SaaS Marketing & SEO Statistics 2026: Spend, CAC and AI Buying Behaviour

B2B SaaS is the most heavily instrumented industry in marketing. Every company knows its CAC, its payback, its net revenue retention. Almost none of them know whether those numbers are good.

That is the honest problem with SaaS benchmarks. The medians move every year, and the gap between a bootstrapped $5m ARR business and a venture-backed one at the same revenue is now larger than the gap between industries.

What has changed in 2026 is where the funnel starts. Half of B2B software buyers now open a chat window before they open Google, and the shortlist they arrive with was assembled by a model that read review sites and comparison pages.

The figures below come from SaaS Capital's 2026 survey of more than 1,000 private B2B SaaS companies, ICONIQ's 2026 State of Go-to-Market, two G2 buyer surveys fielded in March and June 2026, First Page Sage's agency benchmark datasets and live Ahrefs data pulled in August 2026.

Key SaaS marketing statistics for 2026

• The median private B2B SaaS company spends 15% of ARR on selling costs and 8% on marketing, with selling up from 13% the previous year, according to SaaS Capital's 15th annual survey, completed March 2026 [1]
• Equity-backed SaaS companies spend 70% more on sales and 100% more on marketing than bootstrapped peers at comparable revenue (SaaS Capital) [1]
• Median growth for private B2B SaaS fell to 22% in 2025, down from 25% in 2024, per SaaS Capital [2]
• 51% of B2B software buyers now start research with an AI chatbot more often than Google, and 71% use one somewhere in the process, up from roughly 60% seven months earlier (G2, n=1,076, March 2026) [6]
• Generative AI chatbots are now the number one influence on software shortlists at 54%, ahead of review sites at 43% and vendor websites at 36% (G2) [6]
• 45% of buyers say a review site citation is the single most confidence-inspiring signal in an AI answer, rising to 50% among daily power users (G2) [6]
• High-growth SaaS companies draw 60% to 80% of pipeline from sales and channel, against 15% to 20% from marketing, per ICONIQ's survey of 150+ GTM leaders [4]
• Free trial and proof-of-concept conversion has jumped to roughly 50% in 2026, from about 36% in 2025, outperforming SQL and demo paths at 30% to 40% (ICONIQ) [4]
• Sales cycles have shortened by approximately six weeks, but contract durations are compressing at the same time (ICONIQ) [4]
• Evaluation is now the longest stage of the buying journey for 40% of buyers, up from 36%, and 49% have had an approved purchase vetoed by the CFO in the past year (G2, n=1,038, June 2026) [7]
• B2B SaaS organic CAC averages $205 against $341 for paid, the lowest of any of the 29 B2B industries measured by First Page Sage [8]
• G2.com carries 67,750 ChatGPT citations across 39,458 distinct pages, against 35,214 for Capterra and 19,601 for TrustRadius, per Ahrefs Site Explorer data pulled in August 2026 [13]

How much do SaaS companies actually spend on sales and marketing?

Around a quarter of ARR combined, but the split between sales and marketing is nowhere near even. SaaS Capital's March 2026 survey of more than 1,000 private B2B SaaS companies puts the median at 15% of ARR on selling costs and 8% on marketing [1].

Selling costs rose from 13% the previous year. Marketing did not move.

Source: SaaS Capital 2026 spending benchmarks [1].

Sales and support both rose whilst marketing stayed flat. That is a market buying more closing capacity and more retention capacity, and treating demand generation as a fixed cost.

The funding split explains most of the variance anyone will see in their own numbers. Equity-backed companies spend 70% more on sales, 100% more on marketing and 100% more on customer success than bootstrapped companies, and SaaS Capital found 83% of bootstrapped firms operating within two percentage points of breakeven or better against just 52% of equity-backed ones [1].

So when a founder asks whether 8% of ARR on marketing is right, the answer depends entirely on whose money is funding it. A bootstrapped company benchmarking itself against venture-backed spend will conclude it is underinvesting when it is simply solvent.

What is a realistic CAC and LTV:CAC ratio for B2B SaaS?

Lower than most B2B categories, and heavily dependent on channel. First Page Sage's dataset, drawn from client analytics accounts between January 2022 and August 2025 and updated in January 2026, puts B2B SaaS organic CAC at $205 and inorganic at $341, for a blended $239 [8].

That is the cheapest of the 29 B2B industries in the set. Financial services sits at $784, legal services at $749, manufacturing at $723 [8].

Within SaaS, though, the spread by vertical is enormous.

Source: First Page Sage, 22 SaaS verticals [8].

Fintech SaaS costs almost five times what legaltech costs to acquire. That is not a marketing competence gap. It is a function of contract value, compliance friction and how many well-funded vendors are bidding on the same buyer.

