Stringers
Mobile App · Android · written 11-9
杯筊天機 · Shanghai Xiangxiang Network Technology Co., Ltd

A small Shanghai studio built a temple inside your phone: 「杯筊天機」 hit Taiwan's Lifestyle chart, powered by paid acquisition rather than word of mouth

What it is It moves temple visits into your phone: ten deities for fortune sticks, 3D moon blocks, incense offerings to earn merit, and chanting, plus BaZi chart readings and AI interpretation. Revenue comes from memberships, in-app purchases of offerings, and "随喜" (voluntary donations).

Why it blew up It's a small-budget Android ad test. The privacy policy says it connects an attribution tool to measure "our own ads," and two days after a new release it charted in four Chinese-speaking regions at once. The Taiwan Lifestyle Top Free chart has a low bar, a few hundred installs a week is enough to land around #45.

BHow you could follow Don't build another moon-block tool. Either take "altar + membership + voluntary offerings" to the iOS side, where it hasn't gone yet, with AI fortune readings built on the poems, or switch to local deities and devotional customs in Vietnam, Thailand and Malaysia. Both rely on IG / FB and app ad acquisition, and you have until the 2027 Feb 6 Lunar New Year.

Grade B: Chinese folk "online offerings + AI fortune readings" have been proven to make money by FateTell (#40 Taiwan Top Grossing in its category) and CMoney's Wu Cai Shen. Small regions have a low bar, and one person can run it. But this app hasn't yet shown that paid acquisition pays back, and similar apps flooded the store in the second half of 2026. It has no organic traffic, so it won't take off without ads. Almost nobody knows about it yet

Est. monthly revenue (Google Play) under $1KWe estimated this from its rank on country top-grossing charts and real install counts (it isn't on the grossing chart, so confidence is lower), calibrated against real revenue; typical error is 2–3x. Read it as a tier, not an exact number.

Data (updated daily) · first launch 2026-08-29 · 11-20 installs 1,665
Android · Total installs worldwide
01k2k11-0511-0911-1311-1611-201.7k
Data 2026-11-05 → 2026-11-20 (16 days, 16 observations). Google's detail page only refreshes install counts every other day, so a flat day doesn't mean nobody installed.

"Bei Jiao Tian Ji" (杯筊天機) is a Traditional Chinese "online temple" from a small Shanghai company. It has ten deities' fortune-slip libraries, 3D moon-block tossing, a nine-slot altar where you burn incense to earn merit, chanting, BaZi charts and AI fortune readings, and it makes money from subscriptions and in-app offerings. It launched at the end of August, stayed quiet for over two months, then in Nov 5 suddenly showed up at the same time on the Google Play Lifestyle Top Free charts in Taiwan, Hong Kong, Singapore and Malaysia. By Nov 10 it had only 817 installs in total. Here's what it sells, where this wave of traffic came from, whether anyone on this path has actually been paid, and whether you should follow it.

Nov 5: four charts lit up at once

On Nov 4, Bei Jiao Tian Ji shipped build 1.0.6 on Google Play. Two days later it charted for the first time: #56 Taiwan Lifestyle Top Free, #67 Hong Kong, #69 Singapore, #178 Malaysia. Over the next three days it shipped two more builds, 1.0.7 and 1.0.8 (Google Play charts and version history, 11-05 to 11-07).

Before that it had never charted. It launched in Aug 29, and it never appeared on the Taiwan Lifestyle chart in the daily charts from 10-29 to 11-04. Two-plus months of silence, then four Chinese-speaking regions on the same day. That doesn't look like word of mouth spreading slowly. It looks like someone flipped on an ad switch (inference, evidence below).

On Nov 7, the Play listing got two reviews, one saying 「還行」 and one saying 「讚!!!」. By Nov 9, the store page showed 4.8 stars and 23 reviews.

It's published by "Shanghai Xiangxiang Network Technology Co., Ltd." The contact email on Play is a personal gmail, the address is in Baoshan, Shanghai, the phone is +86 mobile number, and the company has only this one app. It looks like a team of one or two people (inference).

