SEO Agency for Gillingham Businesses

Knowing Whether Anything Is Actually Working

SEO reporting has a poor reputation, and largely a deserved one. Monthly documents arrive full of charts, and the recipient is no clearer about whether the money is doing anything.

The problem is rarely a shortage of data. Search Console and Analytics provide plenty. The problem is that the numbers most often reported are the ones easiest to make look positive, and that nobody has agreed in advance what would count as progress.

Good measurement is less about tools and more about asking questions that could produce an uncomfortable answer.

Search Console: What It Actually Tells You

Search Console is the most useful free source available, because it reports what happened in search itself rather than what happened on your website.

Four figures, and each is commonly misread:

• Impressions — how often your pages appeared. Rising impressions mean visibility is growing, but they say nothing about relevance. Impressions for irrelevant queries are noise.
• Clicks — how often someone chose you. The ratio between clicks and impressions indicates whether your title and description are earning the click.
• Average position — the least reliable of the four, for reasons below.
• Queries — the most useful. What people actually searched to reach you. This is where you discover you are visible for the wrong things.

The coverage report matters too, though it is less often opened. It shows which pages are indexed and which are excluded, with reasons. Pages that are not indexed cannot appear anywhere, so a problem here invalidates everything else.

GA4 and What Happens After the Click

Search Console stops at the click; Analytics picks up from there — which pages people landed on, what they did next, whether they completed anything.

The important discipline is to segment. A sitewide total mixes organic, paid, direct, referral and social traffic, and movements in one can be entirely masked by another. Looking at organic traffic specifically, and at the pages that matter commercially, is where the useful information lives.

GA4 is not intuitive, and a good deal of what it reports goes unexamined because nobody has configured it to answer a specific question. It is worth deciding what you want to know before opening it.

Landing Pages Rather Than Sitewide Totals

Total organic traffic is close to useless as a management figure.

A site can gain visitors overall while every commercially important page loses ground, because an old article picked up an unrelated audience. It can lose visitors overall while enquiries increase, because irrelevant traffic fell away.

Reporting at landing-page level resolves this. Which pages are attracting people, which are producing enquiries, and which are attracting attention that goes nowhere. That is a different — and far more actionable — picture than a single line on a chart.

Branded and Non-Branded Visibility

Someone searching your business name has already decided to find you. Something else created that demand: a recommendation, a previous project, an advertisement, a van they saw.

Non-branded searches are people looking for what you do without knowing who you are. That is the demand SEO is actually working to capture.

Combined, the two flatter each other. Branded terms are easy to rank for and convert well, so a report merging them can show healthy performance while new-customer visibility is flat. Filtering brand terms out of the query report takes a minute and changes what you are looking at considerably.

It also matters in the other direction. Rising branded search is genuinely good news — it usually means awareness is growing — but it should be credited to whatever created that awareness rather than to search work.

Reading Position Data Carefully

Average position is the figure most often quoted and least often understood.

It is an average across every impression, and results vary by device, location and search history. A page can be first for someone nearby on a phone and twelfth for someone forty miles away on a laptop. The reported average is a blend of both, so it moves for reasons that have nothing to do with your website.

It also improves when you lose visibility for terms you ranked poorly for, which is not obviously an improvement. And a screenshot of a ranking, taken on one device in one place, tells you almost nothing.

Position data is more useful as a direction of travel across a set of related terms than as a number to report.

Choosing Date Ranges Honestly

Comparison periods are where reporting most often becomes misleading, sometimes unintentionally.

Month against previous month is affected by the number of working days and by seasonality. A trade busy in spring will always look excellent in April and dreadful in November, regardless of any work done.

Year on year removes most of that, because you compare like with like. A rolling three-month view smooths the noise without hiding trends. What is worth avoiding is choosing whichever comparison happens to look best — if the range changes each month, that is worth asking about.

Conversions and What Can Be Tracked

Traffic figures matter far less than what people do afterwards, so it is worth tracking the actions that indicate commercial interest — form submissions, telephone clicks, quote requests, whatever represents a genuine step toward buying.

Some of this is straightforward. Some is not. A visitor who reads a page, notes the number and phones from a different device leaves no trace connecting the two. Offline outcomes are invisible unless somebody records them.

Tracking what can be tracked is worth doing properly. Believing it represents the complete picture is where measurement starts misleading rather than informing.

Traffic Quality

More visitors is not automatically progress. The question is whether they are the right visitors.

Indicators worth watching: whether people arriving on commercial pages go on to enquire, whether they arrive with expectations the page meets, and whether the queries bringing them relate to what you sell. A page attracting substantial traffic and producing nothing is usually ranking for something adjacent to your business rather than central to it.

Reporting Decisions, Not Numbers

The purpose of a report is to inform a decision. If it does not change what anyone does next, it is documentation rather than analysis.

A useful report is short. It says what changed, why it probably changed, what is being done about it, and what will happen next. Numbers support that narrative rather than replacing it.

Thirty metrics with no interpretation puts the analytical work back on the reader, who usually has less context than the person who produced it. If a figure is included, someone should be able to say what it means and what would be done if it moved.

Working With a Kent-Based SEO Agency

Maidstone Digital is based in Maidstone and works with businesses in Gillingham, elsewhere in Kent and further afield. We do not have premises in Gillingham. For search work this makes little practical difference — it is carried out remotely and reviewed through shared data and regular conversation — though we are happy to meet where that is useful.

This page covers SEO and search marketing specifically. If you are considering something broader — web design, Webflow development, design or paid media — our digital agency services in Gillingham page sets out the full range. For more detail on how search work is structured and delivered, see our Search Marketing services page.

SEO Gillingham FAQs

What should an SEO report actually contain?

What changed, why, what is being done next, and the small number of figures that support that. Visibility for terms relating to what you sell, performance of your commercially important pages, enquiries where they can be attributed, and anything moving in the wrong direction. If a metric is included that nobody can act on, it is padding.

Why does our traffic go up while enquiries stay flat?

Usually because the additional traffic is not the right traffic. An article or an unrelated page can attract considerable interest from people with no intention of buying. Look at which landing pages gained and which queries brought them — if the growth is on informational pages rather than commercial ones, that explains it.

How often should we look at the data?

For decisions, monthly is generally enough, with a year-on-year view quarterly. Weekly checking tends to produce reactions to noise. The exception is after a significant change — a migration, a redesign, a major update — when watching indexing and impressions closely for a few weeks is sensible.

Is average position a useful metric?

Only with care. It averages across devices, locations and search histories, so it moves for reasons unconnected to your site, and it can improve simply because you stopped appearing for terms you ranked badly for. Direction of travel across a set of relevant terms is more informative than the number itself, and a single ranking screenshot proves very little.

Should we exclude branded searches from reporting?

Not exclude — separate. Both are worth knowing. Branded search shows whether awareness is growing; non-branded shows whether you are being found by people who did not already know you. Reporting them together makes it impossible to tell which is happening, and it usually makes performance look better than it is.