When I set SEO KPIs with clients, four things always come up. The first is ranking movement on money keywords, say from #10 into the top three. Next come citations and recommendations when buyers ask ChatGPT or Google’s AI Overviews. Click-through rate and impressions are always on the list. Pipeline, the one everyone waits for, means leads who say they found the company on Google or through AI search.
Those are fair asks, but they don’t move at the same time. Impressions and rankings usually shift first, citations and the right visitors follow, and pipeline shows up last. Judge a two-month-old program on pipeline and a healthy one looks broken.
I’m a content strategist at Breaking B2B, writing for B2B SaaS teams like Softr, where one of my articles is worth about $130K a month in traffic value at its peak. This guide covers 19 SEO KPIs in the order they move, how to measure each one free in Search Console, GA4, and your CRM, the three to show leadership, and a scorecard template that sets targets from your own numbers.
What are SEO KPIs?
SEO KPIs are the few search numbers you attach a target to, so you can tell whether SEO is doing the job the business needs from it. A metric is anything Search Console or GA4 can count. It becomes a KPI when it has a target, an owner, and a decision that changes if it moves.
Impressions are a metric. Non-branded impressions for your product category pages, with a target of +30% by the end of the quarter, are a KPI, because a miss would change what you publish next.
Choose your SEO KPIs from your goal
Pick the goal first, then the KPIs. Every goal needs one lagging KPI that proves the result and two leading ones that tell you early whether you’re on track to hit it.
| Your goal | Lagging KPI (proves the result) | Leading KPIs (move first) |
| Leads and pipeline | Pipeline from organic and AI search | Target-keyword rankings, organic demos by landing page |
| Awareness in your category | Non-branded clicks | Non-branded impressions, share of tracked keywords in the top 10 |
| Visibility in AI search | AI recommendations for buyer prompts | AI citations of your pages, roundup listings and brand mentions |
| Local customers | Calls and direction requests from your Business Profile | Map pack rankings, Business Profile views |
Leading and lagging indicators are a familiar idea. What this guide adds is timing, so every KPI below says how long after the work starts you can expect it to change.
How SEO KPIs move: three stages
An SEO program works like a relay. Indexing and impressions run the first leg, rankings and the right visitors run the second, and pipeline runs the anchor leg. Each runner can only start once the last one has handed off, so judging the anchor leg in month two tells you nothing about the race.

The months are typical ranges for a site with some authority, not promises. A new domain runs slower, and a refresh of pages that already rank can skip ahead. The order holds either way.
Every KPI below gets the same treatment. You’ll see what it measures, how long after the work starts you can expect it to change, where to find it for free, and the trap that makes it misleading.
Stage 1 KPIs (months 1–3): Google is finding and showing you
1. Indexed pages and indexing errors
Indexing coverage shows how many of the pages you want in Google are indexed, and why the rest aren’t. A page that isn’t indexed can’t rank, so this is the first number to check after you publish or migrate anything.
When to expect change. Within days to a few weeks of publishing. New pages that sit in “Discovered, currently not indexed” for more than a month are telling you something about quality or internal links.
Where to find it. The Page indexing report in Search Console lists indexed pages and groups the rest by reason. Filter by your sitemap to see only the pages you submitted.
The trap. More indexed pages isn’t better on its own. Old tag pages and thin archives getting indexed can push the total up while your money pages wait.
2. Non-branded impressions
Non-branded impressions count how often your pages appear in Google for searches that don’t include your brand name. They’re the earliest sign that people who don’t know you yet are seeing you.
When to expect change. Usually within the first few weeks after publishing. Impressions often climb weeks before clicks, because new pages tend to appear at low positions, where they collect impressions but few clicks, before they climb.
Where to find it. Search Console has a branded queries filter that splits your data into branded and non-branded views using Google’s own brand classification. In the Performance report, click Add filter, choose Query, and select Non-branded queries. If your property doesn’t show it yet, add a query filter set to “Custom (regex)” and “Doesn’t match regex”, then list your brand name and its common misspellings.

