Is My SEO Agency Working? A Numbers-Based Checklist

You pay the invoice every month. The reports arrive full of activity — posts published, links built, hours logged. And yet you can't shake the question: is any of this actually doing something?

It's a fair question, and you can answer it without becoming an SEO expert. You don't need a gut feeling or a reassuring call. You need three numbers and about ten minutes. Here's the honest math on whether your SEO agency is working — and what to do if the numbers say it isn't.

The Short Answer: Three Numbers, Not a Feeling

Strip away the jargon and there are only three numbers that tell you if SEO is working:

1. Rankings for keywords that could plausibly bring you money — not vanity phrases nobody searches.
2. Referring domains — how many distinct websites now link to you.
3. Real organic traffic — actual clicks landing on your site from Google.

If all three are moving in the right direction over a reasonable window, it's working, and the monthly report is just color commentary. If none of them have budged after six-plus months, no amount of "reporting calls" or "we're building authority" changes the answer.

This is not a vibe check. It's arithmetic. The rest of this piece shows you exactly where to find each number and what "good" looks like.

How Long Is It Reasonable to Wait?

Before you judge anything, be fair about the clock. SEO is slow by nature, and the mainstream industry consensus — Search Engine Land, and echoed by the likes of WebFX, Shopify and Ahrefs — is that it takes roughly 3 to 6 months to show initial measurable results, and 6 to 12 months for meaningful traffic and lead gains. Competitive keywords or a brand-new site can take longer still.

So panicking at month one is premature. But that same consensus cuts the other way: if you're staring at flat numbers well past the six-month mark, that's a legitimate signal, not "normal SEO patience."

What "early" actually looks like

A healthy early engagement is quiet, and that's fine. Expect the first month or two to go into foundation and audit work — fixing technical issues, cleaning up existing pages, mapping keywords. Movement, when it comes, usually shows up first on long-tail terms around months three and four: lower-competition phrases that are easier to win. If your agency told you upfront that the opening stretch would be groundwork, don't hold the quiet against them.

When "be patient" stops being a valid answer

The grace period has an end date. By the six-month mark, you should be able to point to movement on at least one of the three core numbers — a money keyword climbing, new referring domains, or organic clicks ticking up. If all three are still flat after six months of steady billing, "just be patient" has stopped being an explanation and started being a stall. Patience is not the missing ingredient at that point.

The Metrics That Actually Prove Something Happened

This is where most agency reports fall apart — not because they lie, but because they measure activity instead of outcomes. Here's what to demand instead.

Rankings for keywords that could plausibly bring you money

"We improved your rankings" is close to meaningless on its own, because where you rank is everything. A Sistrix study analyzing more than 80 million keywords found the #1 organic result earns an average 28.5% click-through rate, dropping to 11% at position 3 and just 2.5% at position 10. First Page Sage's 2026 data tells the same story from another angle: the top three organic results capture 68.7% of all clicks on a page.

Translation: a keyword sitting at the bottom of page one, or on page two, drives close to no traffic no matter how proudly it's reported. So the question isn't "did rankings improve?" It's "which position band did my commercial keywords land in?" A move from position 18 to position 12 looks like progress on a chart and changes nothing in your inbox. A move into the top three is the only kind that pays.

Referring domains — the metric that correlates with rankings

When an agency says "we built 40 backlinks this month," ask a sharper question: from how many different websites? In Ahrefs' study of roughly 920 million pages, the number of referring domains — distinct linking sites, not raw link count — was the strongest backlink factor correlating with rankings.

Forty links from one low-quality directory is not the same as links earned from eight new, credible domains. The second is worth far more and far harder to fake. "Referring domains, trending up over six months" is the metric to put in your report; "total backlinks built" is the one that flatters an activity list.

Real organic traffic and clicks, not impressions

Impressions are the most seductive vanity metric in the deck. Your impressions can climb while your actual traffic sits perfectly still — you're just appearing in more searches nobody clicks. Insist on organic clicks and sessions from Google Analytics and Search Console, not impressions.

There is one honest caveat here. A February 2026 Ahrefs study of 300,000 keywords found that when Google's AI Overviews appear at the top of a result, click-through rate for the top-ranking organic page drops by an average of 58% (position 2 falls 50.8%, position 3 falls 46.4%). So it's genuinely possible for your rankings to hold while clicks slip, through no fault of your agency. That's a real explanation — but it's one your agency should be able to show you with data for your specific keywords, not a blanket excuse waved over every flat month. Whether blogging is even the right lever for your traffic goals is its own question; we've laid out the honest data on whether blogging moves the needle for SEO separately.

Why "Domain Rating went up" isn't proof of anything on its own

Domain Rating, Domain Authority, and their cousins are the metrics agencies reach for when the real numbers aren't cooperating. Here's the thing: Ahrefs, the company that publishes Domain Rating, is blunt that it's a correlation, not a ranking factor Google actually uses. A 2025 Ahrefs study found only a small positive 0.131 correlation between DR and rankings.

So "your Domain Rating went from 22 to 28" is directional at best and proves nothing by itself. It only means something when it's paired with actual keyword and traffic movement. On its own, it's a score going up on a scoreboard Google doesn't read.

