How to know if SEO is working before rankings move

Clients ask me this in week three, and it is a fair question. How to know if SEO is working is not a rankings question in the first month, it is a data question. Rankings are a lagging indicator. They move only after several other things have already happened underneath them, and the gap between those things happening and rankings responding is where most clients lose confidence and pull the plug too early.

The mistake is watching the wrong metric. If you check your position for a target keyword every few days and it has not moved, you conclude nothing is happening. In most cases something is happening. It simply has not surfaced yet in the one number you are watching.

The sequence Google runs before a ranking changes

Before a page can rank differently, Google has to crawl it again, process the content, decide it deserves reindexing, and then reassess where it sits relative to every other page competing for that query. Each step carries its own delay. A site with a thin crawl budget or low authority might wait days between crawls of a given page. A newly rewritten page needs a fresh crawl before Google even knows it changed.

This is why a page can be substantially improved and still show the same ranking for two or three weeks afterward. The improvement is real. Google has simply not gotten around to reprocessing it yet. Confusing no rank change with no progress is the most common reason clients lose patience with a campaign that is still on schedule.

How to know if SEO is working when rankings have not moved

Google Search Console gives you visibility into the steps that happen before ranking movement, and these are the numbers I check first with a new client.

Impressions are the earliest honest signal. If a page starts appearing for more query variations than it did a month ago, Google has expanded its understanding of what the page covers and is testing it against more searches. That happens well before position improves.

Average position on page two or three matters more than it looks. A keyword sitting at position 18 that moves to position 12 produces no visible traffic change and feels like nothing happened. It is actually the page climbing through the range where most of the ranking work gets done, right before it crosses onto page one where clicks start.

Indexation status is the foundation check. A page that is not indexed cannot rank for anything, no matter how well it is written or optimized. I compare the indexed page count against the number of pages actually submitted in the sitemap early in every engagement, because a gap there means work is happening on pages Google has not looked at yet.

Crawl stats, buried in Search Console settings, show whether Google is visiting your site more or less often than before. An increase in crawl frequency after a technical fix or a content push is a direct sign the fix registered with Google's systems, independent of whatever rankings are doing that week.

What a month one baseline actually looks like

None of these signals mean anything without a starting point to measure them against, which is why the first thing I do with a new client is pull a snapshot of every number above before any work begins. Impressions, average position by query cluster, indexed page count against sitemap count, and crawl frequency all get recorded in week one, not week four, because by week four the early movement has already started and you lose the ability to prove it happened.

I walk clients through this same baseline so they can check it themselves without waiting on a monthly report. It takes about twenty minutes the first time and five minutes after that. The goal is not to turn a business owner into an SEO analyst. It is to replace a vague sense of nothing is happening with an actual number they can watch move on their own schedule, which is also exactly what I cover in more detail in what to expect in month one of SEO, including what a normal baseline looks like versus one with real problems already in it.

A site with structural issues shows a different baseline than a clean one. If the sitemap lists 400 pages and Search Console shows 220 indexed, that gap is the first thing to fix, and it will show up as improved indexation numbers within a few weeks, long before any ranking moves. A site with no indexation gap skips straight to the content and internal linking work, and its early signal is impression growth instead.

A worked example of what the data actually shows

Here is a pattern I see often enough that it is worth walking through in full, using rounded figures rather than a specific client's real numbers. A local service business starts a campaign with 40 pages indexed out of 55 in the sitemap, impressions around 1,200 a month for its core terms, and its best keyword sitting at position 22.

By week three, after fixing a handful of canonical tag errors and resubmitting the sitemap, indexed pages climb to 52. Rankings have not moved at all, and the owner is starting to ask questions. By week six, impressions are up to roughly 1,900 a month, mostly from query variations the site was not appearing for before, things like added location modifiers and related service terms nobody had written content around yet. The target keyword is still sitting at position 22.

By week nine, that keyword has moved to position 14. It is still off page one, still generating no extra clicks, and it is also right on schedule. By week twelve, it crosses to position 8, and click volume on that page roughly doubles. Nothing about week twelve was a surprise to anyone who was watching indexation and impressions from week one. It was the last step in a sequence that had been visibly working for two months already.

Why the order is not optional

These signals move in a fixed sequence because they are dependent stages, not parallel metrics. A page has to be crawled before it can be indexed. It has to be indexed before it can be tested against queries and generate impressions. It has to accumulate impressions and clicks before Google has the behavioral data to justify moving it up in position. Rankings are the last domino, not the first.

Click-through rate at a given position feeds back into that same sequence. A page sitting at position 14 that earns an unusually high click rate for that position, because the title and meta description answer the query better than competitors do, gives Google an early signal that the page deserves to sit higher than its content alone would justify. This is part of why two pages with similar on-page quality can climb at different speeds. The one with a sharper title and description is feeding the system better data at every position it passes through, not just at the top.

In my experience, the clients who stay calm through this stretch are the ones who get a Search Console walkthrough in month one, before anything is expected to move. They know what progress looks like on the way there, not just at the finish line. The ones who only ever check their target keyword's position are the ones who call in week four convinced nothing is happening, right as the underlying data shows the opposite.

Knowing the signals to watch is half the picture. If you're also wondering what a fair SEO price actually looks like, or whether AI overviews have changed what these results mean, both are worth reading alongside this one. More in the SEO Expectations archive.

What to do if a signal is missing entirely

Occasionally none of these signals move, not slowly, not at all, and that is a different situation from the normal lag described above. If indexed page count has not budged after a month and the sitemap gap remains where it started, something in the technical fix did not actually take, or Google has not recrawled the affected pages at all yet, which is worth checking directly through the URL inspection tool rather than assuming the fix will surface on its own. If impressions are flat after two months with no new query variations showing up anywhere, the content itself may not be covering enough distinct angles of the topic to get tested against a wider set of searches in the first place.

The difference between normal lag and an actual stall is whether the earlier-stage signals are moving at all. Lag looks like indexation and impressions climbing while rankings sit still. A stall looks like every signal flat at once, including the ones that should move first. Telling the two apart early is the difference between waiting out a normal delay and catching a fix that did not land before another month goes by on the wrong assumption.

A realistic timeline for each signal

Crawl frequency changes can show up within days of a fix, especially on a site Google already visits often. Indexation of new or changed pages usually takes one to three weeks. Impression growth from expanded query coverage tends to appear in the four to eight week range. Position improvement, especially breaking from page two to page one, typically takes two to four months, depending on how competitive the query is and how much authority the domain already carries.

None of this is a guess. It reflects how the system is built to work, and checking the earlier signals is the difference between knowing your SEO is on schedule and simply hoping it is. For the fuller month-by-month picture beyond just these four signals, how long SEO actually takes breaks down what typically happens in each stretch of a campaign, from the technical cleanup phase through the point where compounding growth actually starts.