How do we figure out what makes our company different?

You figure out what makes your company different by collecting episodes instead of opinions: the specific deals you won, the exact words your best customers used when they chose you, and the moment the founder decided this thing had to exist. Then you keep only what's true, what a buyer actually cares about, and what the option they'd otherwise pick can't honestly claim.

Most teams try it the other way around. They book a half-day offsite, put "what makes us unique?" on a whiteboard, and walk out with a list of adjectives every competitor already uses. The difference was never in the room's opinion of itself. It's sitting in transcripts, lost-deal notes and customer emails, and the job is to dig it out.

Why can't we see what makes us different?

Because you're too close to it. Founders and leadership teams know the product so well that the genuinely unusual parts feel ordinary, and the ordinary parts feel like features worth bragging about. Ask a CEO what makes the company different and you'll get the version rehearsed for the last board meeting: "we're easier to use, we have better support, and our platform is end to end."

Every competitor's CEO says the same three things. That answer is a summary, and summaries drift toward the category average because they're built out of category words. The specific, slightly odd truth that actually wins deals almost never survives being summarized. We wrote about why discovery interviews keep returning answers you already had in why our messaging interviews give us answers we already had.

Where is our difference actually hiding?

In our experience it lives in four places, and none of them is a brainstorm:

1. 1The founder's origin episode. Skip the mission statement and find the moment: what was the founder doing, on what day, when they decided the existing way was broken enough to quit a job over?
2. 2Your best customers' words. The people who renewed, expanded and referred you already explain you better than your website does. We dug into that gap in why customers describe us better than our website.
3. 3The deals you lost. A lost deal tells you which difference didn't land, or which one the buyer never heard.
4. 4The alternative. A difference only exists relative to something. If you can't name what the buyer would do without you, you can't name what's different about you.

How do I find my product's unique value?

Here's the five-step version we'd run if we only had two weeks and no budget:

1. 1Name the real alternative first. April Dunford's positioning method in Obviously Awesome starts here for good reason: list what customers would actually do if you didn't exist. Usually that's a spreadsheet, an agency, an in-house hire or doing nothing, and only rarely your closest competitor.
2. 2Pull ten episodes from your own team. Ask your top reps, "Tell me about the last deal you won that you thought you'd lose. What did the buyer say right before they said yes?" Write down the buyer's exact words before anyone interprets them.
3. 3Interview five of your best customers. Three questions: what were you doing before us, what almost stopped you from buying, and how would you describe us to a peer? Record it and transcribe it.
4. 4Mark the repeats. Read every transcript and highlight any idea that shows up three or more times in roughly the same words. Those repeats are the raw material. Anything only leadership says goes in a separate pile.
5. 5Run three filters. Keep only what's true (you can prove it), what matters (a buyer would change a decision over it), and what's yours (the alternative from step one can't honestly say it). What survives all three is your difference.

Steps two and three do most of the work, and the question style decides whether they work at all:

A team workshop on "our unique attributes" has its place, and it's the last step, after the evidence is on the table. Run it first and it simply re-lists what leadership already believes.

What does a real difference look like on paper?

Take a hypothetical $20M revenue-cycle software company. The leadership offsite produces "AI-powered, easy to implement, best-in-class support." That's three claims every competitor makes, and an AI engine asked to compare them would find nothing to separate them.

Now the episodes. Four of five customers independently mention the same thing: the billing team stopped working Saturdays within a month. Two lost-deal notes say the buyer "didn't believe the implementation timeline." The founder's origin story is a night spent re-keying denied claims at her family's practice. Put those together and the difference stops being a feature list. It becomes a point of view: billing teams shouldn't need weekends to keep cash flowing, and here's the proof it's possible in 30 days. That sentence is true, a buyer cares about it, and the spreadsheet-plus-overtime alternative can't claim it.

Why does this matter more now that buyers ask AI?

Because AI engines repeat the words they can attribute to you, and most companies give them nothing to attribute. In PitchKitchen's 2026 Healthtech Messaging Index, we scored 302 company homepages on 19 criteria. 176 explained clearly what the product does. Only 52 of those 176 also owned a term, frame or category in their own language that an AI engine could quote back. Across all 302, just 63 did.

That gap is what "we can't figure out what makes us different" looks like from the outside. When your homepage is built from category words, an engine asked "who should I talk to about this?" has no sentence that belongs only to you. It lists you as another option, or it lists someone else. If you want a quick read on where your own homepage lands, the free Brand Signal Score runs the same 19 criteria in about two minutes.

What do we do with it once we find it?

Write it down once, in one place, and make everything else inherit from it. The difference you just dug out belongs in your Magnetic Messaging Framework: who you're for, the problem you end in their words, the alternative you replace, and the proof. Your reps, your homepage and the AI tools your team writes with should all pull from that single source, or they'll drift back to the category average within a quarter.

Then test it. Ask five leaders to write the difference in one sentence, separately. If the answers don't match, the decision isn't finished yet, which is the diagnostic we laid out in how to know if you have a positioning problem. If they do match, put the sentence on the homepage and see if a stranger can repeat it back.

Your difference is almost never something you need to invent. It's already there, in the founder's story and your best customers' mouths. Does that make sense? The work is getting it out of their heads and onto the page before a competitor says it first.