Most founders with a stalled growth motion have already tried the obvious fixes. The problem is not the tactics — it's the absence of a diagnosis.
Most founders who come to me with a growth problem have already tried the obvious fixes. They ran more ads. They posted more content. They hired another SDR. Six months later, the numbers look the same.
The problem is not the tactics. The problem is the diagnosis — or rather, the absence of one.
This post is for Seed-to-Series-B founders and operators who know something is wrong but can't put a finger on what. I'll walk through how I actually diagnose a stalled growth motion, the most common places the real constraint hides, and how to tell whether you need an outside perspective or just better internal visibility.
Before you can fix a stalled growth motion, you need to be precise about what stalled.
Growth stalls in one of three places. Acquisition: fewer qualified people are entering your funnel than before (or than you need). Conversion: people are entering but not becoming customers at the rate required. Retention: customers are leaving faster than you can replace them. Each has a completely different fix. Conflating them is how you end up spending six months on demand gen when the real problem is a 40-day payback period killing your ability to reinvest.
The clearest sign of a mis-diagnosed growth problem is a founder who can't name the single number that should be moving. If you can't name it, you don't have a broken growth motion — you have a measurement gap. Fix that first.
There is an enormous market for tactical advice. Run this ad format. Use this email sequence. Attend this growth framework bootcamp. None of it is wrong, exactly — but all of it assumes the problem is execution, not diagnosis.
In my experience across acquisitions, UX work, and retention diagnostics, the constraint is almost never where the team thinks it is.
A B2B SaaS client came to me convinced they had a top-of-funnel problem. Pipeline was thin. The obvious answer was more leads — more outbound, more content, more paid. Before touching acquisition, I ran an audit on what was already happening to the leads they had. Sixty percent of qualified leads were dropping at the pricing page. Not because of pricing. Because the positioning on the pricing page had drifted so far from the language used in acquisition that qualified buyers assumed they'd landed on the wrong product. Fixing the page — not the ads — moved the needle.
The top-of-funnel problem was real. But it was downstream of a conversion problem. If we'd just poured more budget into acquisition, we'd have been accelerating into a leak.
This pattern repeats more often than most teams expect. The constraint is structural, not tactical, and it rarely announces itself.
A proper growth diagnostic has three stages: map, measure, hypothesize.
Draw the path a qualified customer takes from first awareness to paying and staying. Not the ideal path — the actual path. Where do they come from? What is the first meaningful action they take? What does the journey look like between that action and a closed deal? What happens in the first 90 days post-signup?
Most teams find two things during this exercise: steps they assumed existed that don't, and steps that exist but nobody owns.
Once you have the actual funnel mapped, put a number on each transition. What percentage of people who reach step A make it to step B? You're looking for the stage where the drop-off is disproportionately large relative to what it should be. This is almost always your constraint.
A healthy benchmark varies by model, but a rough heuristic: if more than 50% of qualified pipeline drops at any single stage, that stage is your bottleneck — everything else is a distraction until it's resolved.
Not "our content isn't good enough." That's a feeling. A diagnostic hypothesis looks like: "We believe qualified leads are dropping at the pricing page because the positioning language in our ads doesn't match the language on the page, creating a discontinuity that signals product-market mismatch to the buyer."
That hypothesis has an experiment attached to it. You can test it in two weeks. If you can't attach an experiment to your hypothesis, the hypothesis is too vague.
After running diagnostics across acquisition, UX, and retention problems, the same hiding spots appear repeatedly.
Your paid ads talk about one thing. Your homepage talks about another. Your sales deck says something else. Each of these was written at a different time, by different people, without a shared source of truth on who the buyer is and what pain you solve for them. By the time a qualified lead reaches a pricing or demo page, the message no longer coheres. The buyer loses confidence and leaves — not because of price, but because the product no longer seems to match their problem.
This is the most common constraint I find in B2B growth diagnostics, and it's almost never on anyone's radar because it develops gradually.
Teams optimise for what they measure. If you're measuring MQLs but not MQL-to-SQL rate, you will generate large numbers of leads that go nowhere. If you're measuring activation (did the user do the onboarding steps?) but not activation quality (did the user reach the aha moment in the product?), your activation numbers will look healthy while retention quietly collapses.
The diagnostic question is: what is the last metric before the one that matters, and is that metric actually predictive of the one that matters?
Growth motions that worked at an earlier stage don't always work at the current stage. A founder-led outbound motion that closed the first 50 customers becomes a bottleneck when you need to close 500. A content channel that drove signups at pre-PMF stage may not convert at the ticket sizes a Series B company needs. The team keeps investing in it because it used to work and stopping feels like giving up.
The diagnostic signal is a channel with high volume and declining conversion over time. That's usually not an execution problem with the channel — that's a signal the channel is exhausting its addressable audience, or the audience has changed.
