What the typical SaaS tech stack looks like

Three layers emerge from the data. The first is measurement and site infrastructure, which is close to universal: Google Tag Manager appears on 80% of homepages (95% CI 74-85%), Google Analytics 4 on 74%, and a consent management platform on 54%.

The second is paid media. Google Ads conversion tags appear on 56% of homepages, the LinkedIn Insight Tag on 50%, the Meta Pixel on 46% and the Reddit Pixel on 39%. These figures are recorded before cookie consent and are therefore lower bounds; the subsample analysis below shows how much they rise once consent is given.

The third is the marketing layer: automation, account intelligence, chat and routing, testing, customer data platforms and behaviour analytics. It is thinner than vendor landscapes suggest. The median company runs 2 detectable marketing-layer tools and the mean is 2.8. Across all categories, including infrastructure, the median is 8.

No single marketing-layer tool reaches half the sample. Marketo comes closest at 40% (95% CI 33-47%). This long tail is consistent with the vendor landscape tracked by chiefmartec, which counts thousands of marketing technology products, very few of which reach broad adoption.

Adoption by category

Grouping tools into categories shows how mature each layer of the SaaS tech stack is. Figure 2 reports the share of companies running at least one detected tool in each category.

Adoption falls steadily from infrastructure to the more specialised layers. Marketing automation and account intelligence each reach roughly half of the sample. Testing and personalization tools appear on just over a quarter of homepages, and client-side customer data platforms on 14%.

Marketing automation: Marketo leads by a wide margin

Of the 183 companies, 109 load a marketing automation platform that can be detected from the homepage. Within that group the distribution is concentrated.

Marketo accounts for 67% of detected platforms (95% CI 58-75%) and HubSpot for 26% (18-35%). Salesforce Account Engagement (Pardot) and Oracle Eloqua together account for under 8%. The confidence intervals do not overlap, so the gap between Marketo and HubSpot is not an artefact of sample size.

Marketo is detected through its Munchkin tracking script, which Adobe documents as the standard Marketo visitor-tracking method. It appears across security, developer tools, collaboration and data platform companies. HubSpot is the more common platform in two segments: customer support software and AI companies. Both segments have fewer than 15 companies in the sample, so these segment-level patterns should be read as indicative rather than conclusive.

One interpretation, which the data is consistent with but does not prove, is that platform choice tracks go-to-market model: companies selling large contracts through field sales teams standardise on Marketo, while earlier-stage or product-led companies use HubSpot. Readers comparing platforms may find our guides to choosing marketing automation software and CRM vs marketing automation useful.

Account intelligence and conversational tools

Nearly half the sample (48%) loads at least one account intelligence or enrichment tool. 6sense leads at 25% (95% CI 19-32%), followed by ZoomInfo at 21%, Clearbit at 13% and Demandbase at 10%.

These tools travel with Marketo. Of the 73 Marketo companies, 34 (47%) also run 6sense; of the 28 HubSpot companies, 8 (29%) do. The most frequent pairs of marketing-layer tools in the dataset are Marketo with Qualified (35 companies), Marketo with 6sense (34) and 6sense with Qualified (31). Taken together, these three tools form the most common enterprise pattern in the sample.

In the conversational category the shift is pronounced. Qualified appears on 31% of homepages (95% CI 24-38%). Drift, whose domain now redirects to Salesloft following its acquisition, appears on one. Chili Piper, which routes form submissions directly to a sales calendar, appears on 7%. The data shows which tool companies deploy today; it does not by itself show why companies switched.

Advertising pixels

The most unexpected result is the Reddit Pixel. At 39% (95% CI 33-47%) it sits within seven points of the Meta Pixel, and twelve companies load Reddit’s pixel without Meta’s. Adoption is concentrated among developer tools, data and security companies, whose technical buyers are active in Reddit communities.

The extended observation also recorded advertising platforms outside the main library. Microsoft Ads tags appear on 58% of the subsample and X (Twitter) ads tags on 29%, both larger than often assumed for B2B.

Analytics, testing and customer data

Google Analytics 4 appears on 74% of homepages; Adobe Analytics on 3%. Among behaviour analytics tools, Microsoft Clarity (20%) is more common than Amplitude (14%), Contentsquare (10%), Hotjar (9%) or FullStory (6%). Clarity is free, which likely explains part of its reach.

