How long does this process take?

Baseline measurement and first metadata revision take 1-2 weeks. First creative A/B test takes 4-6 weeks to reach statistical significance. Category rank improvement typically materializes in weeks 4-8. A full ASO programme iteration cycle — baseline, metadata fix, creative test, velocity push, measure — runs roughly 8-12 weeks.

Can I do this without paid UA?

Yes, but it's slower. Organic-only ASO compounds, but the first velocity spike that moves category rank usually comes from a coordinated push (press, seasonal hook, Apple editorial submission, or paid). Without any velocity trigger, you can still move search ranks through better metadata, but category rank improvement will be gradual over months.

How do I know if my baseline is broken or just normal?

Compare to the top 3 apps in your category. If their icon uses bright contrasting colors and yours is monochrome; if their first screenshot has a captioned hero benefit and yours shows raw UI; if their title includes a descriptor keyword and yours is brand-only — those are broken baselines. If your metadata, creatives, and review count are directionally similar to the top 3 and you're still at rank 60, the issue is velocity, not baseline.

Should I run A/B tests on iOS before or after fixing metadata?

Metadata first. Metadata changes affect both impression volume and conversion; creative A/B tests isolate conversion. If you run a creative test on under-optimized metadata, you're measuring creative efficacy against a shrinking impression pool — noisy and slow to reach significance. Fix metadata, wait 2-3 weeks for it to stabilize, then run creative tests.

How often should I revisit this process?

Run the full cycle at least twice per year. Between cycles, monitor keyword rank and conversion weekly, and ship minor iterations (one metadata tweak, one creative element) monthly. Category competition shifts faster than most publishers realize — a top-10 category position is durable only if you defend it.

How to Rank Your iOS App on the App Store — A Step-by-Step Guide

This guide walks through the iOS ASO process end-to-end, in the order a working publisher would execute it. It assumes you have an app live on the App Store with some baseline of installs, not a pre-launch app. The methodology is equally applicable to Google Play with adjustments for metadata fields (no keywords field, more description weight).

Before you start — baseline metrics

You can't measure improvement without a baseline. Collect and record the following before you change anything:

Current keyword ranks for every keyword you care about, per target country. Use a dedicated ASO tracker (MWM Scale has a free tier for this).
Current category rank and Overall rank in your top 3-5 markets.
App Store Connect impressions — product page views per day, split by Search / Browse / Referrer.
Conversion rate from product page view to install.
Review velocity (new reviews/week) and rating (rolling average).
Install volume split — organic Search, organic Browse (chart), paid, referral.

Snapshot these and date-stamp them. You'll compare against this in 4-8 weeks to measure the impact of changes you ship.

Step 1 — Keyword research

The goal of keyword research is to produce a target keyword basket — 10-20 keywords you actively optimize for and track. Not 200, not 5. The process:

Generate seed keywords. Start with your category's obvious terms (e.g., "meditation" for a meditation app, "expense tracker" for finance). Expand with synonyms, audience descriptors ("meditation for anxiety"), and outcome-based phrases ("sleep better").
Add competitor keywords. Take your 3-5 direct competitors, pull the keyword basket they rank for, and add the ones you aren't targeting.
Check search volume and difficulty. Use your ASO tool. Target mid-difficulty keywords with real volume — heading-term keywords like "meditation" are crowded; long-tails like "mindful breathing meditation for sleep" are easier to rank for.
Filter for intent. A keyword that drives installs from unqualified users is worse than a keyword with less volume but stronger buyer intent. If your app is a premium subscription, "free meditation" is a poor match.
Prioritize 10-20 keywords. Cap the list. You'll use these to structure your metadata, creative copy, and review acquisition strategy.

Step 2 — Optimize your metadata

With a target keyword basket in hand, place keywords strategically across the iOS metadata fields. The goal is full coverage without stuffing.

App name (30 characters). Your primary keyword goes in the descriptor position. Typical patterns:

BrandName: Primary Keyword — e.g., Strides: Habit Tracker
BrandName — Descriptor — e.g., Calm — Meditation & Sleep
Primary Keyword by BrandName — e.g., Meditation by Calm (less common; weaker branding)

Subtitle (30 characters). Secondary keywords phrased as a benefit. Comma separation is fine and doesn't waste indexation.

Good: Workouts, Sleep, Nutrition
Bad: The Best Health & Fitness App You'll Ever Need (all filler, no keywords)

Keywords field (100 characters, comma-separated, hidden). Singulars not plurals; synonyms; long-tail terms; misspellings. Don't repeat words already in title or subtitle — Apple combines fields. Don't add spaces after commas — that wastes characters.

Promotional text (170 characters, non-indexed, editable without review). Use for conversion copy, seasonal promotions, recent accolades. Not for keywords.

Description (4,000 characters, weakly indexed). First 3 lines matter most (users see them above the "more" fold). Structure:

Lead paragraph: one-sentence value prop.
Benefit-focused H2-style sections (bullets work well).
Features list.
Social proof (reviews, press).
Support contact and review request.