The trap. Since AI Overviews launched, impressions and clicks often move apart. Google counts appearances in AI Overviews and AI Mode inside Search Console’s web search data, so impressions can rise while clicks stay flat. Read impressions as visibility, never as traffic.
3. Target-keyword rankings, current vs. target position
Target-keyword rankings show where the page you built for a keyword ranks today, against the position you agreed to reach. My clients ask about them often, and they’re worth tracking as long as each one is tied to a page and a target.
Rankings often get dismissed as a vanity metric, and they can be one when they’re a site-wide average or a list of 500 keywords nobody chose. Tied to a named page, a money keyword, and a target, they’re one of the clearest early signals you have. “Move the pricing comparison page from #10 into the top three for ‘[competitor] alternatives’ this quarter” is a KPI a CEO can follow and a writer can act on.
When to expect change. Long-tail and low-difficulty keywords can move within the first two or three months. Competitive head terms take longer, and refreshed pages that already rank on page two often move fastest.
Where to find it. In Search Console, filter the Performance report by query and by page, turn on Average position, and compare date ranges. A rank tracker adds daily positions and competitor movement, and rank trackers at every budget start at a few dollars a month.

The trap. Search Console’s average position blends every query a page appears for. A page can rank #2 for a niche phrase and #40 for the term you care about, and the average will flatter both. Always filter to the exact query.
4. Share of tracked keywords in the top 3 and top 10
Your share of tracked keywords is the percentage of the keywords you chose to target that rank in the top three and the top ten. It turns a long ranking report into one line that shows direction.
When to expect change. Once a few target pages start moving, usually in months two and three. It grows in steps as each cluster of pages matures.
Where to find it. Any rank tracker reports the share of tracked keywords by position bucket. For a free version, export Search Console queries for your target list into a sheet and count positions by bucket each month.
The trap. A share is only as good as the list behind it. If the tracked keywords include easy terms nobody buys from, the line will look great and pipeline won’t follow. Review the list with sales once a quarter.
5. Click-through rate on target pages
Click-through rate (CTR) is the share of impressions that turn into clicks. On a target page, it tells you whether your title and description win the click once Google shows you.
When to expect change. Right after you change a title or meta description. Give it two to four weeks of data before you judge it.
Where to find it. In the Search Console Performance report, filter by page and turn on CTR and Average position. Compare CTR between pages at similar positions, because a #2 result will always out-click a #8.
The trap. A low CTR next to a high position often means an AI Overview or other SERP feature is answering the query on the results page. Check the live results before you rewrite a title that’s doing its job.
Stage 2 KPIs (months 3–6): the right people arrive
6. Non-branded organic clicks
Non-branded clicks count visits from searches that don’t include your brand name. They measure the demand SEO creates, as opposed to the demand your brand, sales team, and ads already created.
When to expect change. A few weeks to a few months after impressions and rankings, once target pages reach the top five or so.
Where to find it. In the Search Console Performance report, click Add filter, choose Query, select Non-branded queries, and turn on Clicks. If that option isn’t there, use a “Custom (regex)” query filter that excludes your brand name. Google also added a branded vs. non-branded clicks card to the Insights report.
The trap. Total organic traffic mixes in branded searches, which rise with every product launch, ad campaign, and conference. A healthy total can hide a flat non-branded line, so report the two separately.
7. Engaged sessions on target pages
An engaged session is a visit that lasts longer than 10 seconds, includes a key event, or views at least two pages. GA4 defines engagement rate as the share of sessions that meet one of those conditions, and bounce rate as the share that don’t.
When to expect change. As soon as traffic reaches a page, so it’s a useful early read on whether the right people arrived.
Where to find it. In GA4, open the Landing page report, which sits under Reports in the Engagement or Generate leads section, depending on how your property’s reports are set up. Add a comparison for Session default channel group equals Organic Search and read Engaged sessions and Engagement rate per landing page.
The trap. Engagement shows that people stayed, not that they were buyers. A tutorial can hold readers for six minutes and never send one to a demo. Check the page’s organic signups and demo requests too before you call it a winner.
8. AI mentions and recommendations for buyer prompts
AI visibility for buyer prompts tracks how often assistants like ChatGPT name your brand, and how often they recommend it, when someone asks the questions your buyers ask. A mention is your name somewhere in the answer, and a recommendation is your name offered as the answer.
