What a Monthly SEO Report Should Actually Tell You

Every month a PDF lands in your inbox. It has a green upward arrow, a number like 41,000 impressions, and a line that says 87 keywords improved. You skim it, you feel vaguely reassured, and you pay the invoice.

That report was not built to tell you the truth. It was built to survive your attention span. Almost every article about monthly SEO reports is written for agencies — how to build one, which template to buy, which dashboard tool to license. This one is written for the person receiving it.

Here is how to read the document you already get, spot the five tricks hiding in it, and demand the three numbers that decide whether the retainer is working.

What should a monthly SEO report actually tell you?

A monthly SEO report should answer three questions in the first 30 seconds: how many qualified leads did organic search produce, what changed to cause that, and what happens next month. Everything else is supporting evidence.

The test is brutal and simple. Cover everything below the fold of page one. Can you still tell whether last month was good or bad? If not, the report is decoration.

Most reports fail that test because they lead with activity (blog posts published, links built, pages optimized) and bury outcomes on page nine. Activity is an input. You are not buying inputs. You are buying booked calls.

Which metrics in your report are pure vanity?

Sort every metric in your report into one of three tiers — vanity, diagnostic, money — and you will usually find that 80% of the page count sits in tier one. Vanity metrics move without anything real happening. Diagnostic metrics explain why the money metrics moved. Money metrics are the only ones you can take to your accountant.

The tiers are not a moral judgment. Diagnostic metrics matter — you need them to debug a bad month. The rule is about placement: money metrics go on page one, diagnostics go in the appendix, and vanity metrics go in the bin.

Impressions deserve a special warning. Google's own Search Console documentation defines an impression as 'how often someone saw a link to your site on Google' — and adds that an impression means the user 'has seen (or potentially seen)' that link. Potentially seen. Not read, not clicked. Your impressions can double while your phone stays silent.

Why is 'domain authority went up' not a result?

Because Moz — the company that invented Domain Authority — says plainly on its own Domain Authority page that 'Domain Authority is not a Google ranking factor and has no effect on the SERPs.' It is a 1-to-100 prediction score built from Moz's link index, and Moz explicitly calls it a comparative metric, not an absolute one.

Ahrefs says the same about its own Domain Rating. In Ahrefs' guide to SEO metrics, the company writes: 'We don't recommend using Domain Rating as a standalone metric. It's not a good indicator of a website's quality or legitimacy; it's merely a measure of link popularity.'

So when a slide says DA 18 to DA 24, both vendors whose products generate that number have told you it is not a ranking factor. A DA bump can also come from links you would never want — scraped directories, syndicated press releases, expired-domain spam. The score went up. Your business did not.

Fair use of DA/DR: sizing up competitors before you commit to a keyword. Unfair use: presenting it as a monthly deliverable.

The same logic kills 'keywords improved.' A keyword that moved from position 74 to 68 improved. Nobody clicked it. Nobody will. Ask which keywords moved into the top three and what those specific terms produced in clicks and calls — a list of six real terms beats a count of 87 imaginary ones.

How does a rolling 12-month graph hide a 3-month decline?

A rolling 12-month line chart can show a confident upward slope while the last 90 days are down 30%, because ten months of early growth dominate the visual and the recent dip becomes a wiggle at the right edge. It is the single most common trick in agency reporting, and it is not even dishonest on paper — the data is real.

The fix takes 20 seconds. Ask for the same chart on three fixed windows every month, side by side:

• Last 28 days vs the previous 28 days — catches this month's damage
• Last 3 months vs the same 3 months last year — strips out seasonality (roofing in March is not roofing in December)
• Last 16 months, unsmoothed — the longest window Search Console keeps, so nothing gets cropped

Three related tricks live in the same neighborhood. Rank-tracker screenshots instead of Google Search Console: a rank tracker checks a robot's SERP, not your customers' SERPs, and it has no clicks, no CTR, no leads. Charts with the y-axis truncated so a 2% rise looks like a hockey stick. And 'average position improved from 14.2 to 11.8' — Google defines it as 'the topmost position occupied by a link to your property or page in search results, averaged across all queries in which your property appeared,' so it can improve while your best pages get worse.

If your agency will not export the raw Search Console data alongside the pretty chart, that is your answer. Our Search Console guide for service businesses walks through pulling these comparisons yourself in about ten minutes.

Is your agency counting branded traffic as an SEO win?

This is the most expensive trick on the list, because branded search — people typing your company name — converts at a high rate and makes any report look excellent, while proving nothing about the SEO work. If you ran a radio ad, sponsored a little-league team, or had a good month of referrals, your branded clicks rise. Your agency did not cause that.

In Search Console, filter the Queries report to exclude your brand name and its common misspellings, then look at what is left. That residue is the SEO program. For a dental practice, it is 'invisalign near me,' not 'smith family dental.' For an HVAC company, it is 'furnace not blowing hot air,' not the company name plus 'reviews.'

