SEO Metrics for SaaS: What to Track in Search and AI Answers

SEO metrics are the numbers that show whether search is growing your business. For a SaaS company in 2026, that means more than rankings and traffic. Buyers now research in Google's AI Overviews and AI Mode, ChatGPT, Perplexity, Gemini and Copilot. Many of them never click through to a website. A useful metrics set has to capture visibility in those answers as well as the demo requests and signups that follow.

This guide covers the metrics Flying Cat reports to its B2B SaaS clients. Maeva Cifuentes, Flying Cat's founder, leads strategy on every account, so these are the numbers the team puts in front of CMOs each month. They come in four groups. Business outcomes tell you whether search paid off. Demand metrics show whether more people know your brand. AI visibility metrics show whether assistants recommend you. Classic search metrics show whether Google can find and rank your pages.

The principle behind all of it is simple. Outcomes lead the report, and every other number explains them. D2L grew demo requests from organic search and LLMs 73% in its first year with Flying Cat. That figure, more than any ranking, is what its leadership team tracked.

Leading and lagging SEO metrics

Every SEO metric is either a leading indicator or a lagging indicator. Leading indicators move first. They tell you within weeks whether the work is on track. Lagging indicators confirm the result months later, in pipeline and revenue.

A client who asks after month one how they will know SEO is working needs the leading indicators. A CFO deciding next year's budget needs the lagging ones. A good report shows both and explains how one feeds the other.

Flying Cat's content team works to two primary KPIs on every account. One is the visitor to demo or trial conversion rate. The other is the SQL to opportunity conversion rate, using the client's own definition of a qualified lead. Average position, CTR, time on page and pages per session are secondary leading indicators. They help explain movement in the primary KPIs. The team never reports them as goals.

Business outcome metrics

These are the lagging metrics a leadership team cares about. Set them up first, because every other number in your report exists to explain them.

Conversions from organic and AI search

A conversion is a designated action that signals buying intent. For SaaS, that means demo requests, free trial signups, paid signups and contact sales forms. Clicks, page views and scroll depth are engagement, so keep them out of your conversion count.

In GA4, mark each of these actions as a key event. Google renamed GA4 conversions to key events and now reserves "conversion" for key events shared with Google Ads. For one client, an AI tax research platform for accountants, the main key event is generate_lead for free trial signups, with form_submission for demo requests. Pipeline attribution then runs through Salesforce.

Report conversions by landing page and by page type. The spread between page types is usually wide. At an enterprise procurement platform, one decision-stage page converted at 17%, against 0.24% for high-traffic educational pages. That comparison decided where the content budget went. The full story is in the enterprise procurement platform case study. Demo conversions from AI and search at that client grew 29% in a single month.

Segment AI referral sessions as their own channel too. In GA4, filter session source for chatgpt.com, perplexity.ai, gemini.google.com and copilot.microsoft.com. The conversion rates are often higher than Google's. At the AI tax research platform, visitors from chatgpt.com converted to free trial at 3.3%, against 2.0% for Google. At Leapsome, AI assistant visits converted to demo requests at more than twice the rate in Q3 2025.

Pipeline and revenue from search

Conversions count leads. Pipeline metrics count the leads your sales team accepted. Track sales-accepted opportunities (SAOs) or SQLs sourced by organic and AI search, plus the revenue they close.

Pipeline is the metric that survives a drop in traffic. Leapsome's non-branded clicks fell 36% in 2025 while its SEO-sourced non-branded SAOs grew 47%. A non-branded click became 2.3x more likely to turn into an SAO. Organic and AI-sourced SAOs grew 48% from the first half to the second half of the year. A traffic-only report would have shown a failing program.

Pipeline numbers need a CRM connection. Leapsome's team pulled first-touch attribution data to show which pages had sourced SAOs and customers since 2023. Templates, playbooks and software comparison pages led the list. At D2L, attribution tied LLM referrals to opportunities. Demo requests from ChatGPT grew 3.2x year over year.

One rule keeps pipeline reporting honest. For an AI customer service platform Flying Cat works with, every reported metric carries its HubSpot field name and exact filter. The client must be able to reproduce the number in HubSpot in two minutes, or it stays out of the report.

Self-reported attribution

Analytics cannot see a buyer who asks ChatGPT for a shortlist and then types your URL into a browser a week later. Self-reported attribution fills that blind spot. Add a "Where did you hear about us?" dropdown to your demo and signup forms, with options for ChatGPT, Google, AI Overviews, Perplexity, LinkedIn, a podcast and a colleague.

