Nothing spiked, and that is the finding

When you look at a growth chart, the first question is not how big the gain was. It is whether the gain arrived as a step or as a slope, because those point at different causes and they are worth different amounts. A step means one thing changed: a technical fix landed, a migration completed, a single page caught a wave. A slope means something is accumulating. What our proof page publishes for clawlaw.in is a slope rather than a step. Month by month, the daily click average reads 59, then 154, then 282, then 349, across a window of 2026-07-13 to 2026-10-12 exported on 2026-10-14. No single day carries the result. That is the finding, and it is a duller sentence than any spike would have produced.

How to tell a step from a slope

Three readings separate them, and all three are available from a monthly table without any extra data.

Reading	A step	A slope
Gain between periods	Concentrated in one boundary, flat either side	Present at every boundary
What happens if you remove the best month	The result mostly disappears	The rest still rises
What the cause has to be	One identifiable event, nameable with a date	Something repeated across the whole window

Run the middle test on the four figures above and the shape holds without any one of them. Remove September and July still beats nothing, August still beats July and October still beats August. A step cannot survive that test, which is what makes the test worth running first.

Why a step is easier to produce than a slope

Because a step only requires one thing to go right once. A single technical correction on a large site, a pile of pages suddenly indexed, one piece that gets picked up widely: each produces a large, genuine, datable jump in a Google report. None of them is evidence that the next month will be better than this one. A slope requires the thing that caused last month's gain to still be happening, which is a much stronger statement about a process and a much harder one to arrange.

That asymmetry is the reason absence of a spike is informative rather than disappointing. The chart with no dramatic day in it is making a claim about repeatability. The chart with one enormous day in it is making a claim about that day.

The two questions to ask about any rising chart

• Which period carries the gain. If the answer is one period, you are being shown an event. Ask what the event was and whether it can happen again.
• What the starting conditions were. A multiple depends almost entirely on where it started. The domain behind these figures had pages and history before the window opened, so this is growth on a working site rather than a launch, and the same multiple from a standing start would describe something completely different.

Both questions are answerable from a monthly table with a day count, and neither is answerable from a line chart with no axis labels. When the chart you are shown belongs to somebody else's case study, the six items a credible one has to contain are set out in how to read an SEO case study and how to write one that earns trust, and this reading is the one to take first because it decides what the rest of the document is about.

Applying it to your own numbers

Take your own last four months of non brand clicks a day, write them in a column, and ask which boundary carries the gain. Three outcomes are worth acting on differently. A slope means keep going and measure at the same cadence. A step with an identifiable cause means the cause is now your lever, so find out whether it is repeatable before budgeting as if it were. A step with no identifiable cause means the number moved for a reason outside your work, and the next thing to establish is what it was, because an unexplained rise reverses as easily as it arrived.

Why we published the dull version

We sell this work, so a chart from us is self interested by construction, and the only way to make one worth reading is to publish the shape rather than a highlight from inside it. The appetite for the duller version is measurable rather than assumed: put the question to ChatGPT and Gemini on 15 October 2026 and 68 of 102 conversations named failures and negative results among the things a page would have to show. A monthly table with no standout day in it is the least dramatic object that answers that.

Questions about telling real growth from a one off jump

Is a spike always suspicious?

Not at all. Spikes are real and often valuable, and the only error is reading one as evidence about the next quarter. Report it as an event with a date and a cause, and keep it out of any sentence about the ongoing level.

How many periods does the shape take to show up?

Four will do, because that is enough boundaries to see whether the gain sits at one of them or at every one. With two periods a step and a slope look identical, which is what makes a bare before and after the weakest form of this claim.

My growth is a slope but it is slowing. Is that bad?

A decelerating slope is still a slope. The figures above rise by a smaller multiple each month while the absolute gain stays large, which is the normal arithmetic of a base that is getting bigger.

Can a supplier fake a slope?

It is far harder than faking a step, because every boundary has to rise and the day counts are printed. The practical check is to ask for the monthly table rather than the chart, and to ask which month was the weakest.

What if there is no gain at any boundary?

Then the question moves from reporting to diagnosis, and the first thing to establish is whether the pages are being served at all. Impressions and distinct query counts answer that before clicks do.

The shape of a result decides what it is evidence for

Two charts can carry the same total and mean entirely different things. One of them says a thing happened. The other says a process is running. Suppliers prefer the first because it is easier to produce and photographs better, and buyers should prefer the second because it is the only one that predicts anything. The whole reason the monthly table is published rather than described is that the shape is the claim, and a reader who disagrees with our reading of it can take the four figures and the day counts and reach their own.