A temple in your phone: what it sells

The store page opens with: "For thousands of years, people have walked into temples when torn and undecided, tossed moon blocks to ask the gods, and drawn lots for answers." It moves the whole temple visit onto your phone:

  • Ten deities, each with their own domain: Guanyin, Guan Gong, Mazu, the God of Wealth, the Buddha, the Matchmaker, plus Green Tara, Yellow Wealth God, Third Prince and the Monkey King. The store page reads "For marriage ask the Matchmaker, for career ask Guan Gong, for wealth ask the God of Wealth." Each slip library comes with its legend and a plain-language explanation, in both simplified and traditional Chinese.
  • The full slip-drawing ritual: hold the slip cup and shake until a stick flies out. Then toss the moon blocks 3D to ask "is this the right slip?" Holy, laughing and yin blocks land and tumble.
  • Nine-slot altar: three incense sticks, candles, flowers, oil lamps and fruit baskets, each with a duration and a multiplier. You earn "incense merit" while offerings are up, and tapping the wooden fish adds more. You spend merit on new offerings, and every entry in the ledger is viewable.
  • Chanting hall: Heart Sutra, Great Compassion Mantra, Heavenly Holy Mother Sutra, Taishang Qingjing Sutra, with timed audio and a dedication at the end.
  • The fortune-telling layer: BaZi charts calculated down to the minute (true solar time and late-night Zi hour included), and AI reads your slips line by line from your BaZi, with follow-up questions. Your daily fortune only "reveals" once you toss a Holy block that day.

Money comes from three places: an auto-renewing membership (multiple profiles, cloud backup of records, photograph a physical offering and the AI picks an elegant name and writes a blessing), a one-time "Bone Weight" unlock, and a tip jar called "Voluntary Offering", the equivalent of incense money. On Play, the in-app purchase range is listed at $0.99–$99.99 per item, and there are no ads inside the app (Google Play store page, 11-09). The subscription price isn't visible from outside.

The cost base is light. The privacy policy says the AI runs on Volcano Engine Ark (ByteDance's model platform), the servers run on Cloudflare Workers, there's no account system, data is stored on the phone, and cloud backups are written to the user's own Google Drive. One person can run it (inference).

Its website still links an iOS privacy policy, but on Nov 9 a search by name in the Taiwan, Hong Kong and US App Store turned up nothing, so it's unconfirmed whether the iOS version is live.

817 installs, and a number that doesn't add up

DateTotal installsTaiwan Lifestyle Top FreeHong KongSingapore
11-0541566769
11-0648505969
11-07485664109
11-086054447180
11-096424445187
11-108174561179

(Google Play installs and Lifestyle Top Free ranks. Malaysia 11-05, 11-06 at #178, 168, then it dropped off the chart.)

At first glance, the install count jumped 557 times in a single day on Nov 8. But that number is most likely Play's install count catching up a few days late in one batch, not a real one-day spike (inference). The ranking doesn't add up: from 11-05 to 11-07 it sat around #50 on Taiwan's lifestyle chart, while its neighbors at #40–50 on the same chart were adding hundreds of installs a week. On Nov 10, the #48 app 「旅電 PLUG-INN」 added 405 over 7 days, and the #38 app 「夢時代」 added 461. An app with only 41–48 total installs can't hold that rank. When the count jumped to 605 11-08, its rank only moved from #56 to #44, which suggests the actual pace didn't change much.

So the closer picture is a few dozen to a hundred-plus installs a day, mostly Taiwan and Hong Kong (estimate). It never made any country's Top Grossing chart.

This also shows a fact about small Chinese-language chart markets: Taiwan's Lifestyle Top Free chart sits around #45 with just a few hundred installs a week. The bar is low, a little budget can get you on it, and people browsing the chart will see you.

Traffic came from paid ads, and its own privacy policy says so

Where did this wave of traffic come from? Its Android privacy policy (updated October 2026) contains this line: "To measure the effectiveness of our own ads, we read the Google Advertising ID and send installs and specified conversion events to Adjust."

