Methodology — how bumetric calculates revenue, build cost, audience & risk
Every number on every /p/{slug} page comes from one of the formulas below. Sources, inputs, accuracy bands and limits are public. No black box.
1. Where our data comes from
Every public metric on bumetric comes from one of four sources, all labelled in the snapshot history of each app page:
| Source | What we read | How often |
|---|---|---|
| iTunes Lookup API | iOS app metadata, ratings, icon, screenshots, current price, version, last update date, release notes, file size | On-view, then weekly |
| Google Play Store | Android app metadata, ratings, install band, developer info, last update date | On-view, then weekly |
| App Store Connect / Stripe Connect | Real revenue numbers, supplied by app owners who claim their listing | Pulled the moment an owner verifies |
| Acquired-deal anchors | Founder-declared MRR at the moment an app sold on Flippa / Acquire.com / MicroAcquire | Manually curated, weekly batch |
We do not scrape App Store Connect (it would violate Apple's terms). We do not buy mobile attribution data from third parties. Every estimate is derived from public signals plus the calibration set described in section 5.
2. The BU Score
A single 0-100 number summarising overall app health, shown in the hero of every /p/{slug}. Three weighted components:
- Rating quality (up to 50 points).
rating × 10. An app rated 4.7 stars earns 47 points. - Audience size (up to ~30 points).
log10(ratings_count) × 5. Rewards large rated audiences but discounts marginal gains past ~1M ratings. - Monetisation signal (5-15 points). Tier-based on inferred or verified monthly revenue. Free-with-no-IAP gets 0; verified $100K+ MRR gets the full 15.
Clipped to 100. The tier label (Transcendent / Excellent / Strong / Solid / Emerging / Niche) is a lookup based on the score band.
3. Revenue forecast
The dollar figure in the hero KPI bar and the "Forecast Revenue" chart. Produced in two steps:
- Base estimate from app metadata:
ratings_count, rating velocity (week-over-week growth in reviews), category-typical monetisation curves, in-app purchase signals, ad-supported flag, subscription pricing if visible in the IAP price range. - Calibration adjustment against the verified-MRR anchor set (section 5). The base estimate is multiplied by the category-specific scaling factor learned from anchor apps in the same niche and revenue tier.
No LLM in the loop. Gemini is used elsewhere on the site for ASO rewrites and growth recommendations; the revenue figure is a deterministic calculation.
4. Lifecycle cards
Five synthetic cards appear on every /p/{slug}, in this order: Build cost → Exit value → Audience → Tech stack → Maintenance risk. Each answers a single founder question.
How much did this app cost to build?
Synthetic estimate calibrated against 2024-2026 indie-agency rate surveys.
Formula: base $28K + (screens × $4.5K) + IAP layer + cat_multiplier + platform_multiplier + (age × team_size × $6K/year)
- Screens: count of store screenshots, floor 5.
- IAP layer: +$18K for subscription billing, +$9K for one-time IAP, $0 for free apps.
- Category multiplier: 2.2× for games, 1.6× for finance, 0.9× for utilities.
- Platform reach: ×1.55 if Android (proxy for cross-platform build), ×1.0 for iOS-only.
- Age × team: maintenance proxy. Team size from
developer_apps_count(solo / 2-5 / 6-20 / studio).
Rounded to nearest $5K. Range shown as ±25/35%.
What would this app sell for on Flippa / Acquire.com?
Synthetic estimate calibrated against 2024-2026 indie-app-marketplace closed-deal medians.
Formula: exit_value = MRR × multiple_months
| Monetisation model | Multiple (months of MRR) |
|---|---|
| Subscription | 42 months (~3.5y of MRR) |
| IAP / freemium | 28 months |
| Paid one-time | 21 months |
| Ad-supported | 15 months |
| Free + large audience | 12 months |
Adjusted by: app age (+5% per year, capped 20%), rating quality (4.7+ → +15%, <3.5 → -18%), category demand (Finance +15%, Casual Games -22%), user-base maturity (1M+ ratings → +8%). Hidden when MRR is below the $500/mo marketplace floor.