When to expect change. Slower than rankings and less predictably, often from month three onward. AI answers draw on pages that already rank, third-party lists, and brand mentions, so mentions and recommendations tend to grow once your pages rank and other sites start listing you.
Where to find it. Build a prompt set from the questions sales hears, run it on a schedule, and read the trend rather than any single run, because the same prompt can return a different answer an hour later. Paid trackers automate the runs, and the AI visibility tools worth paying for differ mostly in which engines they cover and how much each prompt costs. I track 33 buyer prompts for my own site in Peekaboo. In the 30 days to October 16, 2026, the prompt asking for a freelance writer who creates bottom-of-funnel SaaS content scored 69 out of 100 for visibility, while the prompt from someone whose CEO asks why competitors appear in ChatGPT scored 0. That gap is the KPI doing its job, because it shows exactly which buyer questions need content next.

The trap. A mention isn’t a recommendation. Being listed seventh in an answer that recommends a competitor counts as a mention in some tools, so check what your tracker counts before you report it.
9. AI citations of your pages
A citation is a link to one of your pages in an AI answer, shown as a source. Citations tell you which pages AI systems trust enough to point to, which is a different question from whether they mention your brand.
When to expect change. Alongside rankings. Pages that rank well for a topic are the usual candidates for citations on related prompts.
Where to find it. AI visibility trackers usually list cited URLs per prompt. For a free check, run your top prompts by hand in ChatGPT, Perplexity, and Google’s AI Mode once a month and log which of your URLs appear as sources. Getting cited in ChatGPT mostly comes down to pages that answer one question clearly, with dated facts and named sources.
The trap. Citations swing from week to week. Report a monthly or quarterly count across the whole prompt set, never a single week.
10. Referring domains to your money pages
Referring domains to money pages are the unique websites linking to your product, comparison, and pricing pages, plus the posts that feed them. It measures authority where you need it, not across the whole domain.
When to expect change. Gradually, from month three onward, as content gets cited, roundups list you, and outreach lands.
Where to find it. The Links report in Search Console shows top linked pages and top linking sites for free. Ahrefs and Semrush add referring-domain counts per URL and new and lost links over time.
The trap. A site-wide link count can grow from footer links, directories, and scraper sites while your money pages get nothing. Count referring domains per target page, and leave Domain Rating off the KPI list, since buyers never see it.
11. Brand mentions and category roundup listings
Brand mentions and listings show how often your brand appears in the places buyers and AI tools read when they compare options, such as “best [category]” roundups, review sites, community threads, and industry newsletters, linked or not.
When to expect change. Slowly, and mostly in response to deliberate work like digital PR, partnerships, and asking to be included in lists.
Where to find it. Search Google for “best [your category]” and “[competitor] alternatives” each month and log which lists include you. Google Alerts catches new unlinked mentions for free.
The trap. Volume means little if the mentions sit in places your buyers never read. One listing in a roundup your buyers read can be worth more than fifty mentions on press-release sites.
Stage 3 KPIs (months 6–12): search brings business
12. Organic signups and demos by landing page
Organic conversions by landing page are the trials, signups, and demo requests that come from organic search, split by the page where each visit started.
When to expect change. Once target pages bring steady non-branded clicks, usually from month four to six. Bottom-of-funnel pages like comparisons and alternatives tend to convert first.
Where to find it. Mark your signup and demo events as key events in GA4 under Admin, then Data display, then Events. Then open the Landing page report, filter to Organic Search, and read key events per page.

The trap. GA4 credits the session, not the buyer. Someone who reads your comparison page today and books a demo next week from a LinkedIn ad shows up as a paid conversion. Asking buyers how they found you, and checking the original source in your CRM, catches the visits GA4 misses.
13. AI referral sessions and conversions
AI referral traffic is the visits and conversions from people who clicked a link inside an AI assistant’s answer.
When to expect change. Once AI answers start citing your pages, since a visitor can only click a link that’s there. The numbers are small for most B2B sites, but these visitors often arrive with a clear question.