The events sent back are clearly listed: completing a bond, first slip draw, viewing the membership page, starting a trial, completing a subscription, the Bone Weight unlock, and Voluntary Offering. This set looks built for conversion-optimized campaigns (such as Google App campaigns), with trials and subscriptions as the goal (inference). Interestingly, its iOS privacy policy says the reverse: "no cross-app tracking." Paid acquisition seems to be on Android only, which skips Apple's "allow tracking" prompt on iOS.

The timing lines up: the privacy policy was revised in October, the build shipped in Nov 4, and four Chinese-speaking regions charted together on Nov 5. There's no direct evidence of which channel was used or how much was spent.

Other channels are nearly empty. Searching TikTok for "杯筊天機" or "擲筊app" turns up nothing from it, nobody mentioned it on X in 60 days, a web search for the exact name "杯筊天機" returns zero results, and no media has covered it. Searching "擲筊" in the Play Taiwan store, it ranks #11, behind 3D擲筊 and CMoney's Wu Cai Shen / Ying Cai Shen, familiar names (Play search results, 11-09).

The product is more complete than its competitors, but that isn't why it got traffic. Differentiation matters for what comes next: whether users stay and whether they pay, which decides whether this ad spend pays back. There's no data on that yet.

On this path, has anyone actually been paid?

It hasn't proven anything yet, but others on the same path have already been paid.

The closest match is FateTell: Moon Blocks, Holy Blocks & Daily Horoscope on iOS. It launched in Oct 31, 2025, and in a year collected 1,063 ratings in Taiwan and 315 in Hong Kong. Its best Taiwan Lifestyle top-grossing rank was #40, Hong Kong #36, and Malaysia #21 (App Store, November). It also runs ads, and users say that's where they came from. A Taiwanese user wrote in March: "I just saw the IG ad with moon blocks and mystery boxes, so I thought I'd download an app for moon blocks to buy scratch cards." A Hong Kong user wrote a one-star review in Nov 6: "The Fb Ig ads promise the moon… then after signing up they start charging an annual fee."

FateTell's reviews show both sides of this business. Paying users are buying this:

"After buying the destiny book, lots of the details are really helpful, and my career is changing a lot this year... some of it even made me cry." (Hong Kong, 6-12) "I check the day's fortune every morning as a little reminder for the day." (Taiwan, 6-10)

Critics focus on three things: price, answering the wrong question, and AI that reads badly.

"After the trial ended, every answer I got was really weird, and your membership fee is outrageously expensive." (Taiwan, 8-06) "It completely ignores the hexagram and slip I drew, answering B when I asked A." (Taiwan, 10-31) "I got a Holy block, but the AI replied: your connection with the gods has been cut... I already got the Holy block!" (Hong Kong, 2-20) "What you buy here is ritual and design. If you spent money, tell yourself that." (Taiwan, 10-17, three stars)

That last line gets at what it really sells: not accuracy, but ritual. Bei Jiao Tian Ji pushes the ritual harder: offerings have timers, burn out, and send reminders to renew incense, with incense reminders on the 1st and 15th of each month. That's the bet.

Another proof point comes from a local Taiwan brand. CMoney's "Wu Cai Shen" offers lamp-lighting, slip drawing and lottery blessings, with 50k+ downloads and 5,130 reviews on Play. "Ying Cai Shen" has 10k+ downloads (Play store page). In Taiwan, people do pay for consumable online offerings like lamp-lighting and incense money.

Crowded or not

ProductPlatformScaleHow it differs from Bei Jiao Tian Ji
FateTell(AURAAI)iOSTaiwan 1,063 ratings, best Taiwan Top Grossing rank #40Moon blocks + AI fortunes + BaZi, already collecting money; no altar or merit
Shengbei Jiaobei: AI fortune-telling app (a Wuhan company)iOSTaiwan 1,511 ratings, 4.8 starsGoes the AI companion route; likewise a mainland team making for Taiwan
武財神 / 迎財神 (CMoney)Android / iOS50k+ / 10k+ downloadsMajor Taiwanese brand, specializing in wealth gods and lamp-lighting
3D擲筊 (BiBiJiJi Studio)Android100k+ downloads, 4.8 starsMoon-block tool only, free
Donggang Haigong Temple - Online Oracle LotsAndroid / iOSMultiple installs in the Play 70k rangeMade by the temple itself, free, carries religious authority
觀運 (HoolyHi)Android50+ downloadsNearly identical feature mix, not yet scaled
Mind Ball, Ask & Throw, Lucky Throw, Throw Jiao·3D Jiao Bei, Kaucim, etc.iOSEach with 0–6 ratingsClustered launches from 2026 7 through November

(Google Play and App Store store pages and charts, 11-09.)

