Stop judging SEO by a deadline. Judge it by evidence.
- 0
Fixed timelines in SEO. Google says some changes land in hours, others take months.
- 3
Links to check, in order: was the work done, did visibility grow, did it hit the business.
- 8/8
Sources agree leads and sales from organic search are the final test. Free tools are enough.
- What did you recommend, and what's actually live?
- How did non-brand clicks change versus last year?
- How many real leads came from organic search?
Flat line. Three very different stories.
You hired an SEO person or agency. There were targets. The date came, and the numbers didn't.
We've paid every month and missed the targets. Is our SEO bad, or were the goals wrong?
Real talk: a deadline can't answer that. Google doesn't even promise one. Its own starter guide says some changes take effect in a few hours and others could take several months.
What a flat line can do is hide what's really going on. Click through the three stories below. Same traffic, completely different reality underneath.
What's hiding under your flat line?
Interactive · tap a storyMost of the work never reached the site.
Recommendations are stuck with a busy developer or waiting on an approval. Traffic can't move if the changes aren't live.
Each story calls for a different move: push for execution, give it time, or change strategy. Firing your SEO in story two is as costly as renewing in story three.
So how do you know which story you're in? I started with what the experts measure.
8 trusted sources. One unanimous answer.
I didn't want another list of 30 metrics. I wanted to know which ones the people who define SEO actually agree on.
So I put Google's documentation for site owners side by side with the KPI guides from Ahrefs, Semrush, Search Engine Land, Search Engine Journal, Seer Interactive, Shopify and HubSpot. Then I counted. Click any bar to see who's in.
How many of 8 sources recommend each metric
Interactive · click a barThe only metric every single source recommends.
Didn't make the cut: rankings, Domain Authority, bounce rate and backlink counts. The sources disagreed, and some advised against them.
Business results win. Ahrefs calls organic conversions "the only indisputable SEO KPI."
Only Google talks to you, the buyer. Every other guide is written for SEOs reporting to their boss. That's the gap this guide fills.
Line these metrics up in the order things actually happen, and you get a chain.
SEO is a chain. Break one link and everything after it dies.
Work has to go live before Google can see it. Google has to show it before people click. People have to click before they buy.
Try it: click any link below to break it, and watch what happens downstream.
Break a link
Interactive · click to break / fixSemrush splits KPIs the same way. Links 1 and 2 are leading indicators: they move early, so you can fix things while there's still time. Link 3 is the lagging indicator: it confirms the result, but it always shows up last.
That's also why a missed deadline is so misleading. It only looks at link 3, the slowest one.
Now let's walk the chain, metric by metric.
7 metrics (+1 if you're local). Grade your SEO report as you go.
Step through each metric. For each one, tell me if your current SEO report actually shows it. You'll get a coverage score at the end.
Work delivered and verified
The share of your SEO's recommendations that are actually live on your site. Sounds basic. It's also the metric almost nobody checks.
Here's why it matters: a lot of SEO work never ships. In a 2024 survey of technical SEOs, the most common answer was that only 40–60% of their recommendations got implemented. Often the blocker isn't the SEO. It's a developer with no time, or an approval stuck in someone's inbox, sometimes yours.
| Recommendation | Owner | Status |
|---|---|---|
| Rewrite titles on 12 service pages | SEO | Live, verified |
| Fix redirects from old product URLs | Developer | Blocked |
| Add a new location page | You | Waiting approval |
- Where
- Ask for a dated change log. Spot-check a few items on the live site.
- Good sign
- Key items live; blocked items say why.
- Red flag
- No log. Just "we did a lot of work."
- Ask
- "What did you recommend, what's live, and what's blocking the rest?"
Sources: Google, Do you need an SEO?; SEOFOMO Technical SEO survey, 2024.
Indexing of your key pages
If Google hasn't indexed a page, it doesn't exist in search. Period. So before you judge rankings or traffic, make sure your service, product and location pages are actually in.
Google recommends checking Search Console about once a month, including the indexing report and Manual Actions. Skip the site-wide page count; Ahrefs says that mostly matters for big sites. You only care about your money pages.
- Where
- Search Console → Indexing → Pages
- Good sign
- All key pages indexed. New ones in within weeks.
- Red flag
- Money pages "not indexed" for months, no plan.
- Ask
- "Are all our key pages indexed? What's the plan for the rest?"
