Still waiting for anything to appear? The first check finishes within the hour — see How long your first check takes.
Your first set of results is a snapshot, not a trend. Reading it as a trend is the most common early mistake, so it's worth knowing what the data can and can't tell you yet.
Why the first numbers look thin
On day one, each prompt has been checked once on each platform. That's a handful of runs — enough to see whether you appear at all, not enough to know whether appearing was typical.
Answers vary between runs on their own. If Nike shows up in the marathon-shoes answer today and not tomorrow, neither result is wrong; the average across many runs is the real signal. A visibility score built from one run per prompt will move a lot in the first week and then settle.
What's worth reading on day one
Do you appear at all? The blunt version of the question. If Nike isn't mentioned in any answer about running shoes, that's a finding you can act on immediately, and no amount of extra data will soften it.
Which sources got cited. The most useful day-one data by far, and reliable straight away — the pages an assistant reaches for don't fluctuate the way phrasing does. If a Runner's World roundup and a Reddit thread are shaping the answer, that's true today and will still be true next week.
Who else showed up. Competitor names in those first answers tell you the real shortlist you're up against, which is often not the one in your internal deck.
What to wait on
Your visibility score as a number to report. Give it a week or two of runs before treating it as a baseline, and longer before reading movement as improvement. How your visibility score is calculated explains what moves it.
Comparisons between prompts. A prompt where you appear and one where you don't may just be run-to-run variance this early.
Average position. Built from the subset of answers that rank brands in a list, so it needs the most data before it stabilises. See Prominence, share of voice, mention rate and average position.
If a prompt shows nothing at all
Some questions return no brands and no citations — not because tracking failed, but because that phrasing didn't prompt the assistant to search the web or name products. Why a prompt returned no results or no citations covers how to tell the difference and how to rephrase.
Two things worth doing now
Check your aliases. If your brand appears less than you expected, this is the usual reason — specifically for names that don't contain your brand name, like Jordan or "the Vaporfly". Anything containing your brand name is already matched automatically.
The good news is that aliases apply to past data too, so fixing them corrects your baseline rather than only helping from here. Why your brand wasn't detected in an answer has the detail.
Look at your prompts again. Now that you've seen real answers, it's much easier to tell which prompts produce useful data and which return generic prose. Adjusting them in week one is normal, and cheaper than living with a list you chose blind.