On the ratio, First Page Sage's guidance is that 3:1 to 4:1 is the healthy band and that anything above roughly 5:1 signals a company sacrificing growth for margin [12]. Their own B2B SaaS client benchmark sits at 6:1 with an average CAC of $728 across 41 clients, which tells you how much definitional choices matter [12].

Treat these as directionally useful, not authoritative. This is agency client data, weighted 75% organic by the firm's own admission, from a company that sells SEO. Their reported campaign ROI of 748% for SEO against 36% for PPC should be read with that clearly in mind [12].

Where does SaaS pipeline actually come from in 2026?

Mostly from sales, not marketing, at the companies growing fastest. ICONIQ's 2026 State of Go-to-Market, based on 150+ B2B software GTM leaders, found high-growth companies sourcing 60% to 80% of total pipeline from sales and channel motions, against 15% to 20% from marketing [4].

That will be uncomfortable reading for demand generation teams. It is also the clearest evidence yet that the 2021 marketing-sourced-pipeline orthodoxy has been quietly abandoned.

Self-serve is the counterweight. High-growth companies project self-serve at around 20% of 2026 revenue, roughly double the 10% their peers expect [4].

First Page Sage's funnel benchmarks across 50+ B2B SaaS clients show why organic keeps its budget even when it sources less pipeline.

Source: First Page Sage B2B SaaS funnel conversion benchmarks, June 2025 [11].

Look at the MQL to SQL row. SEO converts at 51%, PPC at 26%. The gap at the top of the funnel is three to one on visitor-to-lead, and it stays wide through qualification.

That is the real argument for organic in SaaS, and it is not the one most agencies make. Organic does not just cost less per lead, it delivers a lead that survives qualification roughly twice as often.

Do free trials or freemium convert better?

They convert at opposite ends of the same funnel. Freemium wins the signup, free trials win the payment, and opt-out trials win by a distance because the decision to pay is passive.

First Page Sage's 2026 freemium report, drawn from 80+ SaaS clients between 2022 and 2026, sets it out cleanly [10].

Source: First Page Sage, SaaS Freemium Conversion Rates 2026 [10].

An opt-out trial converts to paid at 49.9%, more than thirteen times the traditional freemium rate. It also converts visitors to signups at 2.4%, less than a fifth of freemium's rate.

The trap is obvious once you see both columns. An opt-out trial's conversion rate is flattered because a conversion is counted at a single paid month, which is why First Page Sage flags that the figure must be read alongside churn and LTV [9].

ICONIQ's number sits above all of these and measures something different. Free trial and proof-of-concept conversion at high-growth software companies reached roughly 50% in 2026, up from about 36% in 2025, ahead of SQL and demo paths at 30% to 40% [4]. That is a sales-assisted POC at meaningful ACV, not a self-serve signup, and the two should not be compared directly.

What do retention and growth benchmarks look like now?

Growth is down, retention is holding. SaaS Capital's 2026 survey put median growth for private B2B SaaS at 22% in 2025, down from 25% in 2024, with bootstrapped companies at 20% and equity-backed at 25% [2].

The share of companies reporting flat or negative growth was 7.3%, up marginally from 6.9% [2].

On retention, SaaS Capital's 2025 survey found companies in the $25,000 to $50,000 ACV band reporting median net revenue retention of 102%, with the top quartile at 111% and the bottom quartile at 97% [3]. ICONIQ, looking at a later-stage set, describes net dollar retention settling into a healthy 110% to 120% range [5].

The link between the two is the most useful number here. SaaS Capital calculates that moving NRR from the 90-100% band into the 100-110% band improves growth rate by five percentage points, and that companies with the highest NRR post median growth 173% above the population median [2].

Retention is a growth channel. That is arithmetic rather than a slogan, and it is why customer success spend rose whilst marketing spend did not.

Price point drives most of it. ChartMogul's 2023 analysis of over 2,100 SaaS businesses found only about 2% of B2C-style companies with ARPA under $25 a month reached NRR above 100%, against nearly 50% of B2B companies with ARPA above $1,000 a month [14].

How long is the SaaS sales cycle now, and where does it stall?

Shorter at the top, slower at the bottom. ICONIQ reports sales cycles shortening by approximately six weeks, driven by AI compressing the research phase [4].

But the friction moved rather than disappeared. G2's June 2026 survey of 1,038 buyers found evaluation is now the longest stage for 40% of buyers, up from 36% the year before, ahead of research at 36% and decision at 22% [7].

Once a vendor is selected, IT security review is the biggest source of delay at 39%, rising to 50% among enterprise buyers, followed by budget approval at 32% and implementation planning at 25% [7].