"Marketing audit" is used to mean everything from a 30-minute Zoom call to a six-month agency engagement. Here is what a useful one actually covers.
A diagnostic audit covers three things: what you're doing (channels, content, campaigns in flight), what's happening (traffic, leads, pipeline, closed deals, churn — with transition rates between each), and what it costs (channel-level CAC, not blended CAC).
Blended CAC is the most common way teams hide channel-level problems. If organic and paid are both running, blended CAC can look healthy while paid CAC is 4x payback period and slowly killing the business.
The output is not a list of recommendations. It is a ranked list of hypotheses with evidence attached, ordered by expected impact. Each hypothesis has: the observation, the data that supports it, a proposed experiment, and a success metric.
A good audit takes 2-4 weeks. Less than that and you're summarising feelings. More than that and the team has usually moved on and the findings don't match current reality.
An audit tells you where the breaks are. It cannot tell you whether the market is large enough, whether the product has PMF, or whether the business model is sound. If the underlying product-market fit is weak, a diagnostic will surface it — but no diagnostic can manufacture demand that isn't there.
Finding the constraint is the first half of the work. The second half is executing against the right hypothesis without falling back into the pattern that created the problem.
The diagnostic usually surfaces three to five things that are broken. The instinct is to fix them all simultaneously. Don't. Multiple changes running in parallel means you can't attribute outcomes — you're back to not knowing what worked. Pick the highest-leverage hypothesis, run the experiment cleanly, and measure before moving to the next.
Diagnostics frequently implicate a channel, function, or person that the team has invested in. When the audit shows that the SDR team has a conversion rate problem rather than a volume problem, the head of sales will want to argue about the methodology. When it shows that the content channel is driving traffic but not pipeline, the content team will want to discuss attribution models.
This is normal. A diagnostic finding that nobody defends probably isn't a real finding. But the test of a good hypothesis is whether an experiment can settle it — not whether everyone agrees in advance.
After implementing a change, the temptation is to measure activity (we sent more emails, we published more content, we ran more ads). Measure the transition rate at the constraint you identified. Did the pricing page conversion rate change? Did trial-to-paid improve? Did time-to-close shorten? Activity without movement on the constraint metric means the hypothesis was wrong. That's useful data. Update the hypothesis and run the next experiment.
The decision to bring in outside help is usually made either too early (before you have enough data to know what's wrong) or too late (after you've burned 12 months of runway on the wrong problem).
You have real data — users, revenue, some pipeline — but the growth curve has flattened and internal hypotheses keep failing. You've run experiments and don't know why they're not working. You need someone who has seen this pattern before in a similar context and can compress the diagnostic cycle from months to weeks.
You're pre-revenue or pre-traction. At that stage, what looks like a growth problem is usually a product or positioning problem. A growth consultant working on a product that doesn't have PMF is wasted money. Fix the product first.
You want someone to execute a motion you've already validated. That's not a consultant — that's a hire. A consultant's job is to find the constraint, not run campaigns.
They should start with a diagnostic, not a proposal. If someone comes back to you after a 30-minute call with a full retainer proposal and a channel strategy, they haven't done a diagnostic — they've done pattern-matching against their previous clients. Your constraint may be nothing like their last client's.
Ask them: what would you need to see in the first two weeks to know whether your initial hypothesis is right or wrong? If they can't answer that precisely, they're not working diagnostically.
If you want to see how this plays out in practice, the case studies on this site are documented diagnostics — not agency success stories. The UX conversion engine engagement started with a client who thought they had a traffic problem. The lead acquisition engine started with a client convinced their sales team was underperforming. In both cases, the real constraint was somewhere else.
If you're at that stage — you know something is wrong, you've tried the obvious things, and the curve isn't moving — here's how I work.
Growth stalls in one of three areas — acquisition, conversion, or retention. The most common cause is a positioning or messaging disconnect, not a channel or budget problem. Diagnose which stage has the disproportionate drop-off before touching tactics.
A marketing audit maps your actual funnel, measures drop-off at each stage, and surfaces the constraint causing the stall. A rigorous diagnostic takes 2-4 weeks and covers channel performance, funnel transition rates, and CAC by channel — not blended CAC.
Hire when you have a validated growth motion that needs to scale. Bring in a consultant when the motion has stalled and you don't know why. A consultant finds the constraint and hands you a falsifiable hypothesis — not ongoing campaign management.
A growth marketing consultant diagnoses and redesigns growth systems across acquisition, conversion, and retention — rather than running a single channel. The job is to find where the leverage is across the full funnel, then recommend the highest-impact change.
Pre-PMF. If you don't yet have consistent evidence that a segment of buyers wants what you're selling, no diagnostic can manufacture that demand. Wait until you have traction — even thin traction — before investing in growth diagnostics.