Two things to see clearly. First, more players are coming in: in the second half of 2026, at least seven or eight new Jiaobei / divination-lot apps appeared on iOS, and indie developers in mainland China and Taiwan are all piling in. None of them have proven their numbers yet. Second, only one or two have actually collected money, and both of them run paid ads. Plain "Jiaobei tools" were already ruined by free apps and websites, and temples' official apps are free and carry real endorsement. What earns money is the kind that bundles BaZi, AI lot interpretation, daily fortune and ritual into a membership.

"杯筊天機" itself is still unknown: no media coverage, no social buzz, no disclosed revenue, not on any top chart, and nobody is copying it.

If it were you

B: Worth a try. The opportunity is there, but the weak points are clear. Online offerings plus AI lot readings in Chinese folk belief is a business that FateTell and CMoney have validated, with people paying. Local charts have a low bar, and one person can build and sustain it. The weak points: "杯筊天機" hasn't yet proven that paid acquisition pays back; this niche got more crowded in the second half of 2026; there is no organic spread, so without paid ads there's no traffic.

Who it suits: 1–2 people who understand Chinese folk religion (the division of deities' duties, lot poems, the 1st and 15th of the lunar month) and can write authentic traditional-Chinese copy. At least one of them has run Google app ads or Meta ads and can work out what a trial user costs. A mainland China team can absolutely do this; several on this page already are.

Who it doesn't suit: anyone counting on store search and word of mouth to grow organically (no organic growth example appears in this category); anyone who only wants a Jiaobei tool (the free 擲筊 3D already has 100k+ downloads, and you can toss on any website); anyone who isn't familiar with lot poems and deity systems and relies purely on AI improvisation (half of FateTell's bad reviews are about "asked A, answered B" and "holy blocks said the connection was broken").

How to do it: pick one of two paths. ① Fix what it does badly: it isn't visible on iOS yet, and FateTell has shown that users in Taiwan and Hong Kong on iOS will pay. Build the "altar + membership + donations" model on iOS, have the AI reading strictly follow the lot poem and the Jiaobei result, and let people toss, draw lots and read poems for free first, with payment buying deep AI readings, fate books and offerings. ② Change the region or the deities: the same "altar + consumable offerings + membership" structure could be applied to Vietnamese mother goddess worship and ancestor rites, Thai temple lot-drawing and amulets, or Malaysian Datuk Gong and Nine Emperor Gods (unverified, inferred). Either way, don't copy FateTell's model of charging an annual fee up front and turning hostile once the trial ends. That is where its one-star reviews pile up.

How to acquire users: buy traffic the way these two do. On Android, plug in an attribution tool and send back "first lot draw, trial started, subscription completed" events, bidding on trials or subscriptions. On iOS and Meta, run IG / FB ads, with the creative being the Jiaobei toss itself. What FateTell users remember is "IG push: toss the Jiaobei, open a blind box", and some downloaded it just to "go buy a scratch card." Use the search terms people actually type as store titles: 擲筊, 求籤, 聖杯.

When to start before: the 2027 Lunar New Year falls on Feb 6, when temple visits, lot drawing and fortune-asking peak all year (media runs the "young people draw lots on their phones" story every year). Get it live before late December, start small-budget creative tests in January, and add the working creatives before New Year's Eve. The 1st and 15th of each lunar month come every month, so they can serve as regular mini-peaks to test against.

What would mean you got it wrong: ① By late December, "杯筊天機" drops off the Taiwan lifestyle chart and its cumulative Play installs are still below a few thousand. That means this round of paid acquisition didn't pay back and was stopped, so "heavy ritual" doesn't necessarily convert to payment. ② The reverse: if it gets into the Taiwan or Hong Kong Top Grossing chart, the offerings-plus-membership model makes money, so move fast. ③ If one or two of the batch launched in the second half of 2026 have rating counts climbing past a thousand, the niche is already saturated and only the change-region-or-deity path is left. ④ If FateTell falls off the Top Grossing chart, even the validated one is on the way down, so don't build.