Sources: Google, Get started with Search Console; Google, Debugging drops in Search traffic; Ahrefs, SEO KPIs.
Non-brand clicks, versus last year
Clicks from searches that don't include your business name. People who type your name already know you. Non-brand clicks are new customers who found you because of SEO. That's the whole point.
Total clicks look flat at 1,000. Underneath, SEO is bringing fewer new people. That's what to ask about.
Always compare with the same months last year, not last month, so seasons don't fool you. And use clicks, not impressions: Google confirmed impressions were misreported from May 2025 to April 2026. Clicks were fine.
- Where
- Search Console → Performance → Branded filter: non-brand
- No filter?
- Smaller sites may not have it yet. Filter out your brand name and misspellings instead.
- Good sign
- Up versus last year, or steady while your market shrank.
- Red flag
- Only total traffic, and the growth is all brand.
- Ask
- "How did non-brand clicks change versus last year?"
Sources: Ahrefs, Semrush, Search Engine Land, Shopify KPI guides; Google, branded queries filter; Search Console data anomalies.
Search breadth
How many different non-brand searches send you at least one click. When it grows, Google is showing you for more of what your customers look for. And it usually moves before traffic does. Best early signal you've got.
Ahrefs puts search visibility at the top of its KPI list. The catch: SEO tools measure it with keyword lists your SEO picks. Counting real searches in Search Console uses Google's own data. Free, and impossible to cherry-pick.
- Where
- Search Console → Performance → Queries, non-brand → Export & count
- Good sign
- More searches over time, matching what you sell.
- Red flag
- Growth only in searches unrelated to your business.
- Ask
- "Which new searches are we showing up for?"
Sources: Ahrefs, Semrush, Search Engine Journal, Seer, Shopify KPI guides.
Where you rank
Ditch the single "average position" number. Count how many non-brand searches sit in positions 1–3, 4–10 and 11–20. Now you can see searches climbing toward the top.
Top-3 searches nearly doubled; the top 10 grew from 29 to 49. That's real progress, even before traffic catches up.
Average position hides all this. It can even "improve" while you lose visibility: when low-ranking searches drop off, the average goes up. Google says not to obsess over absolute position, and Ahrefs calls single keyword rankings an inferior metric.
- Where
- Search Console → Performance → Queries + Position → 3 buckets
- Good sign
- Money searches climbing into the top 10 and top 3.
- Red flag
- Only average position, or screenshots of hand-picked keywords.
- Ask
- "How many important searches moved into the top 3?"
Sources: Google, Debugging drops in Search traffic; Ahrefs, SEO KPIs; Search Engine Land, SEO KPIs.
Clicks to pages that sell
Traffic isn't revenue. This one only counts non-brand clicks landing on pages that can make money: services, products, categories, locations, pricing.
Search Engine Land lists "pages generating traffic" as a real-world KPI. Eventbrite judged its own SEO by traffic to event listings, the pages that sell tickets. And HubSpot is the cautionary tale: its blog traffic fell sharply, yet its CEO said in 2025 that only about 10% of leads came from the blog. Viral posts don't pay the bills.
- Where
- Search Console → Performance → Pages, non-brand → Group by type
- Good sign
- Pages that sell gaining non-brand clicks.
- Red flag
- All growth on blog posts nobody buys from.
- Ask
- "How much search traffic lands on pages that sell?"
Sources: Search Engine Land, SEO KPIs; HubSpot, blogging KPIs; Google case study, Eventbrite; HubSpot Q2 2025 earnings call.
Verified leads and sales
Calls, forms, bookings and sales from organic search, confirmed in your own records, and compared with what you pay. The one metric all eight sources agree on.
"Confirmed" is doing the heavy lifting here. Some "conversions" are just page views or button clicks, and Google notes that Search Console and Analytics count visits differently. Your CRM settles it. Because this link shows up last, don't make it your only test early on. But it's the final boss.
- Where
- GA4 → Traffic acquisition → Organic Search → Key events
- Then
- Match against your CRM, call log or sales records.
- Good sign
- Organic leads rising, and sales confirms they're real.
- Red flag
- No key events, or leads your sales team calls junk.
- Ask
- "How many real leads came from organic, and how do you know?"
Sources: Ahrefs, Semrush, Search Engine Land, Search Engine Journal, Seer, Shopify, HubSpot; Google, Using Search Console and Analytics data; GA4 Help, key events.
