A basic audit framework for analysing your PLG status
Most teams know their product-led growth could be better, but they can't say exactly where it breaks. Trials go quiet and users sign up and never return, and the cause is spread across the website, the signup form, the first session, the pricing page and a set of emails nobody has looked at in a year.
An audit pulls those into one view. This is a simple version you can run yourself. It covers eight parts of the user journey, scores each one, and turns what you find into a ranked list of fixes.
What you need first
You need a clear picture of the target user and what they're trying to get done in their first session. You need a working definition of activation, meaning the early action that predicts a user will stay. If you don't have one, the audit should suggest a candidate. And you need your funnel numbers: visit to signup, signup to activation, activation to paid, and early retention. Rough figures are fine. Having no figures is a finding in itself.
Running it
Sign up as a new user with a fresh email address in a clean browser, and record your screen if you can. Go from the homepage to the point where you would upgrade.
As you go, keep a friction log. Write down each step, each decision the product asks you to make, anything that confuses you and anywhere you have to wait, with a timestamp at each stage. Then score the eight areas below from 1 to 5 and rank the fixes by impact and effort.
The eight areas
- First impression. Does the homepage say what the product does and who it's for in plain words? Is the main call to action obvious and the same everywhere? Can you see pricing? Is there believable proof that people like you use it? Watch for jargon headlines, 'book a demo' as the only route in, and hidden pricing.
- Signup. Count the fields. Check whether single sign-on is offered, whether a card is required (and whether that's a deliberate choice), and whether email verification blocks you from getting in. Long forms, and requests for company size or a phone number that never get used, are common problems.
- Onboarding. Look at what happens in the first minute. Good onboarding uses welcome questions to tailor the experience, offers templates or sample data, and has empty states that tell you what to do next. A blank dashboard or a ten-step tour are bad signs, as are setup tasks that need information the user won't have to hand.
- Activation and time to value. Is there a clear moment where the product proves itself, and does onboarding lead straight to it? Time how long it takes, and find out what share of signups get there. If nobody has defined that moment, or value sits behind an integration the user has to set up first, this score will be low.
- UI and usability. Check that navigation makes sense, labels use the customer's words, error messages help, and patterns are consistent. Check accessibility basics like contrast, and check the mobile experience. Dead ends and vague errors cost more than they look.
- Engagement and retention. What brings someone back on day two, and in week two? Look at whether emails and in-app messages respond to what the user has done or just go out on a timer. Look at how easy it is to invite a colleague, since collaboration is often the best retention feature a product has.
- Monetisation. Is the pricing page easy to follow? Do upgrade prompts appear when the user has just got value, or at random? Find out what happens when a trial ends and count the steps in checkout. Surprise paywalls, trials that end without warning and checkouts that need a sales call are the usual problems.
- Measurement and ownership. Is there a tracking plan? Can the team see the funnel by cohort? Does one person own signup to paid, with a target? Are experiments running? If nobody can tell you the activation rate, or two teams quote different numbers, score this low.
Scoring
| Score | Meaning |
|---|---|
| 1 | Missing or broken |
| 2 | There, but causing real friction |
| 3 | Works, with obvious room to improve |
| 4 | Good, with minor issues |
| 5 | Excellent |
The eight scores add up to a total out of 40. From 32 up, the foundations are strong and the next gains come from experiments and expansion. From 22 to 31 it works but leaks, so start with your two lowest areas. Below 22, fix onboarding and activation before you spend more on acquisition.
I'd treat the total as a starting point for discussion. The individual scores and the friction log are more useful.
From findings to a plan
For each issue, estimate the impact (how many users it hits and how hard) and the effort to fix it. Do the high-impact, low-effort fixes in the next two weeks and plan the high-impact, high-effort ones into the roadmap. Batch the small easy ones and park anything low impact and expensive.
What you should end up with is a one-page scorecard, your top five fixes, and a 90-day plan with an owner and a metric for each.
Where products usually fall down
The same few problems come up a lot. Activation has never been defined. Onboarding tours the features instead of getting the user to one result. Upgrade prompts appear before the user has seen any value. The analytics can't answer basic funnel questions. If you can only fix one thing, shorten the route to the moment your product proves itself.
It's hard to see your own product the way a new user does, which is the main limit of a self-audit. An independent audit gives you that outside view and a set of recommendations your team can act on.