new engagement, the first thing I ask for is not the traffic data. I ask for the last twenty inbound enquiries and any recorded calls, and I read the product's listing on G2 or Capterra to see which category buyers already filed it under. Inside an hour you can usually hear the mismatch, because prospects will use a word for your product that appears nowhere on your site.

That word matters more than anything in your analytics. If buyers call it a scheduling tool and you call it a revenue platform, you have two options. Meet them where they are, or pay to teach a market a new word. The second option is a real strategy and an expensive one, which is exactly the trade covered in the decision between creating a category and entering one.

What does the test of who this is not for reveal?

It reveals whether you have made a choice at all. Write the sentence naming who should not buy your product, then show it to your sales team. If they flinch, or if the sentence excludes nobody, your positioning is still a wish rather than a position.

Strong positioning always costs you somebody. A product that is right for a five person team is usually wrong for a two hundred person team, and saying so is what makes the five person team trust you. Vagueness feels safe and reads as weakness, because a buyer who cannot tell whether a product is for them assumes it is not.

I use this in my own practice. I lead with AEO and GEO, AI automation, and B2B SaaS go to market, and I say plainly that Webflow is the build layer underneath rather than the headline. That sentence loses me the occasional enquiry from someone who wanted a pure design shop. It wins me the ones who wanted the thing I actually do, and I made that shift deliberately, which I wrote about in the lessons from repositioning my own practice.

Can your homepage pass the swap test?

Take your homepage hero, swap in a competitor's name, and read it again. If it still makes sense, it says nothing. That is the fastest positioning test I know, and most B2B homepages fail it, including homepages that took a month to write.

The failure mode is always the same. Words that describe a whole category get used as if they describe one product. Powerful, flexible, all in one, built for modern teams. Every one of those survives a name swap, which means none of them are telling a buyer which shelf you are on.

A second version of this test works on your pricing page. Cover the plan names and ask whether someone could tell which plan they belong in. When the answer is no, the confusion is rarely about price. It is about who each plan is for, which is positioning showing up in a place people think of as pricing.

What should you change first, the words or the audience?

Change the audience definition first, because the words follow from it. Pick the segment where you already win, write down what they had before you, and describe yourself as the better answer to that specific situation. Then rewrite the page. Doing it in the other order produces prettier sentences aimed at nobody.

The practical sequence I use is narrow, then name, then prove. Narrow to the situation you win in. Name the category the buyer already believes in. Prove it with evidence from customers who match that situation. Most sites try to do the third step without the first two, which is why proof on a vague page never lands.

Notice that none of this requires new features. Positioning work is mostly subtraction, and the hardest part is agreeing to stop describing the parts of your product that your best customers never use.

Does AI search change how positioning works?

It raises the cost of being vague. When someone asks ChatGPT, Perplexity, Gemini, or Claude which tool fits their situation, the answer is assembled from how clearly sources describe categories, audiences, and fit. A page that refuses to say what it is gives an answer engine nothing to repeat.

This is the part founders underestimate. A human visitor will squint at a fuzzy homepage and click around to work you out. A model summarising your site has no reason to do that work, and whatever it concludes becomes the description a buyer sees before they ever reach you.

So write the boring, explicit sentence. What it is, who it is for, what it replaces. That sentence is what gets quoted, and it is also what stops a page from ranking without converting, which is the same problem wearing a different hat.

How long before you know it worked?

Faster than most rebrands promise. Because positioning changes the questions you get, you can read the first signal in the next ten sales conversations rather than in a quarterly report. Watch whether buyers restate your product correctly and whether the competitor set you expected is the one they mention.

Pipeline metrics move later, and they move for many reasons, so do not wait for them to decide whether the change worked. The early evidence is qualitative and it is honest, because a prospect who describes you correctly back to you has understood something that was previously unclear.

If the questions do not change after a rewrite, you changed the words without changing the position. That happens often, and it is worth admitting quickly rather than defending for two quarters.

What should you do next?

Do the two cheap tests this week. Run the name swap on your hero, and write the sentence about who should not buy. If either one embarrasses you, you have found the highest leverage work available to your marketing team, and it costs nothing but agreement.

If you want an outside read on which segment you actually win in, that is a conversation I have most weeks with founders and marketing leads. Send me your homepage and the last few things buyers said about it, and I will tell you what shelf your page puts you on. Let's chat.