Demand the split every month: branded clicks, non-branded clicks, and the non-branded share of total. If non-branded is flat for six months while total traffic climbs, you are paying a retainer for your own reputation.

One more layer in 2026: AI Overviews sit above the results and eat clicks that impressions still count. Ahrefs compared 300,000 keywords using Google Search Console data and found that the presence of an AI Overview correlates with a 58% lower clickthrough rate for the #1 organic result — position-one CTR on informational keywords carrying an AI Overview fell from 7.3% in December 2023 to 1.6% in December 2025 (Ahrefs, December 2025 data). A report that celebrates rising impressions while clicks flatten is showing you exactly that phenomenon and calling it progress.

And you cannot ask your agency to break out AI traffic separately, because Google does not provide it. Google's own documentation states that sites appearing in AI features 'are included in the overall search traffic in Search Console… reported on in the Performance report, within the Web search type.' Any agency selling you a line item for AI Overview clicks is inventing it.

Which three numbers should be on page one of every report?

Three money metrics, in this order: qualified leads from organic search, cost per qualified lead, and revenue attributed to organic. Everything else is a footnote. If your retainer is $2,500/mo and organic produced 12 qualified leads, your cost per lead is $208 — now you can compare that to Google Ads, to Angi, to your referral cost, and make an actual decision.

Define 'qualified' before month one, in writing, because it is where agencies hide. A qualified lead is not a form fill. It is a form fill or call from a real prospect, in your service area, for a service you sell, that your team could act on. A vendor pitching you SEO software is not a lead. A caller asking for a job is not a lead.

Getting to these three numbers requires three plumbing pieces: call tracking with a dynamic number on the site, form submissions tagged with the landing page and source, and a CRM field for the source on every closed deal. That is a week of setup, not a quarter. If your agency has been reporting for six months and still cannot produce a cost per qualified lead, the setup was never done. Our guide to measuring SEO ROI for a service business has the full build.

Attribution will never be perfect. Somebody sees you in an AI Overview, checks a review site, hears your name from a neighbor, then types your brand into Google. First-touch and last-touch will disagree. Pick one, apply it consistently, and treat the number as directional. Directional beats absent.

One caveat on timing: do not judge month one or month two on these three numbers. A fresh page needs to get indexed, gather links, and climb — that is a two-to-six-month arc for most service businesses, longer in a law-firm or medspa market where every competitor has been publishing for a decade. What months one and two must show is the plumbing working: leads tracked, sources tagged, pages shipped. If the numbers are simply absent by month three, that is not patience — that is a missing scoreboard, and you cannot audit what nobody is counting. Compare it against what a full program should cover in our breakdown of what SEO services actually include.

What should you ask for that your agency is not sending?

Six things, all of which take a competent agency under an hour to produce. If any of them triggers hesitation, you have learned something more valuable than the data.

• Direct read access to Google Search Console, GA4, and any rank tracker — as an admin on your own accounts, not a screenshot
• The non-branded query export (CSV, last 28 days vs prior 28 days) — the actual rows, not a summary
• Leads by landing page, with the query that drove them where it is known
• A written list of what shipped last month — page URLs, not 'content optimization'
• Next month's plan with a hypothesis — 'we expect X because Y,' so it can be wrong
• Losses — pages that dropped, links that disappeared, queries you fell out of

Everything in that list should already be yours. You paid for the content, the links, the accounts, and the codebase. An agency that keeps Search Console access under its own login is holding your data hostage, and that is a contract problem, not a reporting problem. We hand every client admin on everything from day one, which is also why our pricing is published rather than quoted.

What does a report look like when the month went badly?

It says the month went badly, in the first paragraph, with a number. That is the entire test — and roughly zero agency reports pass it, because a report that admits a bad month is a report that invites a cancellation conversation.

SEO has bad months. A core update lands. A competitor buys the market. A dev push breaks canonicals. Seasonality flips. None of that makes an agency incompetent. Hiding it does.

A good bad-month report reads: non-branded clicks fell 22% after the June core update; three of our money pages dropped from the top five to positions 11 to 14; qualified leads dropped from 14 to 9; here is what we think happened, here is what we are testing, and here is when we will know. That is a partner. A green arrow over 'impressions +38%' in the same month is a vendor managing you.

Which is why we back reporting with a 90-day kill switch and no lock-in contract: if a channel cannot show qualified leads in 90 days, it gets cut. A 12-month contract exists so the agency can survive a report like the one above. You do not need to fund that. Read what our SEO service actually includes before you sign anything with a minimum term.

If you want a second opinion on the report you got this month, send it over. We will read it against the three tiers above and tell you plainly which numbers are real, which are theater, and what a fair report for your spend should have contained. Get my free audit.