Google Business Profile actions
Serve a local area? Then plenty of your customers never touch your website. They call or tap "directions" straight from your Google Maps listing. Your profile tracks every one of those actions. Semrush and Shopify both list them as SEO KPIs.
- Where
- Your Business Profile → Performance
- Good sign
- Calls, directions and website clicks up versus last year.
- Red flag
- Never reported, or only "views".
- Ask
- "How many calls and direction requests came from our profile?"
Sources: Google Business Profile Help, performance; Semrush and Shopify KPI guides.
Now the fun part. Let's look at the numbers that make bad reports look good.
Vanity metrics: they look great and prove almost nothing.
These lines show up in a lot of SEO reports. Flip each card to see the catch.
Enough theory. Here's what real companies measured.
12 real businesses. Here's what they measured.
Telecom, marketplaces, local trades, B2B software. The question wasn't how big the wins were. It was which metrics they trusted.
Quick honesty check: not all evidence is equal. Only three of these were proper controlled tests. Filter the table to see who's who.
How strong is the evidence?
Interactive · filter| Company | Business | What they measured | Evidence |
|---|---|---|---|
| Vodafone | Telecom | Leads per visit, sales | A/B test |
| Visual discovery | Search visits per page group | Controlled | |
| Etsy | Marketplace | Search visits per page group | Controlled |
| ZipRecruiter | Jobs | Non-branded traffic, conversions | |
| Eventbrite | Events | Traffic to listing pages, YoY | |
| Rakuten Recipe | Recipes | Search traffic, time on site | |
| HubSpot | B2B software | Blog traffic vs share of leads | Company |
| Steel framing supplier | Local B2B | Leads, Maps views | Agency |
| Elite Tune | Local trade | ROI, organic traffic | Agency |
| TurnPoint Services | Home services | Organic revenue | Self |
| Sosemo clients | Agency portfolio | Non-brand traffic, YoY | Agency |
| HouseFresh | Publisher | Visitors after a Google update | Owner |
Put all twelve side by side and four lessons jump out:
Many SEO changes do nothing. Some hurt.
A Pinterest title test had zero measurable effect. Some Etsy title tests hurt traffic badly. A change log is how you catch the bad ones fast.
The pros measure money, not rankings.
ZipRecruiter tracked non-branded traffic and conversions. Vodafone tracked leads and sales. Eventbrite tracked the pages that sell tickets.
Traffic isn't the business.
HubSpot's blog lost most of its traffic, yet it brought in only about a tenth of the company's leads.
Some drops aren't your SEO's fault.
HouseFresh took about two years to recover from a Google update. Check if your whole market moved before pointing fingers.
You've got the metrics and the proof. Time to diagnose yours.
Remember the flat line? Find out which story is yours.
Three questions, one per link. You'll get a straight answer and your next move.
Visibility is up, but it isn't turning into money.
The SEO work is reaching people. Something between the click and the sale is breaking, or the target was set on the wrong thing.
Next: check your pages that sell, your offer and your lead tracking. Reset targets on Link 2 metrics while you fix it.
Grade your SEO report in section 04 to see your coverage score here.
My own framework, not an industry standard. "Long enough" depends on your site, your market and how much changed.
Run this check every month. In 10 minutes.
The SEO KPI Scorecard
Every metric from this guide in one file. Drop in this year's number and last year's, and the change calculates itself.
Download the scorecard ↓.xlsx · works in Google Sheets
- Where to find each metric + what to ask
- Monthly tracker, automatic year-over-year
- How to read results when targets are missed
- Every source, linked
All sources
- Ahrefs: 12 SEO KPIs You Should (And Shouldn't) Track
- Semrush: SEO KPIs
- Search Engine Land: SEO KPIs to track and measure SEO success
- Search Engine Journal: Top SEO KPIs
- Seer Interactive: 3 SEO KPIs You Should Be Tracking
- Shopify: SEO KPIs
- HubSpot: Blogging metrics and KPIs
- SEOFOMO: State of Technical SEO, 2024
- Vodafone, Pinterest, Etsy
- ZipRecruiter, Eventbrite, Rakuten
- HubSpot blog analysis (Aleyda Solis), HubSpot Q2 2025 earnings call
- Elite Tune and steel framing, Search Engine Land Awards 2024, HouseFresh