What a Real Report Looks Like vs. an Activity List

Hold your last monthly report next to this test. An activity list tells you what the agency did: posts published, links built, keywords "targeted," hours logged. A real report connects every one of those activities back to the three numbers — this content went live, here's the keyword it's now ranking for and in what position, here's the traffic it's pulling.

Multiple SEO-industry sources flag the same pattern as a red flag: monthly reports that list activities without ever tying them to ranking, traffic, or lead outcomes — and reports that only celebrate keywords with negligible search volume, because those are the easy ones to "win." If your report never draws a line from work done to results earned, that's not a reporting style. It's a way of not being measured.

Red Flags Straight From Google's Own Playbook

You don't have to take an outsider's word for what a bad SEO looks like. Google publishes its own guidance, and it's remarkably direct. Google Search Central states plainly: "No one can guarantee a #1 ranking on Google. Beware of SEOs that claim to guarantee rankings, allege a special relationship with Google, or advertise a priority submit to Google."

A few concrete rules straight from that guidance and the wider record:

• Ranking guarantees are a fraud pattern, not a sales style. In one real FTC enforcement case, a court banned two defendants from ever selling search optimization services again, and the FTC returned more than $1.8 million to defrauded small businesses — an average refund of about $158 each. That's not hypothetical.
• Guard your access. Google advises that when an SEO runs an audit, you grant read access to Search Console first — not write access — until trust is established.
• Know your recourse. In a 2026 update, Google went a step further than ever before and explicitly encouraged businesses to file FTC complaints against deceptive SEO providers (1-877-FTC-HELP, or ftc.gov), as Search Engine Journal reported. The same update added warnings about third-party SEO tools that, in Google's words, don't have access to its internal ranking data.

One more access red flag that appears across the industry: an agency that won't give you login and ownership of your own Google Analytics, Search Console, and Google Business Profile — only polished PDFs. Those are your accounts. If you can't get into them, ask why, loudly.

Is Your Price, Contract, and Timeline Even Normal?

Judging value means knowing the going rate. An Ahrefs poll of 439 SEO professionals put the average agency retainer at about $3,209/month, versus $1,348/month for freelancers, with the single most common bracket at $501–$2,000/month. Most small-to-midsize businesses land in the $1,500–$5,000/month range for a legitimate ongoing retainer.

If content is bundled in, Clutch's agency research pegs comprehensive content-inclusive retainers at $3,000–$15,000/month for 4 to 12 published pieces — so if you're paying $4,000 a month and seeing one thin blog post, the invoice and the deliverables don't match. (A serious standalone backlink audit, for reference, runs $300–$800 and ends with a disavow file submitted through Search Console when link quality is genuinely in question.) It's worth understanding what a $3,200-a-month content agency retainer actually includes before you decide yours is over- or under-priced.

Contract length matters too. GoodFirms' 2026 research shows retainer clients stay engaged roughly 56 months on average versus 24 months for project-based work, with about 18% annual churn for retainer agencies versus 42% for project shops. Read that two ways: bailing at month three is often premature, but sitting in year four with flat numbers is overdue for a hard conversation.

And if it is working, take some comfort — dissatisfaction is a real but minority experience. An older Clutch survey of small business owners found 66% were very or somewhat satisfied with their SEO provider, 24% neutral, and only 10% dissatisfied. Worth taking seriously if you're in the 10%; worth remembering if you're just anxious.

The 10-Minute Agency Audit

Here's the part to screenshot. Set a timer and run these five checks today:

1. Pull your money keywords in Search Console. List the 5–10 terms that would actually generate business. Where do they rank — top three, or buried on page two? Only the top three meaningfully drive clicks.
2. Check referring-domain growth. In any backlink tool, look at distinct referring domains over the last six months. Trending up? Flat? Padded with junk directories?
3. Check the organic traffic trend. In Analytics or Search Console, look at organic clicks and sessions — not impressions — over six to twelve months.
4. Confirm you own your access. Log into your own Google Analytics, Search Console, and Business Profile right now. If you can't, that's the finding.
5. Read your contract for the exit. Find the termination clause. If there's a long lock-in and no clean way out, you know that before renewal day, not after.

If three or more of those come back healthy, your agency is very likely earning its retainer. If three or more come back flat or blocked, you have your answer in numbers, not nerves.

If the Numbers Say It Isn't Working

Start with a conversation, not an exit. Bring the three numbers to your agency and ask them to explain the movement — or the lack of it — against the timeline you agreed to. A good partner will walk you through it with data; if AI Overviews are eating your clicks, they'll show you exactly where. A weak one will change the subject back to activity.

If the answer still doesn't add up, get specific about the alternatives before you re-sign anything. Some of the work you're paying a premium for is genuinely worth doing yourself — here's which SEO tasks you can reasonably take back and do yourself. And if you're paying agency-level prices for freelancer-level output, it's worth seeing how agency, freelancer, AI-tool, and managed-worker pricing compares side by side.

The goal isn't to fire anyone. It's to stop guessing. Once you're judging the relationship by rankings, referring domains, and real traffic — instead of by how busy the report looks — you'll know exactly what you're paying for, and whether it's worth paying again.