Where to find it. GA4 now has an AI Assistant channel in the default channel group, which covers referrals from assistants like ChatGPT, Gemini, and Copilot. Open Reports, then Acquisition, then Traffic acquisition, and read the AI Assistant row with key events turned on.

The trap. GA4’s AI Assistant channel excludes Google’s AI Overviews and AI Mode, which GA4 counts as Organic Search. It also misses everyone who read an AI answer and then searched for your brand. A “How did you hear about us?” field on your forms catches those buyers.
14. Self-reported attribution
Self-reported attribution is what buyers say when you ask how they heard about you. It’s the cheapest pipeline KPI there is, and the only one that sees the research buyers do in ChatGPT, in Slack groups, and in roundups before they ever click.
When to expect change. As soon as the question is on your forms. Expect a trickle at first, growing as more buyers find you through Google and AI answers.
Where to find it. Add a required, open-text “How did you hear about us?” field to your demo and signup forms, and have sales ask the same question on the first call. Once a month, sort the answers into Google, AI search, a named article or list, referral, and other.
I’ve seen how much one answer can be worth. A buyer who found an agency through a roundup I wrote told them where they’d found it, and the agency confirmed that one lead converted into a $36K ARR deal.

The trap. Dropdown menus produce lazy answers like “Google”, which can mean an ad, a branded search, or an article. Open text is messier to sort and far more honest.
15. Organic-sourced and organic-influenced pipeline
Organic-sourced pipeline is the value of open opportunities whose first recorded touch was organic search. Organic-influenced pipeline is the value of opportunities where organic search appears anywhere in the journey.
When to expect change. Last of all, in B2B usually from month six onward, because it waits on the sales cycle as well as on SEO.
Where to find it. In your CRM, using the original source field for sourced pipeline and page-view or form history for influenced pipeline. HubSpot fills its Original Source property automatically once the tracking code is installed, while Salesforce needs a lead source field mapped from your forms.
The trap. Each version flatters a different story. First touch overstates SEO when a buyer’s first search was incidental, and influenced pipeline credits SEO for deals it barely touched. Report both, say which is which, and add the self-reported answers from #14 beside them.
16. Paid-search equivalent value
Paid-search equivalent value estimates what your organic traffic would cost if you bought the same clicks through Google Ads. It translates rankings into the language finance already uses for paid search.
When to expect change. It climbs as you start ranking for commercial keywords, usually from month three, and becomes worth showing leadership from about month six.
Where to find it. Semrush calls it Traffic Cost, its estimate of what an advertiser would spend each month in Google Ads to get the same organic traffic, and Ahrefs reports a similar traffic value. For a free version, multiply your top non-branded keywords’ monthly clicks from Search Console by their cost per click in Google Keyword Planner.
These numbers get leadership’s attention fast. The knowledge management software roundup I wrote for Softr brought in traffic Semrush valued at about $130K a month at its peak, and my authenticator apps article for Zapier saves the company more than $20K a year in ad spend for the same visitors.
The trap. Traffic value is modeled, not money in the bank. Present it as “what we’d pay for these visitors in ads”, never as revenue.
17. Organic CAC, payback, and ROI
Organic customer acquisition cost (CAC) is what you spend on SEO divided by the customers it brings. Payback is how many months of their revenue it takes to recover that cost, and ROI compares the revenue with the spend.
When to expect change. Only once search has brought in closed deals to count, which in B2B usually means after the first year.
Where to find it. Your spend records and CRM. The math is simple, and the hard part is agreeing which deals count, so measuring content marketing ROI and proving it to leadership walks through the formulas and the attribution choices behind them.
The trap. SEO costs land up front, and the returns compound for years, so a first-year ROI understates every program that works. Show the trend by quarter, and set the payback expectation before the program starts.
Always-on KPIs for the team (not the board)
18. Core Web Vitals
Core Web Vitals are Google’s three page-experience measurements. A page passes when Largest Contentful Paint is 2.5 seconds or less, Interaction to Next Paint is 200 milliseconds or less, and Cumulative Layout Shift is 0.1 or less, measured at the 75th percentile of real visits.