Ship the metadata change as a new app version submission. Expect initial keyword rank shifts within 48-72 hours, full stabilization in 7-14 days.

Step 3 — A/B test your creative

Once metadata has stabilized (2-3 weeks after shipping), start creative testing. Apple's Product Page Optimization lets you run 2-3 concurrent variants per test, with the system auto-terminating at significance.

Icon tests. Highest-leverage single surface. Common hypotheses:

Character vs. letter-mark (does a character-driven icon beat an abstract logo?)
Colour contrast (brighter, more saturated icons often win on the crowded chart)
Focal element (one clear central shape vs. composite)

Screenshot tests. First two screenshots carry 60-80% of conversion weight. Common hypotheses:

Caption-forward (short headline text overlaid) vs. UI-forward (raw product screenshot)
Benefit statement (what the user gets) vs. feature statement (what the app has)
Device frame (phone mockup) vs. no frame (full-bleed)

Apple's PPO only rotates variants to a fraction of traffic, so reaching significance on lower-volume apps can take 4-6 weeks. Don't kill a test prematurely — conversion deltas of 5-10% look noisy short-term but compound enormously over months.

We observe that most apps underweight the icon test. A 15% icon-driven CVR lift on a 10-million-impression-per-year app is 150k incremental installs — usually more value than an entire creative refresh of the screenshots.

Step 4 — Drive review velocity

Rating and review velocity is a chart-ranking signal, not just a conversion signal. Apps with stale reviews underperform apps with fresh ones.

The iOS review prompt API (SKStoreReviewController) is the only clean mechanism. Apple rate-limits prompts to 3 per user per year, so be strategic:

Trigger on success, not arrival. Prompt after the user completes a key success state — a workout finished, a habit logged for the 10th day, a file exported. Not on app launch.
Wait 7-14 days post-install. Early prompts catch uncommitted users; too-late prompts miss the window.
Gate on engagement. Users who've used the app fewer than 3-5 sessions are statistically likely to leave negative reviews — skip them.
Don't chain with support. Never prompt for a review immediately after a negative support interaction.

Well-tuned prompting yields review velocity of 0.5-2% of new installs. Poor prompting yields 0.1-0.3%. The delta matters.

Step 5 — Fix retention leaks

Retention is both a ranking signal and the base rate for LTV. Audit where users drop off:

Day 1 drop-off: usually onboarding friction. Shorten permission asks, remove non-essential signup requirements, get to "first value" in under 60 seconds.
Day 7 drop-off: usually habit formation failure. The user didn't build a reason to return. Push notifications, streaks, or scheduled content can help.
Day 30 drop-off: usually feature completion or content exhaustion. Add progression, new content drops, or social mechanics.

A 5-point improvement in D7 retention (from 20% to 25%) usually drives 15-25% category-rank improvement over 4-8 weeks, holding everything else constant — because the behavioural signal propagates into ranking slowly.

Step 6 — Measure, iterate, track

Monthly reporting cadence:

Keyword ranks (per country, per keyword) vs. baseline
Category rank and Overall rank vs. baseline
Impressions, conversion rate, new installs vs. baseline
Review velocity, rating, new review count
Retention curves (D1, D7, D30)

Quarterly:

Revisit keyword basket — are new opportunities opening up?
Revisit competitor creative — are they iterating faster than you?
Revisit localization — are secondary markets under-invested?

Annual:

Repeat steps 1-5 from scratch. Your baseline, competitor landscape, and search-demand patterns have all moved.

Timeline and realistic expectations

A disciplined first-cycle ASO programme typically delivers:

Week 1-2: baseline captured, metadata v2 drafted.
Week 3: metadata v2 submitted, passes review.
Week 4-6: keyword rank shifts visible; first creative test starts.
Week 8-10: creative test reaches significance; winning variant rolled out.
Week 10-12: category rank improvement becomes measurable, 10-30% depending on starting position.

Publishers expecting top-10 category rank in 8 weeks from a fresh ASO programme usually underestimate the velocity component. Ranking compounds — the sixth quarter of a disciplined ASO programme usually delivers more rank movement than the first.

Common pitfalls

Shipping a full metadata overhaul and creative test simultaneously — you can't attribute which change drove what.
Copying a top competitor's metadata verbatim — Apple penalizes near-duplicate metadata, and you inherit their keyword strategy, not necessarily the right one for your app.
Running 5 creative tests concurrently — attribution gets messy, significance stretches, and you learn less faster than running them in series.
Treating review velocity as a one-time project — it's a sustained programme. Apps with fresh, recent reviews rank better than apps with stale ones, regardless of total count.
Ignoring non-English markets — localization is often the highest-ROI quarter of work for mid-sized apps.

Where to go next

If you publish games: ASO for Mobile Games.
To understand the signals you're optimizing against: How the App Store Ranking Algorithm Works.
For the broader reference: The Complete ASO Guide.
To run live keyword and rank tracking: MWM Scale's ASO tool is free for up to three competitor apps.