When SaaS traction stops behaving as expected
Find out what changed before deciding what to fix
A SaaS company can have a product that worked, customers who valued it and a sales engine that produced growth — and then find that traction slows or becomes erratic.
The tempting response is to start fixing things.
Product thinks about the roadmap. Sales questions the pitch. Marketing changes positioning or acquisition. Leadership wonders whether the ideal customer profile — the ICP — is still right.
Any of them may be right.
Weakened traction is a symptom, not a diagnosis.
Traction Lab helps SaaS companies find out what has actually changed — using evidence from the customers, prospects and users who can distinguish between the possible causes.
So you can decide what to change on evidence, rather than on competing internal explanations.
What may have changed
The customers who created your traction may not be the customers who can create the next wave
Sometimes there is nothing fundamentally wrong with the product.
It may still work extremely well for the ICP that created the company’s original growth. There may simply be fewer of those customers left to win.
That can happen surprisingly early. A post-Series-A SaaS company can have found genuine traction and still discover that the market population fuelling that traction is becoming insufficient for its next stage of growth.
The company then needs to find its next source of customers.
A new ICP may have different problems and needs, expect different outcomes, or require a different product experience.
What worked before is evidence of what worked for the previous market. It is not proof of what will work for the next one.
Sometimes the company chooses to change ICP
A SaaS company may have plenty of growth left in its existing market and deliberately decide to pursue another one. SME to corporate, for example.
The strategic decision may make perfect sense.
The risk is changing the ICP without establishing whether the product still fits.
The new customers may share much of what existing customers value. Or their essential requirements may differ — in onboarding, time-to-value, UX, security, integrations, reliability, support or the core product itself.
The question isn’t whether the new ICP would like the existing product. It is: what does this ICP need in order to adopt, use and keep buying it — and how well the current product delivers that?
The answer may imply small changes or substantial ones. It may be possible to serve both populations with one product, or it may not.
We find the evidence before prescribing the architecture.
And sometimes the ICP changes without anybody deciding to change it
This is a different problem.
The company still believes it is serving one ICP, but part of the commercial organisation starts bringing another one to the product.
Sales begins prospecting larger companies. A marketing campaign attracts a different population. Growth, SEO or GEO changes who enters the funnel.
Nobody necessarily announces an ICP change.
The product starts being tested by people it wasn’t designed around.
The symptoms may appear later: falling conversion, failed trials, onboarding friction, slower time-to-value, different objections, poor adoption or churn.
That creates an important diagnostic question:
Is the product wrong for the ICP — or is the commercial system bringing the wrong ICP to the product?
Those are very different problems to solve.
How we find out
Talk to the people who can discriminate between the possible explanations
You don’t necessarily need dozens of user interviews.
You need the right comparisons.
Depending on the company, its lifecycle and the problem we are investigating, that might mean talking to some combination of core users, power users, recently onboarded customers, churned customers, competitor users, non-converting testers or prospects, corporate users, or people representing a possible next ICP.
We rarely need all of them.
The purpose is not to collect opinions.
It is to compare behaviour and experience across carefully selected populations and find out:
- Who actually created your historical traction?
- Who are you attracting now?
- What do the different populations genuinely need?
- What makes one adopt while another walks away?
- Which parts of the product create the outcome they value?
- Where do expectations and experience diverge?
- And which differences actually matter commercially?
A handful of well-selected conversations can sometimes reveal more than a large undifferentiated research programme.
Evidence strong enough to change the roadmap
Most companies I have worked with do not systematically test their important roadmap assumptions directly with the users who can validate or challenge them.
That doesn’t mean every feature should be put to a customer vote. Quite the opposite. Testing everything would be slow and wasteful.
The discipline is to identify the assumptions where being wrong would materially matter — and get evidence strong enough to carry the decision before committing significant product effort.
That can also expose something else.
Years of feature requests, important customers, Sales needs, competitive responses and sensible internal ideas can gradually make a product more complicated while obscuring the outcome it originally delivered particularly well.
The question then becomes:
Which parts of this product genuinely earn their place against the needs and outcomes of the ICP whose traction matters now?
That can mean adding something. It can also mean simplifying.
A larger corporate customer, for example, may require considerably more from security, integration, administration and support while simultaneously expecting a simpler experience and faster time-to-value from the ordinary user.
Company size does not tell us what the product should become.
The users do.
What you should know at the end
The purpose isn’t a research report. It is to make consequential decisions.
You should have materially stronger evidence about:
- Who generated your historical traction.
- Who you are actually attracting and selling to now.
- Whether those are still the right people.
- What problem or needs matter most to them.
- What outcomes they expect.
- Where the current product satisfies those requirements and where it doesn’t.
- What is most likely weakening traction.
- And what needs to be changed or tested next.
Just as importantly, explanations held by Product, Sales, Marketing, Growth, Customer Success and leadership can be tested against the same outside evidence.
The goal isn’t to make everyone agree.
It is to make disagreement better informed by reality.
Why Traction Lab
I have spent much of my career around the moments when companies have to work out what the market is actually telling them before making the next important decision.
I co-founded and helped build a technology company from zero, raised millions in funding, found product-market fit across demanding European and US software markets, and experienced both scaling traction and what happens when a company subsequently gets important things wrong.
Since then, I have worked with hundreds of founders, technology leaders and product teams.
The method is deliberately practical.
We start with what the company already knows. Product, Sales, Marketing, Growth, Customer Success and leadership often hold different pieces of the answer.
Then we identify what we don’t know, which uncertainties actually matter, and which people outside the building can give us evidence strong enough to resolve them.
We don’t ask customers to run your company.
And we don’t replace experienced judgement with research.
We make sure that when an important decision rests on an assumption about the market, reality gets a serious vote before you bet on it.
The longer-term opportunity
Fixing a traction problem is the immediate job.
There can be a useful consequence.
A company that learns how to identify its highest-risk product and market assumptions, find the right people to test them with, and use that evidence across Product, Sales, Marketing and leadership has acquired something more durable than the answer to one problem.
It has strengthened its ability to detect when the market is changing — and respond before assumptions harden into expensive decisions.
That capability can compound.
It starts with something much less abstract:
Traction isn’t behaving as expected. Let’s find out why.
Address
44, rue Laborde
75008 Paris
France