Most pre-PMF startups treat positioning as work for later. The pitch deck gets a tagline, the website gets something vague, and the plan is to figure it all out once the product is more defined. That delay is costly, and not for the reason you might think.
Why positioning matters before you have the answers
The argument for deferring is intuitive: if the product is still changing, committing to a position just creates something else to redo. What it misses is that you are already positioning yourself with every call, email, and demo. The question is whether you are doing it deliberately or by accident.
A weak or absent position makes product discovery harder. When you cannot articulate who the product is for and why they would choose it over the obvious alternatives, you can’t tell signal from noise. The early customer who stretches to fit, the champion inside a company that is wrong for you, the feature request from someone you should never have signed up: all of these compound a signal problem you already have.
Positioning is the lens you use to evaluate what you are learning. Without one, every data point looks equally useful.
The trap: trying to nail it too early
There is an opposite mistake. Founders who do think about positioning often treat it like a final answer: they pick a category, write a statement, and hold it at arm’s length because changing it would feel like failure.
That framing makes positioning fragile. At the pre-PMF stage, a position is a hypothesis about who needs what you have, why they would pay for it, and what frame of reference makes the value obvious. Hypotheses are meant to be tested and updated.
Positioning before PMF is not about being locked in. It is about being deliberate enough to learn something from every conversation.
The output is not a brand platform. It is a working document that says: we think we are for these people, solving this problem, and they currently handle it this way. Let’s find out if that’s true.
Four things to decide now
You do not need a full brand architecture before you have product-market fit. You need four decisions, held lightly, that let you run experiments and read the results clearly.
Who is this actually for? Not the broadest possible answer. The narrowest one you can defend. A startup for “small businesses” is positioning by default, not by design. Pick the most specific version of your best early customer and test every assumption against that profile.
What problem does it solve, in their words? This is the language test. You know you have traction with a frame when prospects say something close to your own description back to you, without prompting. Until that happens, you have a positioning hypothesis, not a position.
What is the obvious alternative they have today? The strongest early positions define themselves against a known behavior: the spreadsheet, the manual process, the incumbent tool the buyer hates but still uses. Naming the alternative makes your value concrete. It also tells you which comparisons are relevant and which are distractions.
Why would a buyer believe you over a competitor? This one is hard early. You may not have the case studies. What you do have is the mechanism: how the product actually works differently, who built it and why, what you refuse to compromise. Reason-to-believe can be structural even before it is proven by track record.
These four feed directly into a positioning statement template if you want a structured output from the exercise.
Holding your position loosely
Knowing when to update is the real skill. Two signals are worth watching.
The first is systematic misfit. If your sales conversations regularly surface a different use case, or if users are solving a different problem than the one you built for, the position has drifted from reality. That is data, not failure. The right response is to update the hypothesis and run the next round of conversations against the revised frame.
The second is the language gap. If the words you use to describe the product have to be translated by every prospect into something that resonates, the frame is off. The goal is the buyer who says “yes, exactly that” without you having to explain what you meant.
A useful exercise here is perceptual mapping: plot where your early customers perceive you relative to the alternatives they considered. If the cluster lands somewhere different from where you positioned yourself, the map is telling you something worth acting on.
What not to do
A few patterns that consistently make this harder.
Positioning against future competitors instead of current alternatives. The buyer is not comparing you to who you might face in three years. They are comparing you to what they use today.
Choosing a position based on what sounds ambitious rather than what is true. Bold framing is not a substitute for a real, defensible claim. The brand positioning guide has a useful test: can a competitor put your position on their site without it sounding false? If yes, you have not found an edge.
Writing the position for investors before writing it for buyers. These are different documents. The investor story is about market size and trajectory. The positioning is about why a specific person would switch. Conflating them produces something that serves neither audience well. For more on how positioning and marketing strategy diverge, brand strategy vs. marketing strategy draws that line clearly.
How Strynal approaches early-stage positioning
We work with founders who are still figuring out what they are building and for whom. An engagement here does not start from a finished product brief. It starts from conversations: with the founders, with early customers, sometimes with people who looked at the product and walked away.
From that, we build a working position. Not a final answer. A tested hypothesis with enough structure to run experiments, brief a designer, and get a homepage that does something other than exist.
Because strategy and positioning sit alongside design and build inside Strynal, the position we help you set gets stress-tested immediately in the work it has to direct. If it can hold up a homepage, a pitch deck, and a first outbound sequence, it is probably real. If it can’t, we find out before you do.