The SaaS Organic
Funnel Benchmark
What B2B SaaS organic funnels actually convert at — reported as distributions rather than a single misleading average. The methodology is published below; contributions are open and findings are not yet collected.
Why most SaaS benchmarks are useless
Most published SaaS conversion benchmarks report a single average across companies with wildly different motions, price points and lead definitions. Because the underlying data is heavily skewed, that average describes almost nobody. This benchmark reports medians and quartile boundaries by segment, and publishes the definitions contributors mapped their data onto.
There is a second problem, less often admitted. Every SaaS company defines a lead differently — some count a newsletter signup, others only a demo request from a qualified account. Averaging across those definitions produces a precise-looking number comparing fundamentally unlike things. We cannot eliminate that variance, but we can define the terms up front, require contributors to map onto them, and report the residual uncertainty rather than hiding it.
Definitions fixed in advance
Published before collection so contributors map onto the same terms, and readers can judge whether those terms match their own.
| Term | Definition used | Commonly confused with |
|---|---|---|
| Organic session | Non-paid search entry, excluding branded queries | All non-paid traffic including direct |
| Lead | Identified contact who requested contact or a trial | Any form fill, including content downloads |
| Qualified lead | Lead meeting the contributor’s own ICP criteria | Marketing-qualified by score alone |
| Customer | Closed-won with first payment received | Signed contract or verbal commitment |
| Organic CAC | Fully loaded content and SEO cost ÷ organic customers | Media spend only, excluding people cost |
| ACV | First-year contract value, excluding services | Total contract value across multiple years |
| Comparison query | Query naming two or more products, or “alternatives” | Any bottom-of-funnel query |
| Sales cycle | First identified touch to closed-won | Opportunity creation to close |
The third column exists because those confusions are the most common source of non-comparable benchmark data. Contributors confirm their mapping explicitly at submission.
Reported by segment, never pooled
A $2K-ACV self-serve product and a $200K enterprise platform have nothing useful to say to each other in a shared average.
By ACV band
The strongest determinant of funnel shape. Self-serve, mid-market and enterprise motions convert on entirely different curves and are reported separately.
By motion
Product-led and sales-led companies produce incomparable lead-to-customer rates, because the word “lead” means something different in each.
By stage
Early-stage companies with small samples show extreme variance. Reporting them alongside scaled companies distorts both.
Bars show the layout each metric will use. No values are shown because none have been collected — a distribution with invented numbers would be worse than no chart at all.
Metrics the first edition will cover
Committed in advance. Any metric dropped for insufficient sample will be listed as dropped, with the reason.
Visitor-to-lead by segment
Organic session to identified lead, split by ACV band and motion, reported across the full percentile range.
Lead-to-customer by segment
The stage where definitional variance bites hardest, which is why the mapping is confirmed at submission.
Comparison query contribution
Share of organic-sourced demos originating from comparison and alternatives queries versus educational content.
Organic vs paid CAC
Within the same company, fully loaded — the comparison most SaaS teams cannot currently make internally.
Time to first organic customer
From programme start to first attributable closed-won, which sets realistic expectations for new investment.
Content volume vs outcome
Whether publishing volume correlates with organic pipeline at all, once ACV and stage are controlled for.
Contributors get the useful part
Readers get the distribution. Contributors get their own position within it, which is the part that actually changes a decision.
Submission takes about twenty minutes and requires figures most SaaS teams already have in their CRM and analytics. Data is anonymised on receipt, no individual company’s numbers are published, and you can withdraw before publication for any reason.
There is no obligation attached and no sales sequence behind it. If contributing required sitting through a pitch, the sample would be biased toward companies willing to sit through a pitch — which would be a methodological problem as well as an irritating one.
What contributors receive
- Full findings ahead of public release
- Your own percentile position per metric
- Your segment cut alongside the aggregate
- The underlying dataset, anonymised
- Withdrawal rights up to publication
- No sales follow-up unless you ask
About the benchmark
What will the SaaS benchmark actually report?
Organic funnel performance by company stage and motion: visitor-to-lead and lead-to-customer rates, the share of demo requests originating from comparison versus educational queries, and how organic CAC compares to paid within the same company. Reported as distributions, not single averages.
Why report distributions rather than averages?
Because SaaS conversion data is heavily skewed and an average is actively misleading. Knowing the median and the quartile boundaries tells a team whether they are behind, typical or ahead. Knowing that “the average B2B SaaS conversion rate is 2.4%” tells them almost nothing useful.
How will you handle the fact that every SaaS company defines a lead differently?
By defining the terms in the submission form and requiring contributors to map their own stages onto those definitions. It is imperfect, and the published methodology will say so. Benchmarks that ignore definitional variance produce precise-looking numbers that compare unlike things.
Can we contribute our data?
Yes, and contributors get more out of it than readers do — an early copy plus their own position within the distribution, which is the part that actually informs a decision. Submissions are anonymised and no individual company’s figures are published.
Will there be a minimum sample before you publish?
Yes, and it will be stated with the results. Publishing a “benchmark” drawn from a handful of companies is how unreliable numbers enter circulation and get cited for years. If the sample is too thin to support a claim, we will report that rather than publish anyway.
Related material
Sources & references
- Definitions published on this page prior to data collection; revisions will be dated here.
- Gartner, “The B2B Buying Journey” — buying group size and pre-contact research behaviour.
- Google Search Central, Search Console documentation — organic session definitions.
- Oneskai revenue SEO methodology — attribution model used for organic CAC definitions.
- Oneskai B2B SaaS practice — the engagement context these definitions were developed in.
Find out where you actually sit
Contribute anonymised funnel data and receive your percentile position per metric alongside your segment cut — before anything is published publicly.