Indexed is not the same as found.

Both of our sites are crawled daily and fully indexed. Over ninety days they recorded one search query between them. Those two facts are not in tension, and understanding why they sit together comfortably is most of what a new site needs to know about its own data.

The numbers, without decoration

Forty-seven published pages across two domains. Sampling index status directly, seven of nine checked URLs came back submitted and indexed, each crawled within the previous two days. Robots allows everything, canonicals are self-referential, sitemaps fetch without error. There is nothing technically wrong.

In the same window, one site recorded thirty-three impressions and the other recorded zero. Between them, exactly one search query appeared in the data, at position 49. The pages that did surface were the homepage and three section pages — the ones someone would reach by typing the brand name, not by asking a question.

Why this is the expected shape

Indexing answers a storage question: does the engine have a copy of this page and is it eligible to appear. Ranking answers a competitive one: for this particular query, out of everything eligible, is this the best result. A new domain passes the first and loses the second to pages that have been accumulating links, engagement, and revision history for years.

So the correct expectation for a young site is precisely what we are seeing: fully indexed, effectively invisible. There is no intermediate state where a page ranks slightly. It either clears the bar for a query or it does not appear at all, which is why impression counts on new sites jump from nothing to something rather than climbing smoothly.

The number that actually carries information

Not impressions. Not position. The count of distinct queries recorded. One query in ninety days is not a statement about our ranking; it is a statement that our pages are not yet considered a candidate answer for anything. That is a content-market fit signal, and it is available long before traffic is.

When that count starts climbing — five queries, then twenty, all sitting at position 40 or worse — something real has changed even though traffic is still zero. The engine has begun matching the pages to intents. Position improves from there far more predictably than eligibility appears in the first place.

Why we stopped reading daily numbers

At this volume a single day is seven sessions on one site and four on the other, and a day where nobody happened to scroll produces a hundred percent bounce rate and a zero second average — figures that look alarming and mean nothing. We spent part of an afternoon investigating exactly that, and the twenty-eight day view showed normal engagement the whole time.

The threshold worth applying is simple: if a metric can be moved by one visitor, it is not yet a metric. It is an anecdote with a percentage sign attached.

What to do while the data says nothing

Fix what is objectively wrong, then stop refreshing. Our audit found real defects worth correcting — titles truncated in results, a redirect listed in the sitemap contradicting its own canonical, section pages shipping without an H1, topic pages too thin to be anyone's answer. None of those were the reason for zero traffic, and all of them are worth fixing anyway, because they are the conditions under which a page becomes eligible when authority does arrive.

The rest is publishing and time. That is an unsatisfying answer for anyone who would rather have a lever, which is exactly why so much SEO advice sells one. The honest version is that a new site's search data has very little to say for its first few months, and the discipline is to keep shipping while it says nothing.

← All notes