aso-research5.91 KB
View saved version →
---
name: aso-research
description: Researches and improves an app's visibility on the App Store and Google Play with Applyra's data (keyword rankings and their history, keyword difficulty and traffic, store autocomplete, niche analysis, competitors, top charts, the ASO Health audit of a listing), and scores a draft title, subtitle, short description or keywords field before it ships. Use when the user asks why their app dropped in search, which keywords to target, how they rank against a competitor, what to write in their store listing, or anything about App Store Optimization for an app tracked in Applyra.
---
# ASO research with Applyra
Applyra tracks the user's apps on the App Store and Google Play every day. Its tools read that data,
and score listing text on the same engine that audits live listings. No tool publishes anything to
App Store Connect or Google Play: the user ships changes themselves.
## Ground rules
- **Start from `list_applications`.** It gives each app's numeric `id`, which every app-scoped tool
takes, with its store, country and language. Never guess an ID. `add_application` and
`add_competitor` are the exceptions: they take the store bundle ID.
- **Every number comes from a tool call.** Say which tool it came from, with the store, the country
and the date of the data. Quote difficulty and traffic as the 0-100 scores Applyra returns.
- **Rankings are per store and per country.** Keep markets apart, never blend them into one figure.
- **Let `check_metadata` count lengths.** The stores count UTF-16 code units, not characters, so a
count done by hand is wrong in many languages.
- **Ask before changing the account.** `track_keywords`, `untrack_keyword`, `set_keyword_favorite`,
`add_application`, `add_competitor` and `remove_competitor` change what Applyra tracks, and
`add_note`, `update_note` and `delete_note` change the notes on its charts. Every other tool only
reads.
- **Reuse past work first.** `list_keyword_inspections`, `list_niche_analyses`,
`list_metadata_simulations` and `list_autocomplete_history` return what was already run.
## What moved
1. `list_keywords` for the app: current rank, the apps just ahead and behind, the top 5.
2. `get_keyword_rank_history` on the keywords that moved, with the `keyword_id` from `list_keywords`.
It covers the last 30 days unless a range is given.
3. Report before and after, largest drops first. A move of one or two places is daily noise: lead
with the large ones.
4. When many keywords fell on the same day, `get_app_score_history` shows whether the app's overall
visibility fell with them. A drop across the board on one date usually comes from the store, not
from the listing: its `events.store_updates` list the store algorithm updates of the market.
## What a change did
1. `list_notes` shows what the user noted (a campaign, a metadata change, a launch); `compare_events`
gives the visibility before and after each note and each app update, or, with `keyword_id`, one
keyword's rank. 7 days on each side by default, 14 or 30 for a slower effect.
2. Read each comparison with its `alongside`: another event or a store algorithm update in the same
days is mixed into it. Say so rather than credit the change with the whole move.
3. A user who describes something they did on a given day can have it noted with `add_note`, after
asking.
## Keywords worth targeting
1. Collect candidates: `run_autocomplete` on the app's core terms (what people type), and
`run_niche_analysis` on its topic (clusters of keywords with an opportunity score). A fresh niche
analysis takes a few minutes.
2. `inspect_keyword` each candidate: difficulty, traffic, KEI, the 20 apps ranking for it.
3. Prefer real traffic at a difficulty the app can win. `get_aso_health` calibrates its targeting
verdicts on the app's size (tier 1 Emerging, 2 Growing, 3 Established): the same keyword can be
reachable for an established app and out of reach for a new one.
Preview of the first 4,000 characters. Read full instructions →