When to expect change. Any time you ship site changes. Check it after every redesign, new script, or plugin.
Where to find it. The Core Web Vitals report in Search Console groups URLs as good, needing improvement, or poor, using real-user data.
The trap. Google treats page experience as one signal among many, so passing won’t lift rankings on its own. Treat it as a smoke detector, something the team watches so nothing slows the money pages, and keep it off the leadership slide.
19. Google Business Profile actions (local businesses only)
For businesses that serve customers in a place, Business Profile actions count the calls, direction requests, and website clicks that come from your listing on Google Search and Maps.
When to expect change. Within weeks of profile updates, new reviews, and local page improvements.
Where to find it. Open your profile’s Performance view and choose a date range. Data is only available for verified profiles.
The trap. Profile views rise with any local search volume. Calls and direction requests are the numbers that turn into customers.
SEO KPIs that look good and mean little
Some numbers belong on the team’s dashboard as diagnostics and nowhere near a leadership report. Each of these has a better KPI to report in its place.
- Total impressions. They rise with branded searches and AI Overviews. Report non-branded impressions instead.
- Site-wide average position. It averages hundreds of queries nobody chose. Report rankings for your target keywords instead.
- Total keywords ranked. Thousands of positions on page six add nothing. Report the share of your tracked keywords in the top 10.
- Domain Rating or Domain Authority. They’re third-party scores Google doesn’t use. Report referring domains to your money pages.
- Total organic traffic. It mixes brand demand with demand SEO created. Report non-branded clicks.
- Bounce rate on its own. A reader who finds the answer and leaves counts as a bounce. Report engaged sessions next to organic signups and demos.
Traffic as the north star is the first of the SaaS SEO mistakes that keep traffic from turning into demos, and these numbers are how it creeps back into reports.
The 3 SEO KPIs leadership should see
Leadership needs three numbers, each with last month, last year, and the target beside it. The rest belong to the team.
- Non-branded organic conversions. Signups and demo requests from searches that don’t include your brand name, which shows SEO is creating new demand, not just catching people already looking for you.
- Pipeline from search. Organic-sourced and organic-influenced pipeline from the CRM, plus the answers on your forms that name Google or AI search.
- One visibility number. Positions for your target keywords when the program is built on Google search, or how often AI assistants recommend you for buyer prompts when AI visibility is the goal.
Here’s how that looks on one slide. The numbers below are made up to show the layout.
| KPI | This month | Last month | Same month last year | Target | Comment |
| Non-branded organic demos | 14 | 11 | 6 | 15 | Two new comparison pages drove most of the increase |
| Pipeline from search (sourced / influenced) | $48K / $130K | $41K / $112K | $15K / $60K | $50K / — | Three deals named an AI assistant on the form |
| Target keywords in the top 3 | 9 of 25 | 7 of 25 | 2 of 25 | 12 of 25 | Pricing comparison moved from #9 to #3 |
In the first three months, before pipeline has had time to show up, swap rows one and two for early numbers like non-branded impressions and indexed pages, and say so in the comment. Leadership can live with early numbers as long as they know the program is still in its first months.
How to set SEO KPI targets
Set targets from your own baseline, not from industry benchmarks. Published benchmarks average sites with different authority, industries, and SERP features, and a CTR benchmark set before AI Overviews can make a healthy page look broken.
Start with the last three months of each KPI, then set three versions of the target.
- Conservative. What happens if the plan ships on time and Google changes nothing.
- Expected. What you’d forecast after looking at similar pages you’ve published before.
- Optimistic. What happens if the best pages break into the top three.
Ranges protect everyone, because three months of data is too little for a single confident number. Set ranking targets keyword by keyword, with the page and the position written down, and review them every quarter.
What to report when clicks fall and impressions rise
When clicks drop while impressions hold or grow, AI Overviews are the usual reason. Google answers more of the query on the results page, counts your appearance as an impression, and the searcher never needs to click.
Report it in three steps. Show impressions and clicks on the same chart so the gap is visible. Then show the KPIs that capture the value clicks used to carry, which are non-branded signups and demos, citations of your pages in AI answers, and form answers that name AI search. Finally, check whether conversions fell with clicks, because a drop in informational clicks often leaves demos untouched.
