Why did my App Store keyword go to Not ranked overnight?

A rank only counts inside the top 100, so a keyword that falls past position 100 reads as not ranked rather than as a lower number. A whole set of keywords flipping to not ranked on the same day usually means the store re-indexed, not that your listing broke.

How do I tell a store-wide update from a problem with my own app?

Use a control. Compare the same app on a store you did not touch over the same days. If both stores fall, the cause is your app. If one falls while the other stays flat, the cause is that store. The algorithm change detector answers the same question per store and per country.

Should I rewrite my metadata the day my rankings drop?

No. A release re-triggers indexing, so you lose the ability to tell the store's behaviour from your own change. Wait for two or three daily readings, confirm with a control, then decide. The rankings are already gone and a two day delay costs you nothing.

Do rankings come back after a store-wide update?

Not always. A store-wide recalibration can push keywords out of the top 100 in a single day and hold the new level for good. Treat it as a new baseline to work from, not a dip to wait out.

How to diagnose a ranking drop, step by step

Your positions fell. Before you rewrite anything, find out whether it was you or the store, using a control, the change detector and your release history.

Your positions fell and you want to fix it. This guide finds out what actually happened before you touch the listing, in about fifteen minutes. It needs a tracked app with at least a few days of ranking history behind the drop.

Diagnosing a ranking drop, in order. The order is the whole point:

Measure what dropped, across the set rather than one keyword.
Rule out ordinary noise.
Ask whether the store moved, in this market.
Use a control: another store, or your competitors.
Only then look at what you shipped, your rating and who overtook you.
Decide on day three, not day zero.

1. Measure what dropped

Start on the dashboard home. Biggest drops lists the keywords where your position fell the most this week, with a count. Then open the Keywords Tracker and sort by Trend (7d).

Read three things off the table:

How many keywords moved. One, or most of them.
How far. The Position column, and the Out pill on any keyword that was ranked and no longer is.
When. The seven day trend sparkline, or the rank history on the keyword itself.

A rank counts only inside the top 100. Below that the Position column shows a dash, which is why a drop past position 100 looks like a cliff rather than a slope.

What good looks like at this step: a sentence of the form "eleven of my fourteen keywords lost more than twenty places between Tuesday and Wednesday". If you cannot write that sentence, you do not have a drop yet, you have a feeling.

2. Rule out noise

Individual ranks move on their own, every day, with nothing happening. On apps that shipped no listing change at all, most keywords do not move from one day to the next, and a small minority move five places or more, a little more often on Google Play than on the App Store. The figures are in how long ASO changes take to take effect.

Track twenty keywords and, on a completely uneventful day, one or two will move five places and about one will move ten. That is not a signal, that is what ranks do.

Stop here if: one keyword moved and the rest are flat. Come back in three days.

3. Ask whether the store moved

Open the algorithm change detector and pick your store and your country. It measures daily search volatility per market and marks each day either Calm or Update detected, with a ratio saying how far the day ran above that market's own normal: 1 is an ordinary day, 3 is three times the usual movement.

Two shapes of update are worth recognising. Both move thousands of indie apps at once, and neither is ever announced:

A recalibration. Keywords fall straight out of the top 100 in a single day on one store while the other does not move, and the new level holds. One on the App Store is analysed day by day in why your rankings dropped overnight.
A worldwide update. Every major market of one store moves on the same day, several times beyond its ordinary level, while the other store reads a normal day. One on Google Play is analysed in the Google Play algorithm update.

If the day is marked Update detected in your market: you have your answer. Nothing you do to your listing brings those positions back, because they did not leave for a reason that lives in your listing. Treat the new level as your baseline and go back to work on the terms you can still win.

4. Use a control: the other store, or your competitors

If the detector shows nothing, build the control yourself. This is the fastest diagnostic in ASO and it costs nothing.

If you ship on both stores, look at the same app, over the same days, on the store you did not touch.

Both stores fall. The cause is you. Go to step 5.
One falls, the other is flat. The cause is that store.

If you are on one store only, use your competitors instead. Open a tracked competitor in Competitors Analysis and read the Head-to-head tab over the same days. If everyone in your niche fell together, it was not your subtitle. This is one of the reasons to track competitors before you need them, covered in turning a competitor gap into a plan.

A number that moves is a story. A number that moves while its control stays flat is evidence.

5. When it really is you: four causes to check in order

Four causes, in the order they actually occur.

You shipped. Check the release history for your own app and line the dates up. A release can re-index your listing and shuffle rankings for several days even when the metadata did not change. But remember that a metadata change lands on the first daily reading: a drop that starts a week after your release probably is not your release.

You edited the field that was working. On iOS the keyword field and the subtitle carry most of your indexing, and removing a term costs you its position as fast as adding it won it: a keyword dropping out of an iOS title lost 18 positions on the first reading and settled about 22 worse, in the same measurement. There is no grace period.

Your rating fell. Conversion feeds ranking on both stores, so a slipping rating costs you positions on contested terms. The Store appeal axis of ASO Health tells you when your rating is holding installs back.

Someone else moved. Look at the Ahead column in the tracker: if a new app is sitting above you on the keywords that fell, you did not drop, you were passed. Their release history will usually say what they did.

6. One more check if the data itself stopped

If positions did not fall but simply stopped updating, the cause may be on your account rather than on the store. A free plan keeps running as long as you use it, and one button brings a paused account back. The free plan also keeps seven days of history, so a comparison older than that belongs to the paid plan. Both cases are covered in the inactivity pause and plans and limits.

7. Decide, on day three

Whatever you concluded, do not ship on the drop day. A release re-triggers indexing, so you add your variable on top of whatever moved and you can no longer read either. Wait for two or three daily readings first.

Then:

Store-wide event. Change nothing because of it. Re-check your difficulty scores before trusting an old shortlist: a recalibration rewrites difficulty across the whole store, in every tool on the market, so a "keywords I can win" list built before the event is stale.
Your release. Revert the field you changed, or accept it and measure for two weeks.
Your rating. Fix what the recent reviews complain about, then ask happy users at the moment the app has just delivered.
A competitor. Go to the head-to-head and pick the gaps you can actually close.

And one thing you can cross off straight away: what you added or removed in your tracker. The visibility score reads a stable set of keywords from your app's niche, the same set every day, so your own tracking choices never move it either way.