How many users does this app actually have?
Lifetime downloads, monthly active users, and daily active users — derived from ratings count and category retention benchmarks (Adjust 2024, AppsFlyer 2025, Branch 2026 mobile reports).
Lifetime downloads: ratings_count × downloads_per_rating × age_decay. Productivity ≈ 75 installs per rating; Casual Games ≈ 220.
MAU: lifetime × D30_retention × churn_decay(age). Productivity D30 = 30%, Casual Games D30 = 6%.
DAU: MAU × dau_mau_ratio. Social = 55%, Productivity = 45%, Casual Games = 25%.
Numbers are US-storefront-equivalent. Real global audience can be 5-10× larger for internationally-distributed apps. We say so on the card.
What is this app built with?
Heuristic guess with declared confidence band (High / Medium / Low).
- Description scan wins instantly: "Built with React Native" / "Flutter app" / "Made with Unity" in the store listing → 95% High confidence.
- File-size band: <30 MB → native; 30-60 MB → Flutter or SwiftUI; 60-130 MB → could be either; 100+ MB game → Unity; 400+ MB → Unreal.
- Platform + era: iOS post-2020 default to SwiftUI; pre-2020 to UIKit; Android default to Kotlin.
- Category nudge: Finance / Medical / Business skew native (compliance audit).
Defaults to native; React Native / Flutter only headlined when there's a strong signal. Alternate stacks shown as chips. We do not disassemble binaries.
Is this app still alive?
0-100 risk score with verdict (Active / Slowing / At risk / Likely abandoned). Calibrated against 2024-2025 store-ranking-decay reports.
| Days since last update | Base risk |
|---|---|
| <30 days | 5% — Active development |
| 30-90 | 15% — Healthy cadence |
| 90-180 | 30% — Slowing |
| 180-365 | 50% — Concerning gap |
| 365-730 | 72% — Likely deprioritised |
| 730+ | 88% — Likely abandoned |
Adjustments: Editor's Choice / curated featuring -10%, <500 ratings + 2y+ old +10% (solo-project pattern), 10y+ old + 1y+ no update +5% (API drift).
5. Verified-MRR calibration
Without ground-truth, any revenue estimate is just an opinion. Our anchor set is the moat:
- App owners who claim their listing can verify real revenue through three paths: Stripe Connect OAuth (read-only), App Store Connect login (read-only), or owner-submitted screenshots that we OCR and audit.
- Acquired-deal anchors: when an app sells on Flippa / Acquire.com / MicroAcquire and the listing discloses real MRR, we add it to the calibration set.
- Each verified anchor calibrates the forecast for every similar app in its category and revenue tier.
- The anchor set grows weekly. More anchors per niche → tighter confidence band on every other app in that niche.
This is the part nobody can shortcut. Anyone can copy a scraper. The verified-MRR set takes years of trust-building to assemble.
6. Accuracy & confidence bands
Every forecast is labelled with a confidence band derived from how many anchors exist in the same niche:
| Anchors in category | Typical accuracy | Confidence |
|---|---|---|
| 10 or more | ±15% | High |
| 3 to 9 | ±25% | Medium |
| 1 to 2 | ±40% | Low |
| 0 | category median fallback | Indicative only |
A "Low" confidence label means the dollar number is directionally right (correct order of magnitude, correct relative ranking versus peers) but the absolute figure should be read with healthy skepticism.
