Review miningThe 2026 keyword and ASO strategy guide
What is in this resource:
How to build your ASO strategy from end to end
Learn to pick target keywords, write metadata, follow rank and base every decision on data.
From data to action
Simplify the keyword data and stay in line with App Store rules.
Tools that lift rank
Carry the keywords drawn from competitor reviews into your metadata and take your app out of the cusp band into the top 10.
Why a yearly plan, and why quarterly decisions
How to build a year-long ASO plan: what happens each quarter, which decisions wait for which data, and how to admit what cannot be planned.
A year-long ASO plan is possible, but writing its moves out in advance is not. The reason is simple: what you do in month three depends on what you measured in month two.
So the yearly plan is built as a decision calendar rather than a list of moves. You write down which question each quarter will have answered by its end - not what the answer will be.
That distinction prevents the plan's largest failure mode: staying loyal to a list of moves written in advance. When the measurement shows a move did not work, the plan should not compel you to repeat it.
The second benefit is setting expectations correctly. "By the end of the year we will be in the top 10 for these phrases" is a forecast, not a commitment, and forecasts miss. "By the end of the year we will know which kind of move works for us" is an outcome you can commit to.
Quarter one: set up the measurement
The only aim of the first three months is regular measurement and a first loop of moves. No large targets are set this quarter.
- Month 1 - pool and listWiden the pool from five channels: your own metadata, rival metadata, search suggestions, category charts, rival reviews. Measure it, build a six-row working list, and add five to eight rivals.
- Month 2 - the first movesOne move a week, one field per release. Record the control group's average before and after every move. Four moves this month, all four recorded.
- Month 3 - the first inferenceThe first patterns emerge from eight to ten recorded moves: which field works, what the market's tempo is, which phrases have walls in front of them.
- End-of-quarter decisionDo we add a second country or go deeper in the first? The answer depends on whether the band in the first country is still full.
Quarter two: widen the scope
By the second quarter the loop has settled and scope can widen. Widening has two directions and they should not be taken at once.
The first direction is countries. Adding a second storefront means re-measuring all of your phrases there. Adding a country costs more than adding a phrase, which is why they go in one at a time.
The second is phrases. New candidates move from the pool onto the working list. That raises measurement cost, but not as much as a country does.
Which comes first depends on the first quarter's output. If the band in the first country has emptied - nothing left waiting to be spent - it is time for the second country. If the band is still full, adding phrases is unnecessary; finishing the current list is more productive.
This quarter also runs the metadata audit across every country: keyword fields, repeats, spaces, stem duplicates and empty countries. That audit usually finds more characters than expected, and every character it finds makes room for a phrase from the list.
The end-of-quarter decision: do we start review mining? The answer depends on your rivals' review volume; where it is low, phrases from that channel stay statistically weak.
Quarter three: go deeper
By the third quarter six months of measurement have accumulated and the time series can be read.
This quarter's first job is band persistence: which phrase spent how long in which band? Phrases that stayed at 11-30 are your natural band; ones oscillating between the top 10 and the band are on the edge and the most fragile group.
The second is scanning for silent losses: which phrases were higher six months ago than they are today? That list is usually surprising, and it contains losses that can still be undone.
The third is revisiting the wall list. If your rating count has grown, phrases that looked unreachable may now be reachable.
The fourth is working out the win rate per kind of move. About twenty moves have been made in six months, which is a meaningful number per type. Dropping the lowest-rate type is this quarter's highest-return decision.
The end-of-quarter decision: which piece of work do we stop? A country, a group of phrases, or a kind of move. Stopping frees more time than adding.
Quarter four: consolidate
Nothing new is started in the last quarter; what has been won is defended and the ground for next year is prepared.
The first job is defence. The top-10 phrases are checked weekly. If one has fallen into the band, the cause is usually the last move's side effect and caught immediately it can be undone.
The second is seasonal preparation. Many categories move in the last quarter of the year. A metadata move for a seasonal phrase has to be made a few weeks before the season, because the index takes time to settle.
The third is gathering the record. The year's moves, external events and measurement gaps are collected into one document. That document is next year's starting point and it keeps the knowledge in the team when somebody leaves.
The fourth is writing next year's decision calendar. Which quarter answers which question - questions, not moves.
What is deliberately not done this quarter: a large metadata revision. Rewriting every field during the busiest part of the year pollutes the measurement window at the worst possible time.
