What belongs in a week two report for a site nobody has linked to yet

Report two fields in the first fortnight of a brand new site: impressions, and how many of your pages Google has kept in its index. Those are the two that can move and the two that mean something. Clicks and average position are the ones people ask for, and on a domain two weeks old neither is answerable yet. The aiknowsus.com case on our proof page is the worked example, covering 2026-10-04 to 2026-10-12, 9 days with data, exported on 2026-10-14. Across those nine days the impressions by day read 5, 4, 6, 7, 19, 84, 1,100, 2,685 and 2,055, and the clicks across the whole window total 3.

Field one: impressions, read as a count of appearances

An impression means one of your pages was in the results for something somebody typed. It is the first field that can move on a new domain, because being shown needs the page indexed and judged relevant enough to appear, and nothing more.

Report it as a total for the window with the day count attached, and report the daily figures rather than only the total. The nine numbers above make the point better than any summary: the first five days sit in single digits, and the last three are in four figures. A total of 5,965 alone would hide the fact that almost all of it arrived at the end.

Field two: what Google kept

The second field is the one most week two reports omit, and it is the one that explains the first. Open the Page indexing report in Search Console and read the indexed count. Compare it with the number of pages you published. The gap, and the reason named beside each excluded group, is the most actionable thing available in the first fortnight.

This matters because impressions are downstream of it. A page that is not in the index records nothing however good it is, so a flat impressions line on a new site is usually an indexing question rather than a quality one. Checking which of your pages Google has indexed takes one screen and answers it directly.

The two fields to leave out, and what they would say if you did not

Clicks first. In the nine day window above the click column reads 3. Published in a week two report as a headline, 3 reads as failure, and it is not a finding about the work at all: clicks follow position, and position takes longer than nine days. The column belongs in the appendix with the day count beside it, not in the summary.

Average position second. The best day of that window, 11 October 2026, recorded 2,685 impressions at an average position of 41.6. That figure is real and it is averaged across every impression in the window, in every place and on every device, so on a tiny base it moves for reasons that have nothing to do with anything you did. It is a trend field over months and it is noise over days.

What the report should actually say

Four lines, and they can be written before the data arrives because the shape does not change.

• The window, as two dates, and the days with data.
• Impressions for the window, with the daily figures attached.
• Indexed pages against published pages, with the reason for any gap.
• One sentence naming what cannot be answered yet, and when it will be.

That fourth line is the one that keeps a stakeholder calm in week six. A report that answers two questions and names the two it is not answering yet is a report nobody has to interpret. A report that quietly leaves out clicks invites the question at the worst possible moment.

Why the wrong pair costs a month

A new site judged on clicks in its first fortnight produces a predictable reaction: the pages are declared not to be working and the plan is changed. The plan was fine. The measurement was early.

That reaction has a real cost, because what the first fortnight is for is finding the indexing problems while they are still cheap to fix. Time spent rewriting pages that were never indexed is time spent on the wrong layer, and the indexed count would have said so on day three.

For the fuller version of which figures survive being reported and which mislead when they are, the scorecard of metrics worth reporting sets out the separation, and the ninety day view of the same sequence is in what to expect in the first 90 days.

The field with no console behind it

One more field belongs in a week two report on a new site and it does not come from Search Console at all: whether an assistant names you when somebody asks a buying question in your category without using your name. There is no report to open, so the only way the field exists is if somebody asks the questions and writes down what came back, with the date and the apps named. That is the measurement we sell, and on a two week old site it is the one field where you can establish a baseline of zero and have it mean something.

Questions about reporting on a site in its first fortnight

How soon should the first report go out?

At day nine or ten, because that is when a performance export has enough rows to show a shape. The export does not carry the most recent day or two, so a report written on day three is reporting on day one.

What if impressions are still in single figures?

Then the indexed count is the field to act on rather than the impressions. Single digit impressions with most pages unindexed is an indexing problem with a named cause in the report. Single digit impressions with everything indexed is a relevance problem, which is a different week's work.

Is three clicks in nine days a bad result?

It is the expected result, and it is published rather than hidden for that reason. Clicks are a function of position, and a nine day old domain has not had time to earn one. The figure becomes meaningful when the window is long enough for position to have moved.

Should the report include a forecast?

Include the next reading date instead. All 102 build spec conversations run on ChatGPT and Gemini on 15 October 2026 asked for a date, a last verified line and a refresh frequency, and a stated next reading is the honest version of what a forecast is reaching for.

Does any of this change for a site with existing pages?

Yes, and that is the point of asking first. A domain with an index footprint and links can be read on clicks from the start, because it has a base to compare against. The two field report is for the case where there is no base yet.