This is Flying Cat's main method for connecting AI visibility to pipeline. The AI customer service platform added the question to its web form and began reporting on it in its dashboard in October 2026. The answers are imperfect, because people forget and pick the first plausible option. Read the field as a trend line across quarters, alongside GA4 and CRM source data.

Sales calls add color. In one D2L deal that closed in 11 days, the provost had asked ChatGPT how Brightspace compared with the incumbent platform before the first call. That answer reached the senior leadership team. No analytics tool would have recorded that touchpoint.

Cost per acquisition and content efficiency

Cost per acquisition is your total SEO and content cost divided by organic and AI conversions. Include agency fees, content production, placements, tools and team time. Weigh it against volume, since a low cost on a handful of demos says little about growth.

Content efficiency is the average result per page across your site, measured in conversions as well as visits. When the average falls, refresh and consolidate existing pages before you publish more.

Demand metrics

Demand metrics show whether your brand is growing in buyers' minds. They lag behind visibility work, because a buyer who sees you in an AI answer or a ranking often comes back later by name.

Branded search demand

Branded queries include your company name, product names and their misspellings. Rising branded search means more people already know you when they start looking.

Google Search Console now splits this for you. The branded queries filter in the Performance report uses an AI-assisted system to classify queries as branded or non-branded, including misspellings and product names. Google announced it in November 2025 and made it available to all eligible sites in March 2026. Sites below its volume thresholds still need a manual regex filter.

Report branded and non-branded clicks as separate lines. Non-branded clicks are a leading indicator of new buyers finding you. Branded clicks are a lagging indicator of demand you have already created.

Direct traffic

Direct traffic has become an AI metric. Many visits from AI assistants arrive without a referrer and land in GA4 as Direct. Flying Cat tracks direct traffic growth as one of its four AEO signals for that reason.

The measurement shifts fast. In October 2026, after ChatGPT rolled out a new interface, measured ChatGPT sessions fell by roughly half in one week across several of Flying Cat's B2B SaaS clients. Those visits moved into Direct. ChatGPT still drove demo requests at the August and September pace. When referral data breaks like this, demo requests and engaged sessions are the steadier signal. Direct traffic absorbs the difference.

AI visibility metrics

AI visibility metrics measure how assistants describe your brand when buyers ask about your category. Flying Cat measures AEO results with four signals. AI visibility is how often the brand is recommended across an agreed prompt set. Citations count the client's own content used as a source. Direct and branded traffic growth shows the downstream demand. AI bot traffic shows how often AI systems crawl the site for retrieval and training. Mentions and citations rank ahead of clicks in the team's reports, because the answer itself shapes the shortlist.

Flying Cat's AEO guide chapter on measurement covers the setup in more depth.

Mention rate across a prompt set

Mention rate is the share of tracked AI answers that name your brand. You start with a prompt set, which is a list of the questions your buyers ask assistants, such as "best contract management software for mid-market companies". The prompt research chapter of Flying Cat's AEO guide explains how to build one.

AI answers vary from one run to the next, so a single check means little. Flying Cat runs each prompt about 75 to 80 times a month across eight AI assistants. A 40% mention rate on a prompt means the brand appeared in 40% of those answers that month.

Mention rate is the first AI number in Flying Cat's reports. D2L went from appearing in 12% to 33% of tracked AI answers. It also moved from 11th to 7th among learning platforms. The enterprise procurement platform went from being named in 76% to 88% of ChatGPT category answers in four months.

Break mention rate down by assistant, because each one draws on different sources. In its first two months with Flying Cat, the AI tax research platform's overall brand presence rose from 55.3% to 64.4%. Perplexity presence rose from 15.1% to 42.6% over the same window. Copilot presence rose from 59.9% to 80.6%.

Citations

A citation is a link to a source page inside an AI answer. Citations show which pages, yours and third parties', shape what assistants say about you. Track three numbers.

• Citation rate. The share of answers that cite your own domain. The AI tax research platform's citation rate tripled from 7.8% to 23.9% in two months. Its Google AI Mode citation rate went from 8.7% to 33.5%.
• Own-site share of cited sources. Your domain's share of every source cited in your category. For an ecommerce marketing platform, it rose from 14.9% to 20.5% in July 2026, while competitor sources fell from 6.1% to 3%.
• Cited third-party pages. The roundups, review sites and publications assistants cite most for your prompts. These become your placement targets. The enterprise procurement platform earned placements on those sources. ChatGPT answers citing its own site rose from 25% to 41%.