Here’s a sentence you can adapt for your report, with your own numbers swapped in.
“Clicks fell 18% as AI Overviews answered more top-of-funnel searches on the results page, while non-branded demos held steady and two buyers named AI search on our demo form.”
Agreeing SEO KPIs before the work starts
If you’re a freelancer, an agency, or a new hire, agree on KPIs in writing before you start. Without that, month two tends to get judged on revenue.
When I set KPIs with clients, we agree on four things. We name target keywords with a starting position and a target position, the AI prompts where we want to be cited or recommended, the CTR and impressions we expect for the pages we touch, and how we’ll capture leads who name Google or AI search. Each KPI also gets a timeline, so nobody expects pipeline in the first month.
Write the timeline into the proposal. A line like “Months 1–3 are judged on indexing, non-branded impressions, and target-keyword rankings” saves a lot of awkward calls later.
SEO KPIs for SaaS teams
SaaS teams can measure more of the funnel than most, because the product records what happens after a signup. Add these to your report from about month six, once signups start coming from search.
- Trials and demos by landing page. The SaaS version of #12, split by plan or product line if you sell more than one.
- Product-qualified leads from organic. Organic signups who reach the activation point your product team uses, such as creating a first project or inviting a teammate.
- Organic CAC payback against paid. Months to recover organic CAC compared with paid search, which is the comparison a CFO asks for at budget time.
The SaaS SEO process for B2B teams shows where these KPIs plug into the wider program, and the SaaS SEO checklist covers the GA4 and Search Console setup they depend on.
Free SEO KPI scorecard (Google Sheet)
This scorecard turns the guide into a one-page monthly report. Make a copy and fill in your own numbers.
- Pick your KPIs. Choose a goal from the dropdown, and the Sheet lists the KPIs that fit it, with where to find each one in Search Console or GA4.
- Baseline and targets. Enter your last three months, and the Sheet suggests conservative, expected, and optimistic targets. A ranking targets table holds each keyword, its page, the current position, and the target position.
- Monthly tracker. The team view, with every KPI by month and a color for on track, close, or behind.
- Leadership summary. The three KPIs, with last month, last year, the target, and a comment, ready to paste into a deck.
- KPI library. All 19 KPIs from this guide, with where to find each one, when to expect change, and whether it belongs in the team or leadership view.


Make a copy of the SEO KPI scorecard
Frequently asked questions
What’s the difference between an SEO KPI and an SEO metric?
A metric is any number your tools report, such as impressions, sessions, or backlinks. KPIs are the metrics with a target and an owner, chosen because they show progress toward a business goal. Every KPI is a metric, but only a handful of metrics should be KPIs.
How many SEO KPIs should you track?
Track three for leadership and five to eight for the team, chosen from your goal. The 19 in this guide are a menu, not a checklist, and tracking all of them at once spreads attention across numbers that don’t change any decision.
When should you start reporting SEO KPIs?
From the first month, using early KPIs like indexing, non-branded impressions, and target-keyword rankings. Waiting until pipeline appears leaves leadership with nothing to read for half a year, and starting early sets the expectation that KPIs change as the program matures.
How do you report SEO KPIs to a CEO?
Show three KPIs on one slide, each with last month, last year, the target, and a one-line comment. Lead with pipeline or conversions, keep rankings to the keywords that bring buyers, and say how long the program has been running and what should change next.
Which AI search KPIs should you track?
Start with AI recommendations for a fixed set of buyer prompts, AI citations of your pages, and AI Assistant sessions and conversions in GA4. Add the answers that name AI search on your demo form, since that’s where most AI-driven pipeline shows up.
Pick the KPIs for your stage
Begin with the goal, check how long your program has been running, and report the KPIs that should be changing by now. Leadership sees three numbers, and the team watches the rest.
If you need to show leadership what SEO is doing for pipeline, book a call with me. I’m a content strategist at Breaking B2B and a B2B SaaS writer, and on the call, we’ll look at what you report today and pick the KPIs that fit your stage.