- No new scope in the last quarter; what was won is defended.
- Seasonal moves go in a few weeks before the season.
- The year's record is gathered into one document; it is needed for handover.
- A large metadata revision is not made in the busiest part of the year.
What cannot be planned
For a yearly plan to be honest, the things it cannot plan have to be written down too.
Rival movement cannot be planned. A rival refreshing their metadata, being pulled, or a new app entering the category changes your ranks while you do nothing.
Changes on Apple's side cannot be planned. How often the index refreshes and how heavily fields count change from time to time, and they are not announced.
The product itself cannot be planned - at least not by ASO. A rating drop after a release weakens your strength across every phrase and metadata cannot compensate for it.
The severity of seasonality cannot be planned. Whether a period will be busier or quieter than last year cannot be known in advance.
Those four are why the plan is a decision calendar rather than a list of moves. A decision calendar withstands that uncertainty; a list of moves does not.
The rhythm that stays constant
The quarters change but the weekly and monthly rhythm stays the same all year. It has to, because the comparability of the measurement depends on it.
Weekly, about two and a half hours: half an hour to look at the measurements and pull the band, an hour to choose the move and write the metadata, half an hour to ship, half an hour to compare a week later.
Monthly, about two and a half hours: an hour to re-scan the pool, half an hour to update the rival list, an hour on review mining.
Quarterly, about an hour: band persistence, silent losses, win rates per move type, the wall list, and one decision to stop something.
About a hundred and forty hours a year in total. That number is the plan's most concrete part, and the comparison is made from it: where else could those hundred and forty hours have gone?
| Rhythm | Time and work |
|---|---|
| Weekly | 2.5 hours - measure, choose, move, compare. |
| Monthly | 2.5 hours - pool, rival list, review mining. |
| Quarterly | 1 hour - persistence, silent losses, one thing stopped. |
| Yearly | ~140 hours in total. |
What will be known at the end of the year
The only thing a plan can commit to is which questions will have been answered by the end of it.
Which kind of move works for this app? A record of thirty or forty moves produces a win rate per type. That answer is in no general guide, because it belongs to the app.
What is the market's tempo? The control group's average weekly movement tells you how large your moves have to be. If the market moves three places a week, a two-place gain cannot be measured.
Which phrases are out of reach? The ones above the strength threshold have been set aside and no time is spent on them. That separation is the year's largest saving.
Which country is most productive? First-page phrases won per country decides where the time goes - and it is usually not the country with the most downloads.
And which work did we stop? The quarterly stop decisions have brought the scope down to something genuinely workable by the end of the year.
The one thing that cannot be committed to is a particular rank. Apple's algorithm is not published and what rivals will do is nobody's to control; a plan promising a first place is promising something it cannot keep.
Frequently asked
- Is a yearly plan really possible?
- As a decision calendar, yes; as a list of moves, no. What you do in month three depends on what you measured in month two, and a list written in advance compels you to repeat a move even when the measurement contradicts it.
- When should I add the second country?
- When the band in the first has emptied - when nothing is left waiting to be spent. If it is still full, adding a country divides your attention and ends with half a job in each.
- How many moves happen in a year?
- One a week is about forty. If your release rhythm does not allow that the number drops, and choosing the move becomes far more critical.
- What if the plan does not hold?
- Built as a decision calendar there is no such thing as not holding - the questions still get answered, the answers just come out differently than you expected. What fails to hold is a list of moves, and that should not have been written in the first place.
- Can I commit to a particular rank by the end of the year?
- No, and anybody who does should not be trusted. Apple's algorithm is not published and what rivals will do is not in your control. What can be committed to is knowing which move works for you.
Explore our resources
rankcusp's concrete return: why search visibility is your next growth lever
Ready to turn ASO into your next growth lever? This report explains how to turn keyword tracking from a maintenance task into a concrete return. We show what a cusp band word means for downloads once it reaches the top ten, and why phrases drawn from competitor reviews bring cheaper traffic. Download "rankcusp's concrete return" and turn ASO into a measurable channel!
See resource
The rankcusp guide: keyword strategy (the cusp band)
Discover rankcusp's new keyword strategy guide! This resource shows step by step which words to set realistic targets on and how to carry cusp band keywords into the top ten. Build your own ASO road map with rankcusp, raise your app's visibility in search and catch the searches hidden in competitor reviews before anyone else. We are with you at every step!
See resource