Citations also show which of your pages earn their keep. D2L's monthly AI citations of d2l.com pages grew nearly 4x in a single quarter.

Share of voice

Share of voice is your brand's share of all brand mentions across your prompt set, measured against a defined competitor list. Mention rate tells you how often you appear. Share of voice tells you how you compare.

The ecommerce marketing platform appeared in 41.7% of tracked AI answers in August 2026 and held the number-one share of voice at 23.0% among 16 tracked brands. For the AI tax research platform, citation share of voice went from 9.4% to 28.4% in two months. The same report showed it gaining 9.1 points of AI presence while its largest legacy competitors gained 2.3 and 2.8 points.

Position and sentiment

AI ranking position is unstable from one run to the next. Flying Cat reports visibility percentage across many prompts and treats position as a secondary signal. A useful version is the share of mentions where your brand leads the answer. At the enterprise procurement platform, 55% of AI mentions led the answer, up from 50% a month earlier.

Sentiment and accuracy deserve a monthly read of sample answers. Check that assistants describe your pricing, features and positioning correctly. A high mention rate built on an outdated description of your product will cost you deals.

AI bot traffic

Server logs and CDN analytics show how often crawlers such as GPTBot, OAI-SearchBot, PerplexityBot and ClaudeBot request your pages. Retrieval bots fetch pages to answer a live question. Training bots collect content for future models. A page that AI bots never request cannot be cited, so this metric flags crawl and access problems early.

Classic search metrics, read correctly

Google search remains a major source of organic demand for B2B SaaS. The classic metrics still belong in your report. AI features have changed how several of them behave, so each one needs context.

Impressions and clicks

Impressions count how often your pages appear in Google results. Clicks count the visits that follow. Both live in the Search Console Performance report.

AI features have pulled the two apart. Google began counting AI Mode clicks and impressions in Search Console totals in June 2025. It reports AI Overviews the same way, mixed into the main totals. A page cited in an AI Overview earns an impression even when nobody clicks. Pew Research Center found that Google users clicked a traditional result in 8% of visits to results pages with an AI summary, against 15% without one. Ahrefs measured position-one CTR as 58% lower when an AI Overview appears, comparing December 2025 with December 2023.

Read rising impressions with flat clicks as a visibility signal, then check conversions. For the ecommerce marketing platform, Flying Cat changed reporting in September 2025 to measure clicks, conversions and page-one visibility in place of raw impressions.

Keyword rankings and page-one visibility

Rankings show whether your target pages gain ground on the queries you chose. Track a defined keyword set in a rank tracker and count how many keywords sit in the top three and on page one. Rankings still feed AI answers, because AI Overviews and AI Mode draw on pages that rank in Google.

Weight your keyword set toward money pages. Mixmax grew keywords in positions 1 to 3 by 462% in ten months. Blog-assisted demo requests grew 500% over the same period. The rankings told the story because they were on topics close to a buying decision.

Average position in Search Console blends every query a page appears for, including irrelevant ones. Use it to compare a page against itself over time. Use your rank tracker for target keywords.

Click-through rate

CTR is clicks divided by impressions. It shows whether your titles and descriptions earn the click at a given position. Compare CTR within the same position band and query type. A page whose CTR drops while its position stays flat has usually gained an AI Overview above it, which calls for a check of the live results page before you rewrite the title.

Referring domains and authority scores

Referring domains count the unique websites that link to you. Growth in relevant referring domains to your priority pages is a leading indicator of rankings. Leapsome moved part of its content budget into link building. Flying Cat built around 50 targeted links to its priority pages. Referring domains grew about 30% in twelve months.

Third-party authority scores, such as Ahrefs Domain Rating, Semrush Authority Score and Moz Domain Authority, estimate link strength. Each tool calculates its own score. Google uses none of them. They help you compare sites within one tool. Report them as context and keep them out of your goals.

For AI visibility, track mentions on cited third-party pages separately from SEO links. A placement on a roundup that assistants cite can lift your mention rate without passing any link value.

Indexing and Core Web Vitals

The Page indexing report in Search Console shows which URLs Google has indexed and why others are excluded. Track the count of indexed priority pages and investigate any money page in "Crawled, currently not indexed".