And the money got harder. Nearly half of buyers, 49%, had an already-approved purchase vetoed by their CFO in the last year, 75% now expect positive ROI within six months of signing, and 70% say the pace of AI innovation is pushing them toward shorter contracts [7].

For marketers this reframes the job. Content that wins the shortlist is no longer sufficient, because the deal now dies in a security questionnaire or a finance review.

How much do G2 and Capterra influence AI recommendations?

A great deal, and it is measurable on both sides. G2's March 2026 survey found 45% of B2B buyers say a review site citation is the most confidence-inspiring signal in an AI answer, rising to 50% among self-identified daily power users [6].

Review sites rank second only to AI chatbots as an influence on shortlists, at 43% against 54% [6]. In the separate June survey the two had converged, with review sites at 38% and AI chatbots at 37% [7].

The supply side matches. Ahrefs data pulled in August 2026 shows how heavily AI platforms lean on these domains.

Source: Ahrefs Site Explorer AI citation counts, subdomains mode, August 2026 [13]. Note that G2, Capterra and Software Advice are all part of the same group.

Three of those four properties belong to G2. That is a concentration risk worth naming out loud, because a review strategy in 2026 is largely a G2 strategy whether vendors like it or not.

The effect on outcomes is real. G2 found 69% of buyers chose a different vendor than expected because of AI guidance, and 33% purchased from a vendor they had never previously heard of [6].

What is AI search doing to SaaS category and comparison keywords?

It is moving the decision upstream of the click. Two thirds of buyers now open with a commercial prompt: 33% start category-based, 31% competitor-based, and only 22% with requirements or process questions [6].

Those are exactly the queries that used to land on a "best CRM software" listicle or a "Salesforce vs HubSpot" page. The classic category terms still carry serious volume, per Ahrefs data for the US in August 2026 [13].

Source: Ahrefs Keywords Explorer, US, August 2026 [13].

Note the difficulty scores on the comparison terms. "Hubspot alternatives" has a difficulty of 5 and a $5.00 CPC, "salesforce vs hubspot" a difficulty of 13. Cheap pages to rank, expensive clicks to buy, which is a strong signal that intent is concentrated there.

They are also the pages models read when a buyer asks for a comparison. G2 found that comparing vendor strengths and weaknesses is the most common reason buyers use AI chatbots for software research, ahead of basic category learning [6].

The practical implication is unglamorous. Comparison and alternatives pages have gone from a low-priority SEO chore to the primary input into an AI-generated shortlist, and 41% of buyers now run Deep Research tools regularly, meaning those pages get read in full rather than skimmed [6].

How to read these numbers

Almost none of this is a census. SaaS Capital surveys self-selected private companies, ICONIQ surveys its own network and portfolio-adjacent operators, and G2 surveys a panel of buyers who agreed to answer questions about software buying.

First Page Sage's benchmarks are agency client data, not industry data. The sample is 41 to 86 companies depending on the report, skewed toward firms that hired an SEO agency. Their SEO ROI figures should be treated as a claim, not a measurement.

Survey-reported buyer behaviour has a known optimism bias. People overstate how systematically they research, so the 51% AI-first figure is better read as "AI is now a mainstream first move" than as a precise share.

Ahrefs AI citation counts are estimates derived from a sampled prompt set, not a full accounting of what every model cites. Useful for relative comparison between domains, much weaker as absolute totals.

Finally, benchmark medians hide bimodal distributions. A 22% median growth rate in a population containing both 5% growers and 150% growers describes almost nobody.

What this means going into 2027

Marketing budget stayed flat at 8% of ARR whilst sales and customer success both rose. Expect that pattern to continue until marketing teams can attribute AI-influenced pipeline, which G2's qualitative interviews identify as the single loudest frustration among B2B marketers.

Comparison content is becoming the most valuable asset in SaaS SEO. Low difficulty, high commercial intent, and read directly by the models building shortlists.

Review platforms are consolidating into a chokepoint. With G2, Capterra and Software Advice under one roof and G2 alone holding nearly 68,000 ChatGPT citations, a thin review presence is now a discoverability problem rather than a social proof problem.

Retention remains the cheapest growth lever available. A ten point move in NRR buys five points of growth on SaaS Capital's numbers, a better return than most demand generation programmes will produce for the same money.

And the bottleneck has moved to the back half of the funnel. If evaluation is the longest stage for 40% of buyers and half have watched a CFO kill a signed-off deal, the marketing content that matters most in 2027 may be a security whitepaper and an ROI calculator rather than another category guide. If you want help working out how findable your SaaS business really is in search and AI answers, you can book a call with me.