SaaS Marketing Statistics 2026: CAC, Retention, Trials

SaaS Market Size and Growth

Software spending is still growing at double-digit rates. Gartner’s October 2026 forecast puts worldwide software spending at $1.46 trillion in 2026, up 15.2% against 14.5% for all IT spending. Its last worldwide public cloud forecast, from November 2024, expected spending on SaaS to reach $299.1 billion in 2025, the largest segment of the public cloud market. These are forecasts rather than measured sales, and they cover all software, not only the subscription companies the rest of this page describes.

Growth at individual SaaS companies tells a different story. Benchmarkit’s 2026 survey found median ARR growth of 20% in calendar 2025, down every year from 30% in 2022. The 2025 KeyBanc Capital Markets and Sapphire Ventures survey projected a rebound to 20% from 15% in 2024. ChartMogul’s billing data shows how quickly growth fades: among 700+ private software companies that grew at least 20% in 2024, the median company kept 43% of its growth rate, so a company growing at the median 65% in 2024 grew 28% in 2025, and only 18% held or improved their growth rate.

AI-native companies are the exception. In High Alpha’s 2025 survey, companies with $1 million to $5 million ARR whose products have AI at the core grew 110% year on year, against 40% for those using AI as a supporting feature, and the gap held above $50 million ARR (38% against 16%). Reaching scale is still rare: in ChartMogul’s data on 6,525 software companies, 1 in 10 reach $10 million ARR within ten years of first monetizing.

CAC and Payback Benchmarks

Benchmarkit’s survey reports acquisition efficiency several ways, each with its sample size. In calendar 2025 the median B2B SaaS company needed 16 months of gross margin to recover what it spent winning customers, down from 18 months in 2024, and spent $1.63 in sales and marketing for each $1 of new-customer ARR, down from $2.00. The median Magic Number rose to 1.37, above 1.0 for the first time in the survey’s four-year series. Benchmarkit attributes the gain mainly to lower sales, marketing and R&D spending rather than to faster growth.

Contract size moves payback most: 11 months below $5,000 ACV against 22 months at $50,000 to $100,000. Fast growers were also the most efficient, with a 10-month median for companies growing more than 50% a year. For the wider set of acquisition-cost figures, including paid-media cost per lead and the CAC that public companies report, see our CAC benchmarks.

Payback definitions differ. Private companies usually divide spend by new-customer ARR, while public companies usually use net new ARR after churn, and Benchmarkit warns that the two are not comparable: 77 public SaaS companies showed a 34-month median in its 2025 edition against 18 months for private respondents. High Alpha’s 2025 survey shows why payback and retention belong together: companies with net revenue retention above 106% and payback under 10 months grew a median 71%, against 10% for those with retention below 98% and payback over 15 months.

Trial-to-Paid and Free-to-Paid Conversion

Trial and freemium conversion is where SaaS marketing and product meet, and the most quoted figures rarely say what they measure. ChartMogul’s Conversion Report, built on a January 2026 survey of 200 B2B software products run with ProductLed, defines free-to-paid conversion as the share of leads or free signups that become paying customers within six months. On that definition the median was 8%.

The entry model matters more than any single benchmark. Free trials were the most common entry point (57% of products, against 26% for freemium), most often lasting 14 days. Trials that ask for a credit card converted 30% of signups, more than five times trials without one, but they attract fewer signups. Per 1,000 website visitors, the survey’s averages give 10.5 paying customers for card-required trials against 5 for freemium and 3.6 for free trials.

Platform data shows how short the window is. In ChartMogul’s billing data, B2B trial conversions cluster around day 7, with a median 2.5% converting in the first week. Larger buyers rarely go fully self-serve: 80% of free-trial products in the survey bring in people when an enterprise user arrives, and ICONIQ reports 56% conversion from free trial or proof of concept at AI-native companies above $100 million ARR, against 32% for others.

Churn, GRR and NRR Benchmarks

Retention weakened in 2025. Benchmarkit’s median gross revenue retention, the share of recurring revenue kept after churn and downgrades, fell from 88% to 84%, the largest drop in its four-year series, and the top quartile fell from 95% to 91%. Median net revenue retention, which adds expansion, edged up to 102%. Smaller contracts churn more: companies selling below $5,000 a year kept a median 80% of revenue, against 91% at $50,000 to $100,000.

Go-to-market motion shows up in retention too. Product-led companies kept a median 79% and hybrid companies 80%, against 89% for sales-led companies (Benchmarkit’s chart; its text says 88%). KeyBanc Capital Markets and Sapphire Ventures describe gross retention near 90% and net retention above 100% in their 2025 survey.

ChartMogul’s billing data, drawn from the companies on its platform, shows lower retention than Benchmarkit’s survey. Among B2B SaaS companies above $250,000 ARR on ChartMogul, median NRR was 82% in 2025, and AI-native companies kept a median 40% of revenue gross, although AI products priced above $250 a month retained 70%. Retention also predicts growth: in ChartMogul’s data for the first half of 2024, companies with NRR of 100% or more grew a median 48%, about double the rest.

For what these rates mean for customer value, see our customer lifetime value benchmarks.

Net Revenue Retention at Listed SaaS Companies

Listed SaaS companies report net revenue retention in their SEC filings, each on its own definition: some measure annual recurring revenue and some revenue, some use a trailing 12-month average, and Zoom reports only its enterprise customers. The figures are not directly comparable, but they show the range. Platforms that grow with customer usage sit highest, with Snowflake at 126% and MongoDB at 122%, while products sold mainly by licence or seat sit near or below 100%: Zoom’s enterprise customers at 99% and Asana at 97%.