Only 27% of companies run a testing or personalization tool, led by Optimizely (14%) and VWO (8%). Client-side customer data platforms appear on 14%, led by Segment (11%). The CDP figure should be read with care: many data-mature teams now collect events server-side and activate audiences from a data warehouse, which is not visible from a website. A low client-side figure is therefore not evidence of low investment in customer data. Our customer data platform guide covers the trade-offs between the two approaches.

How large is a typical stack?

Counting all detected tools, the mean company runs 7.8 and the median 8. Counting only the marketing layer, the mean is 2.8 and the median 2. The distribution is wide: 37 companies show no marketing-layer tools and 11 show seven or more.

The largest observed marketing-layer stacks belong to Goldcast, Guru and Freshworks (eight tools each), followed by Gong, Klaviyo, Smartsheet, GitLab, Culture Amp, Algolia, Whatfix and Yotpo (seven each). The companies with no detectable marketing-layer tools include developer-first businesses such as GitHub, Vercel, Sentry and CircleCI, and AI companies including Anthropic, Mistral, Hugging Face, ElevenLabs, Harvey and Cursor. For companies that grow mainly through product usage, a minimal stack can be a deliberate choice rather than a gap.

Differences by segment

Mean marketing-layer tools by segment range from 3.3 in security, collaboration, fintech and customer support to 1.9 among AI companies. Security companies lean most heavily on the Marketo, 6sense, ZoomInfo and Qualified pattern: ten of fourteen run Marketo and nine run Qualified. Most segments contain fewer than 30 companies and several fewer than 15, so differences between segments are directional. The searchable table below allows readers to inspect any segment directly.

Three observed stack patterns

The following patterns are our interpretation of the clusters in the data, offered as a framework rather than a finding.

The enterprise account-based stack

Marketo, an intent platform such as 6sense or Demandbase, ZoomInfo, Qualified, and LinkedIn and Google advertising, often with a testing or session replay tool. Most common in security, data platforms and collaboration software, where contracts are large and sales teams target named accounts.

The HubSpot growth stack

HubSpot for automation and forms, a free behaviour analytics tool such as Clarity, and Google and LinkedIn advertising. More common among customer support, AI and earlier-stage companies.

The product-led minimal stack

A tag manager, analytics and few or no marketing-layer tools. Common among developer tools and AI companies whose growth is driven by product usage and community.

Cookie consent and page coverage: subsample results

To test how far the main figures understate real deployment, 73 companies were observed a second time with an extended protocol: the homepage before consent, the homepage after accepting the cookie banner, the pricing page and the demo request page. A cookie banner with an accept control was found and accepted on 45 of the 73 sites.

Accepting consent increased the number of detected tools on 22 of the 45 sites with a banner. Across the subsample, the mean number of detected tools rose from 9.3 before consent to 11.0 after it, and to 11.4 when pricing and demo pages were included. The share of companies with at least one advertising pixel rose from 63% to 81%.

Two conclusions follow. First, the advertising figures in the main study are conservative; true pixel deployment is materially higher. Second, page coverage matters less than consent: pricing and demo pages added at least one tool on 21 of 73 sites, but the average increase was small. Homepage observation captures most of a company’s marketing stack.

The extended observation also detected vendors outside the original library, including G2 buyer-intent tracking (26% of the subsample), the Vector visitor identification tool (16%) and interactive demo platforms such as Navattic and Storylane.

Of the 45 sites with a consent banner, 27 loaded at least one advertising pixel before consent was given. Consent requirements differ by jurisdiction and many companies apply region-specific rules, so this figure, observed from a single location outside the EU and US, should not be read as evidence of non-compliance.

Explore the data: all 183 companies

Methodology

Sample frame

The sampling frame comprised 215 B2B software companies drawn from the Forbes Cloud 100, the BVP Nasdaq Emerging Cloud Index and a list of publicly traded enterprise SaaS vendors. After exclusions the analytic sample is 183 companies across eleven segments. The frame over-represents well-funded, growth-stage and public companies and is not a random sample of all SaaS businesses.

Data collection

Main study: each company’s homepage was rendered in a desktop Chromium browser with full JavaScript execution on 2 and 3 October 2026, from a single location in India, without user interaction and without accepting cookie banners. Observations were recorded three to four seconds after navigation completed. Subsample: 73 companies were observed again on 3 and 4 October 2026 using the extended protocol described above.