7. Worked example — Snapchat
Take Snapchat (/p/447188370). Here is every number on that page traced to its source:
| Field | Value | Source |
|---|---|---|
| Rating | 4.5★ | iTunes Lookup → averageUserRating |
| Ratings count | 5,723,148 | iTunes Lookup → userRatingCount |
| Last update | 2026-05-07 | iTunes Lookup → currentVersionReleaseDate |
| BU Score | 87 / Strong | 4.5 × 10 + log10(5.7M) × 5 + 15 monetisation = 87.4 |
| Forecast Revenue | $24,108 / mo | Estimated_revenue from category-calibrated model; not a reliable global figure (Snapchat is a global super-app, our number reflects US storefront signal only) |
| Build cost | $230K range $170-310K | $28K base + 5 screens × $4.5K + 180 MB asset bump + Social category ×1.7 + Android cross-platform ×1.55 + 14y maintenance × ~5-person team × $6K/y |
| Exit value | $500K · 21× MRR | $24K × 21 (Social model × age × rating × category) |
| Audience | 565M lifetime / 23M MAU / 6.9M DAU | 5.7M ratings × ~90 installs/rating × age decay (US-storefront equivalent) |
| Tech stack | Native iOS (Swift/UIKit), Medium 55% | 180 MB bundle on pre-2020 iOS app, no React Native description marker |
| Maintenance risk | 5% Active | Last update 2026-05-07 = <30 days ago → base 5%, no adjustments |
Two things worth noting in this example: the forecast revenue figure ($24K/mo) is much lower than Snap Inc's company-wide revenue because we model App-Store-visible signal only. And the build cost of $230K reflects what a competitor would pay to clone the iOS app today — not what Snap actually spent originally.
8. Data freshness
Every /p/{slug} page is live. When you open a page for an app we haven't scanned recently, our worker fetches new metadata in the background and saves a fresh snapshot before you finish scrolling. The "Last update" timestamp on the page is the moment we wrote that snapshot.
Cron jobs additionally re-scan high-traffic apps weekly to keep the snapshot history smooth. As of 2026-05 we are running a one-shot metadata refresh of the full iOS catalog to fix stale last_update_date fields on older apps.
9. When NOT to trust us
- Paid acquisition spend. An app that buys 100K downloads at $3 CPI shows up the same as an app with 100K organic downloads. Revenue is real either way, but the unit economics are not the same.
- Refund rate. Apple and Google publish gross revenue. Net revenue after refunds is typically 5-10% lower; we report gross.
- Geographic split. We give a single US-storefront-equivalent figure. International apps with most users outside the US can be 5-10× larger globally.
- Future revenue. The forecast is a current-state estimate, not a prediction of next month. App revenue is volatile, especially in categories with seasonal demand.
- Internal-only apps. A "personal app" used only by the developer's family has the same App Store signal as a real product — we can't tell from public data.
- Brand-new apps (<30 days). Too little ratings velocity to make a meaningful estimate. Forecast is shown but with "Indicative only" confidence.
- Apps with verified MRR > $1M/mo. Mega-tier apps are calibrated against few anchors; bands widen.
If you spot a clearly wrong number, see section 11.
10. Spotted an error?
Two options:
- Claim your app and verify the real number through Stripe Connect or App Store Connect (free, ~30 seconds, replaces our estimate everywhere).
- Email us with a screenshot or evidence. We respond within two business days and update the page (or remove it) within a week.
The reputation of this site depends on our numbers being closer to ground truth than competing scrapers. We take corrections seriously.
11. FAQ
Why does the forecast for [Snapchat / Spotify / Disney+] look "too low"?
How is "verified MRR" different from "estimated revenue"?
Why do build cost and exit value use different formulas?
exit_value / build_cost = N× return.Can I download the raw data?
How do I know the calibration anchors aren't cherry-picked?
What's different from Sensor Tower / data.ai?
12. Changelog
2026-05-23 · Methodology page rewritten. Added sections on the five lifecycle cards (Build cost, Exit value, Audience, Tech stack, Maintenance risk), a worked example with Snapchat, "When NOT to trust us" callout, and an FAQ.
2026-05-18 · BU Score weights documented in full. Confidence-band table added.
2026-05-11 · Page launched.