I fix B2B SaaS
onboarding.

Signups arrive and never become customers. I read the whole path — landing page to the first moment your product is worth paying for — and tell you the one thing to change first, and why that one and not the other thirty-nine.

98%

of new users go inactive within two weeks — for half of all products measured

Amplitude, 2,600+ companies ↗
7%

returning on day seven is already the top quarter for activation

Amplitude, Product Benchmark Report ↗
−47%

day-one activation in B2B SaaS against the all-company benchmark

Amplitude benchmarks, mid-size B2B SaaS ↗
[ 01 ] The finding

Your signup form
is already fine.

The teardown →

The usual answer to a leaky trial is to shorten the form. So I built an instrument that counts everything countable on an activation path and pointed it at seven well-known B2B products.

0510Monday.com1Amplitude1Asana1Intercom2Pipedrive3ZendeskBot challenge before the formHubSpotConsent wall, never reached
Measured 2026-08-13. The axis runs to ten, roughly what a heavy B2B form asks — and nobody is near it.

What any tool can count

01Landing
02Invitation
03The form
No automated audit goes past here

Where the trial is won or lost

04Empty workspace
05Getting real data in
06Inviting the team
07First job actually done
08Day two
Three stages are cheap to observe. The other five need an account, a colleague and a second day, so no instrument has ever scored them.

The form is solved almost everywhere. The trial is lost after it — and none of that is countable, which is why it is still judged by a person.

That judgment is what I sell.

[ 02 ] Where this comes from

Case studies.

View all →

Onboarding

Platform Redesign — onboarding, after
Platform Redesign — onboarding, before
BeforeAfter
01 · 2024

Platform Redesign

OpenLM — Enterprise SaaS · Engineering

A monolith broken into modular services, with a new navigation backbone and the platform's first onboarding layer. Shipped after 4.5 weeks of research with 30+ license and IT managers.

Read the case →

Homepage

Homepage Dashboard — homepage, after
Homepage Dashboard — homepage, before
BeforeAfter
02 · 2024

Homepage Dashboard

OpenLM — Enterprise SaaS · Engineering

A legacy dashboard of raw tables became a configurable widget surface with role-aware defaults — after usability testing killed the first concept outright.

Read the case →

Concept → shipped

New Navigation — concept → shipped, after
New Navigation — concept → shipped, before
BeforeAfter
03 · 2024

New Navigation

OpenLM — Enterprise SaaS · Engineering

31 services across three competing mental models, resolved into one flat list. New-user time-to-find fell from 60–90s to 25–40s across four iterations.

Read the case →

Same read, pointed at your trial.

Ten working days from the first call. $1,800, fixed.

Send me your product URL →
[ 03 ] How it runs

Ten working days.

01

A 30-minute call

What counts as an activated account for you, and what your numbers say. If they already answer the question, I will tell you so and we stop there.

02

The instrument runs, then I do

Machine first, for the countable part. Then I use the product as a new customer would, with nothing set up — the part no tool can do.

03

The verdict, in writing and out loud

The written audit, the Figma, and a 45-minute walkthrough with your team.

And you leave with

  • The measurementsEvery step from your homepage to first value — what is asked, in what order, and where a visitor stalls.
  • The one thing to change firstAnd why that one, not the other thirty-nine. This is the part you are paying for.
  • The reworked flow in FigmaScreens, order and empty states, drawn well enough to build from rather than argue about.
  • The wordsLabels, empty states, the first-run message — usually written by whoever was free that day.
  • A build checklistFor whatever you already run. I am not selling you another tool.
[ 04 ] Price and fit
$1,800

The audit, fixed. 10 working days. No hourly billing, no scope conversation

from $6,000

The build-out afterwards — optional, and never assumed

What the build-out covers

  • Every screen and state of the path, not the one change: signup, empty workspace, first run, the invite, the first real job, day two
  • The states that get skipped and then get built badly — empty, error, loading
  • The copy for all of them, not just the one screen
  • Working sessions with your engineers while it is built, so it ships as drawn
  • One pass over what actually shipped, with a list of what drifted

Three to four weeks, because the number of screens on the path differs by product. I do not write the code.

Built for

  • B2B SaaS with a self-serve trial or free tier
  • Roughly 10–50 people, past your first customers
  • Signups arriving, too few of them ever becoming real
  • No designer in-house, or none with room for this

Not built for

  • Sales-led products with no self-serve path
  • Pre-launch products with no users to observe
  • Teams who want a tool rented to them monthly
  • Anyone who wants the deck rather than the change

If the audit does not give you at least one change you intend to make, do not pay for it.

I would rather carry that risk than argue about it later.

Send me the URL and I will look before we speak.

Two sentences is enough. You get a reply from me, not a scheduler.

Send me your product URL →
[ 05 ] Questions

Before you ask.

What is a B2B SaaS onboarding audit?
A fixed-price read of everything between your landing page and the moment a new account is worth paying for. You get the one change to make first, why that one, the reworked flow in Figma, and a build checklist for whatever tooling you already run.
How is this different from the free onboarding-audit tools?
Every automated audit is already built and already free. I ran mine across seven well-known B2B products and it found almost nothing — the signup form is solved almost everywhere. What is not automatable is judgment about the twenty minutes after it.
How much does it cost and how long does it take?
$1,800 and ten working days from the first call. If you then want the whole path designed and shipped rather than the one change, that build-out starts at $6,000 over three to four weeks — optional, and never assumed.
How much of my team's time does this take?
About two hours in total: a 30-minute call at the start, a 45-minute walkthrough at the end, and one person reachable in between for a test account and whatever your analytics already say. I do not run workshops and I do not need your roadmap.
What do you need from us to start?
A working trial or free-tier signup I can go through as a stranger, one sentence on what counts as an activated account for you, and whatever activation numbers you already have. If you have none, that is common and the audit still runs — the walkthrough is the measurement.
What happens after the audit?
You own everything: the written verdict, the Figma, the checklist. Most teams build the one change themselves, which is the point. If you would rather I designed the whole path instead, that is the build-out from $6,000 — never assumed, and never a retainer.
Will you sign an NDA, and do you publish what you find?
Yes to the NDA. Nothing about your product, your numbers or your screens goes on this site or anywhere else without your written permission. The public teardowns are of products I have no relationship with, audited from the outside like any other visitor.
We already have a designer. Why would we need this?
Usually because they are booked on the product your customers already paid for. Onboarding is the one flow nobody owns: it is built once at launch, then bolted onto for two years. An outside read of it costs less than a fortnight of the argument about who should do it.

Where do your signups go quiet?

Two sentences is enough. Send a product URL and I will run the instrument over it before we speak — whether or not you go ahead.

Goes straight to me. Used to reply to you and nothing else.

Prefer email? maxim@maximcocetov.com