Detection rule

A vendor was recorded as present when at least one network resource requested by the page, or the source of at least one script element, matched a domain or path signature unique to that vendor, for example munchkin.marketo.net for Marketo or js.qualified.com for Qualified. The main library covers 50 vendors in eleven categories; the extended protocol added 17 further vendors. Mentions of a vendor in page text, alt text or metadata were not counted. A raw-HTML baseline on the same frame detected a mean of 3.4 tools per site, against 7.8 under rendered collection, confirming that most marketing tags are injected client-side.

Reliability

Seventy companies were observed independently in both the main study and the subsample, on different days and through separate collection runs. Of the tools recorded for these companies in the main study, 97.1% were detected again on re-observation, and the mean Jaccard similarity between the two observations was 0.93. Differences arise mainly from tags that load conditionally, such as rotating advertising tags.

Research team

The study was designed and led by Shreepad Pujari, who developed the sample frame, the detection rules and the reliability check, and directed the analysis.

Exclusions and data cleaning

1. A vendor’s own product was excluded on its own domain, for example HubSpot on hubspot.com.
2. Domains redirecting to another company in the frame were merged: Drift and Clari resolve to Salesloft, SendGrid and Segment resolve to Twilio, and Chorus resolves to ZoomInfo.
3. Three domains were removed: one served a bot-verification page, one resolved to an unrelated company with a similar name, and one redirected to a parent company outside the frame.

Definitions and statistics

The marketing layer comprises marketing automation, customer data platforms, account intelligence, chat and routing, testing and personalization, and behaviour analytics. Category share is the percentage of companies with at least one detected tool in the category. Marketing automation platform share is calculated over the 109 companies with a detectable platform. Confidence intervals are 95% Wilson score intervals.

Limitations

• Only tools with a public web footprint can be observed. Absence means not detected, not not owned.
• Main-study figures are recorded before cookie consent and are lower bounds, particularly for advertising pixels.
• Data was collected from a single location; companies that deploy tags by region may differ elsewhere.
• The data describes a single point in time. We plan to repeat the study quarterly.

Data access, licence and citation

The full dataset is available as a CSV download and is published under a Creative Commons Attribution 4.0 licence. You may use, share and adapt it, including commercially, with attribution and a link to this page. Charts on this page may be embedded with the same attribution.

Disclosure. Unified Platforms is a marketing agency that implements and operates several of the tools named in this study. We have no commercial relationship with any vendor that affects the results, and no vendor reviewed or funded this research.

Related reading: martech stack, marketing technology stack, martech audit, martech consolidation, marketing technology trends and marketing attribution.

If you want an independent review of your own stack against this benchmark, our marketing automation team can run one.

Frequently Asked Questions

What is a SaaS tech stack?

A SaaS tech stack is the set of software a SaaS company uses to build, market and sell its product. In marketing it usually covers analytics, advertising, marketing automation, account intelligence, chat, testing and customer data tools. In this study of 183 B2B SaaS companies, the median company ran 8 detectable tools on its homepage, 2 of them in the marketing layer.

Which marketing automation platform do most SaaS companies use?

Marketo. Among the 109 companies with a detectable platform, 67% use Marketo and 26% use HubSpot. Salesforce Account Engagement and Oracle Eloqua together account for under 8%.

What does a typical B2B SaaS tech stack include?

Almost always Google Tag Manager, Google Analytics 4 and a consent manager; usually Google Ads, LinkedIn and Meta pixels. Enterprise-focused companies commonly add Marketo, an account intelligence tool such as 6sense or ZoomInfo, and Qualified for chat.

Is Drift still used by SaaS companies?

Rarely. Drift’s domain now redirects to Salesloft, and one of the 183 companies studied still loads Drift. Qualified appears on 31% of homepages.

Can I use this data?

Yes. The dataset and charts are published under a Creative Commons Attribution 4.0 licence. Please credit Unified Platforms and link to this page.

How was the data collected?

Each homepage was rendered in a desktop browser, and a vendor was recorded only when a network request or script source matched a signature unique to that vendor. A subsample of 73 companies was re-observed before and after cookie consent and on pricing and demo pages, and 97% of original detections were confirmed. The full methodology is set out above.