The Core Web Vitals report covers three metrics. Largest Contentful Paint measures loading. Interaction to Next Paint measures responsiveness. It replaced First Input Delay as a Core Web Vital in March 2024. Cumulative Layout Shift measures visual stability. Google says Core Web Vitals are used by its ranking systems and also warns against chasing perfect scores for SEO alone. Treat them as a pass or fail check on your templates.

Search Console retired its separate Mobile usability and Page experience reports. Use Lighthouse or a crawler such as Screaming Frog for mobile checks. Rendering checks need a crawler too, since content that loads only through JavaScript can be invisible to AI crawlers that do not render it. Flying Cat's technical SEO service runs these audits.

Engagement rate

GA4's engagement rate is the share of sessions that lasted longer than 10 seconds, triggered a key event or included two or more page views. It replaced bounce rate as GA4's default engagement measure. Use it to compare pages of the same type and to spot content that fails to meet intent. It is a diagnostic metric for your content team.

Metrics to stop treating as ranking signals

Bounce rate, dwell time, time on page and pages per session appear in many SEO metric lists as ranking factors. Google has said repeatedly that it does not use Google Analytics data in rankings. John Mueller called the idea that Google looks at analytics bounce rate a misconception in 2022, as Search Engine Journal's review of the evidence documents.

These numbers also mislead on their own terms. A buyer who reads your pricing comparison, finds the answer and leaves counts as a bounce. A visitor who clicks through five pages because navigation is confusing looks highly engaged. Keep them as content diagnostics and remove them from your SEO goals.

How Flying Cat reports SEO metrics to SaaS clients

The metrics only help when they arrive in a format leadership trusts. Flying Cat's reporting follows a few rules across every account.

• Outcomes first. The enterprise procurement platform's reports led with demo requests from AI and search, with AI visibility as the leading indicator underneath. D2L's quarterly reviews tracked recommendation share across AI assistants next to pipeline.
• Month over month monthly, year over year quarterly. Monthly reports compare against the prior month. Year-over-year comparisons go in the quarterly business review, where seasonality evens out.
• A baseline before the work. In the first 90 days, the team sets a baseline in month one. It ships quick wins from the technical audit and keyword mapping in month two. In month three the team compares results against the baseline.
• Reproducible numbers. Every metric names its source, field and filter, so the client's team can rebuild it.
• What didn't work. Each report includes a standing section on what underperformed and what the team is changing.

Each quarterly business review covers performance against goal and a recap of the work. It also refreshes the onboarding assets, which are the content audit, competitor analysis, backlink audit and technical audit review. The review closes with an updated strategy and goal and the next two months of topics, links and refreshes. Flying Cat's digital analytics service sets up the GA4, CRM and AI visibility tracking behind these reports.

Frequently asked questions about SEO metrics

Which SEO metrics should a SaaS company track first

The metrics closest to revenue come first. Track conversions from organic and AI search, such as demo requests and signups, plus the pipeline they source. Then track leading indicators that predict those outcomes, including AI mention rate, citations, non-branded clicks and page-one visibility.

What is the difference between SEO metrics and SEO KPIs

A metric is any number you can measure, such as impressions or referring domains. A KPI is a metric you have tied to a business goal and a target, such as visitor to demo conversion rate. Treat the rest as diagnostics. A handful of KPIs keeps a report focused.

How do you measure SEO in AI search

Track mention rate, citations and share of voice across a prompt set in an AI visibility tool. Segment AI referral traffic in GA4 and watch direct and branded traffic growth. Add a "Where did you hear about us?" field to your forms to catch buyers who researched in an assistant without clicking.

How often should you report on SEO metrics

Monthly reporting with month-over-month comparisons works well for SaaS teams. Save year-over-year analysis and strategy changes for a quarterly review. Flying Cat usually sees AI visibility start to move within the first 60 days. New content often needs three to six months to rank for competitive terms.

Build a metrics set that ties search to pipeline

Start with the outcome your leadership team already tracks, which for a SaaS company is usually demo requests or signups. Make sure GA4 and your CRM record it by source. Add the "Where did you hear about us?" field. Then layer on AI visibility and classic search metrics as leading indicators that explain the outcome.

If you want help setting this up, Flying Cat offers SEO strategy and GEO and AEO for B2B SaaS companies in English, French, German, Italian and Spanish. Its case studies show how these metrics look over a full engagement. Book a call and the team will show you which numbers it would put in your first report.