Auditing a B2B SaaS

The method of a growth audit stays the same for every B2B SaaS. What changes with the motion is where you find the evidence: a product-led funnel lives in product events, a sales-led funnel lives in the CRM and in the calls. Audit a sales-led company from its analytics, or a product-led one from its pipeline report, and you will write a tidy document about the wrong stage.

For this chapter, say a B2B SaaS at 2M ARR with a free trial and a small sales team that picks up accounts above 50 seats. That shape forces you to audit both motions and the handoff between them, which is where the most interesting findings tend to sit.

Measure which motion actually brings the revenue

Companies describe themselves by the motion they want. Your first job is to measure the one they have. Pull twelve months of new ARR and split it by whether anyone from sales touched the account before the first payment.

SELECT
  CASE WHEN EXISTS (
    SELECT 1 FROM sales_activities a
    WHERE a.account_id = s.account_id
      AND a.occurred_at < s.first_paid_at
  ) THEN 'sales_touched' ELSE 'self_serve' END AS motion,
  CASE WHEN s.seats_at_purchase >= 50 THEN '50+' ELSE 'under_50' END AS segment,
  COUNT(*) AS new_accounts,
  SUM(s.first_year_arr) AS new_arr
FROM subscriptions s
WHERE s.first_paid_at >= CURRENT_DATE - INTERVAL '12 months'
GROUP BY 1, 2
ORDER BY new_arr DESC;

Those four rows often contradict the pitch deck. A “product-led” company may close most of its ARR through founders on calls, and a “sales-led” one may have a steady stream of small accounts buying alone that nobody looks at. A touch doesn’t prove the sale needed it, so keep this as a map of where to dig. The audit then treats each motion as its own business.

	Product-led	Sales-led
Where the evidence is	Product events, billing	CRM, call recordings, billing
Funnel stages	Signup, activated, paid, expanded	Lead, MQL, SQL, opportunity, won
Clock	Days or weeks	Months
Who to interview	Recent self-serve buyers, users who dropped during the trial	Reps, won and lost prospects
Classic blind spot	Counting signups without asking who signed up	Stage definitions that vary from rep to rep
What CAC must include	Marketing, onboarding effort, the cost of serving free users	Marketing, loaded salaries and commissions of the sales team

Audit each motion on its own clock

Sales-led cohorts need time before they say anything. Take the median length of the sales cycle from won deals, then exclude every cohort younger than that from your conversion rates. Otherwise the last two quarters look like a collapse when they are simply unfinished. Group by the date the lead was created, because grouping by close date mixes the campaigns of several quarters.

Then test the stage definitions. Ask two reps what makes a lead an SQL, and open twenty recent opportunities to see whether the required fields are filled. If the answers differ, every conversion rate above that stage is an opinion, and that is a finding in its own right.

Product-led cohorts move faster and fail differently. Most of the signal sits in the first days of use, so the work is in the activation chapter: the moment of value, the time it takes to reach it, and the share of signups that fit the target at all.

Compare the two motions on the same arithmetic:

CAC = (acquisition spend + loaded salaries of the people involved) / new customers
Payback in months = CAC / (new MRR per customer x gross margin)

Judge the result against the company’s own history, its cash position and the other motion. A published average tells you little about a business with your deal size and your cycle.

On inbound, open with a verdict

On an audit I ran for a B2B SaaS, traffic had good bases and conversion had barely been worked on, so the whole analysis centered on conversion and went well beyond a SEO checklist. I also wrote down what I had not dug into, and kept recommendations out of the areas where someone else was better placed. Your reader needs to know both.

B2B SaaS has a specific trap here: the people who read are often not the people who buy. A handful of pages can carry a large share of the traffic and attract the wrong company size, because the topics are too operational. Join first page, company size and CRM outcome, then read the result by page family. Illustrative numbers for our 2M ARR example:

First page family	Signups	Qualified	Won	Reading
Operational how-to posts	420	18	2	Volume, wrong company size
Feature pages	90	35	9	Small, and it sells
Alternatives and versus	40	22	7	Underbuilt

Some top-of-funnel posts will bring signups and few qualified leads, so judge them on qualified leads. Then walk the path. Compare the declared TOFU, MOFU and BOFU labels with what each page really does, as covered in the funnel chapter. Look for CTAs that send someone back up the funnel, or that sit at the bottom of the page where nobody sees them and nobody tracks them. BOFU belongs on landing pages, and if conversions happen on feature and pricing pages, articles should link to them by topic.

Long cycles make attribution worse. A demo request that follows months of reading gets credited to a brand search under last click, and the content that created the demand disappears from the report. Before concluding anything on channels, go through the tracking chapter. The quick fix is to track the CTAs and navigation of the blog and add an acquisition analytics tool. The advanced one is multi-touch, for example 40% first touch, 40% last touch and 20% shared across the middle touches.

For the competitive part, pick about three companies, each for a stated reason: market leader, fastest riser, different content strategy. B2B positioning drifts easily, so finish with the gap between the target you state and the one the market perceives. The positioning chapter gives the grid.

Audit the handoff between the two motions

In a hybrid company, money leaks at the point where a self-serve account becomes a sales account. Nobody owns that point, so it rarely gets looked at.

Check	How to measure it	What a finding looks like
Routing rule	List the signals that send an account to sales	Routing relies on company size alone and ignores product usage
Speed	Median time between the signal and the first human contact	Hot accounts wait days, and the trial ends first
Identity	Join product accounts and CRM records on domain	Duplicates inflate both funnels
Discounts	Compare prices on sales-touched and self-serve accounts of the same size	Reps discount deals that would have closed at list price
Expansion	Find who is responsible for seat growth on self-serve accounts	No owner, so growth happens only by accident

When the data can’t settle whether sales adds value on a segment, recommend a test. Hold a random share of mid-size accounts out of routing for at least one full sales cycle, then compare paid conversion and ARR per account between the two groups.

What the deliverable adds for a SaaS

Keep the usual audit structure and add one page per motion, one page on the handoff, and a list of what you left out of scope. Write every finding in the same four parts:

Finding	Evidence	Consequence	Action
Sales calls accounts that buy alone	New ARR by motion and segment	Reps have less time for 50+ seats	Route under 50 seats on product signals only, test for one cycle
How-to posts bring the wrong size	First page joined to CRM outcome	Traffic grows, qualified leads stay flat	Link these posts to use cases, invest in versus pages

It all ends in a prioritized action sheet, with each line assigned to a motion so that marketing, product and sales each know what they own.

Your next step: run the new ARR query above and show the four rows to the founder before any other analysis. The gap between that table and what the team believes tells you which motion to audit first.