HubSpot, which sells marketing, sales and service software, reported 103.5% net revenue retention for 2025 across 288,706 customers paying an average $11,414 a year. At the self-serve end, Zoom’s online customers churned 2.9% of recurring revenue a month in the quarter to July 2026, while Salesforce put its attrition, the annualized contract value lost to cancellations and reductions over 12 months, at about 8%.

Expansion Revenue and Pricing

Expansion is now a large share of new revenue. In Benchmarkit’s 2025 data, upsell and cross-sell to existing customers made up a median 40% of total new ARR, rising to 62% above $50 million ARR. Hybrid companies that combine product-led and sales-led motions led at 54%, while product-led companies sat lowest at 36%.

Pricing architecture shows up in NRR. Companies with usage-based-only pricing had a median NRR of 108% in 2025, against 98% for seat-based subscriptions. For AI features, most vendors still bundle: 67% of SaaS companies with AI functionality did not charge for it separately, and among those that did, subscription plus usage was the most common model at 39%.

Timing matters as much as packaging. In ChartMogul’s data on 10.5 million purchases, new customers added seats at a rate of 0.83% in their first month, falling to 0.09% by month 11, and 40% of first-year plan upgrades came at the annual renewal in month 12.

Sales and Marketing Spend and Channels

The median B2B SaaS company in Benchmarkit’s survey spent 35% of revenue on sales and marketing in 2025, down from 37%, the first decline in the series. Companies growing 31% to 50% spent the most (44%), while those growing faster than 50% spent 35%. Hybrid product- and sales-led companies spent more than sales-led ones (39% against 34%), which Benchmarkit puts down to product-led companies adding sales and account management as they move to larger customers.

The nine listed companies on this page show a wide spread. Calculated from 10-K income statements, sales and marketing ran from 17.2% of revenue at Twilio to 51.5% at Asana in their latest fiscal years, with HubSpot at 44.1% and Salesforce at 34.5%. All nine cut the share year on year, Freshworks most (from 54.2% to 47.1%), and HubSpot went from 46.4% to 44.1%.

Marketing budgets alone are far smaller. Gartner’s 2026 CMO Spend Survey puts marketing at 7.8% of company revenue across industries, mostly at companies above $1 billion revenue, so it is a cross-industry figure rather than a SaaS benchmark. On channels, respondents to High Alpha’s 2025 survey named events as effective more often than any other channel in every ARR band below $50 million; above that, SEO and organic search led.

For paid-social costs on the main B2B channel, see our LinkedIn Ads benchmarks; for wider B2B data, see our B2B marketing statistics.

How SaaS Buyers Research: AI Search, Reviews and Content

How software buyers find vendors is changing faster than most marketing measurement. In G2’s March 2026 survey of 1,076 B2B decision-makers, 51% said they start software research with an AI chatbot more often than with Google, and 71% use AI chatbots somewhere in the process, although 80% still use Google too. AI chatbots were the most cited influence on vendor shortlists (54%), ahead of software review sites (43%) and vendor websites (36%), and 69% said a chatbot led them to a different vendor than planned. G2 runs a software review site, and its survey also found review-site citations to be the signal buyers trust most in an AI answer (45%).

Buyers also want less contact with sellers. In Gartner’s survey of 646 B2B buyers in August and September 2025, 67% preferred a rep-free buying experience, up from 61% in its previous survey, and 45% had used AI during a recent purchase.

Content teams use AI heavily but report mixed results. In the Content Marketing Institute and MarketingProfs survey of more than 1,000 B2B marketers for 2026, 9 in 10 use AI to create content, but fewer than 4 in 10 say it has improved performance. For how AI answer engines are changing search, see our AI search engine statistics; to build content that answer engines cite, see our content engine service.

AI in SaaS Go-to-Market

AI is now standard in SaaS products and go-to-market teams, though most vendors do not charge for it separately. ICONIQ reports that 70% of companies have moderate or full AI adoption in go-to-market workflows, led by meeting transcription, lead generation and content creation, while account-executive quota attainment stayed flat (58% in 2025 to date against 59% in 2024). In Benchmarkit’s survey, 35% of companies first added AI to their product in the past 12 months, and the median company directed 10% of R&D spend to AI-specific work.

Investment is rising. CMOs in Gartner’s survey put 15.3% of marketing budgets into AI initiatives in 2026, but only 30% report mature AI readiness, and over half of companies in the KeyBanc and Sapphire survey planned to raise AI budgets by 20% or more.

AI-native companies grow faster from the start: in ChartMogul’s data they were 3x more likely to reach $1 million ARR within six months, while 36% of High Alpha’s respondents now call AI core to their product. For an AI program tied to go-to-market metrics, see our AI transformation services.

Outlook: 2026 to 2027

Forecasts and plans point to more growth than companies have recently achieved. Gartner expects software spending to reach $1.68 trillion in 2027, up 14.6%. SaaS companies in Benchmarkit’s survey planned median growth of 26% for 2026 against the 20% they achieved in 2025, the fourth year in a row that plans ran ahead of results, and companies relying on outcome-based pricing planned 50%. Respondents to the KeyBanc and Sapphire survey expected EBITDA margins to turn positive and account-executive payback to shorten to 18 months by 2026.

Widely Quoted SaaS Figures, Checked

Several SaaS statistics circulate with a source that does not publish them, or measures something else. Each one below is checked against the publisher’s own report.

Sources

Every figure on this page comes from one of the sources below.