Android store page ↗

Research notes: competitor table, acquisition evidence, full research report
Competitors
NameWhereLaunchedScaleHow it differs from this one
FateTell: Chinese Divination Blocks & Daily Horoscope (AURAAI LIMITED)iOS2025-10-311,063 Taiwan ratings and 315 Hong Kong; best Taiwan category Top Grossing rank 40, Hong Kong 36, Malaysia 21 (appstore library)Jiaobei plus AI daily fortune; validated that it can earn money on iOS. No altar, no merit economy, no chanting, no ten-god lot library
Holy Cup Jiaobei - AI Divination Fortune Telling (Wuhan Hei Bo Cai)iOS2024-04-211,511 Taiwan ratings averaging 4.8 stars, 1,821 China store ratings (iTunes lookup 11-09)Lot drawing plus AI emotional companionship / role-play, following the chat-companion route; also a mainland China team targeting the Taiwan market
武財神 / 迎財神 (CMoney)Android武財神 is the earlier one; 迎財神 was still being updated in 2024武財神 Play 50k+ downloads, 5,130 reviews; 迎財神 10k+, 991 reviewsA major Taiwanese local player focused on the God of Wealth, incense-lamp lighting and lottery blessings, with ads and in-app purchases. It has proven that the online lamp-lighting incense-money model works
3D擲筊 (BiBiJiJi Studio)AndroidUnclear; 2026-07 still being updatedPlay 100k+ downloads, 1,370 reviews, 4.8 starsOnly a 3D physical Jiaobei tool, with no lot library, altar, AI or payment system
Donggang Haigong Temple - Online Oracle LotsAndroid2019-10Play 72,846 installs (googleplay library); 322 iOS ratings; included in Marie Claire Taiwan's 2026 Threads app recommendationsAn official app from a single temple, with faith-based endorsement, free, and not relying on AI or in-app purchases
溫府千歲 - 擲筊、求籤 (e68Club)AndroidAround 2011Play 50k+ downloads, 539 reviewsOne temple, one deity, free
心誠則靈 (Utoappia)AndroidUnknownPlay 10k+ downloadsOnline worship with ads plus in-app purchases; few features
觀運: 3D real Jiaobei, lot drawing, BaZi / Zi Wei / tarot (HoolyHi)Android2026Play 50+ downloadsNearly identical feature combination, but not yet measured
Web-based Jiaobei (piliapp, temples' own websites offering online tossing)WebAlready aroundUnknownFree and one-off, with no retention or monetization design
Where users come from
  • Confirmed Paid acquisition (Android, presumed to be Google App Campaigns): the Android privacy policy (2026-10) states that it reads the Google advertising ID and sends install and conversion events (joining, first lot draw, viewing the membership page, starting a trial, completing a subscription, one-time purchase of 称骨, donations) to Adjust, for "measuring ads from our own campaigns." It had zero chart appearances in its first two-plus months after launch; after 11-04 it went live on 1.0.6 and charted in all 11-05 four Chinese-language regions at once. Our Meta / Google / TikTok ad libraries don't track it, so there is no direct evidence of channels or budget.
  • Verified store search: on Play Taiwan, searching '擲筊' (throwing blocks) ranks #11, ahead of 3D Throwing Blocks, CMoney Wucai God / Welcome Wealth God, etc. (11-09)
  • Inferred Exposure from small-category charts: Taiwan lifestyle free chart ranks 40–50 only pick up a few hundred installs a week (neighbour "旅電" +7 over 405 days, "夢時代" +461), and charting there gives a little browse-the-charts traffic.
  • Inferred TikTok / X / Threads / media: TikHub searches for "杯筊天機" and "擲筊app", x-radar searches over 60 days, and quoted Brave searches all returned no content about it.

Why 杯筊天機 (com.beijiaotianji.app) showed up on the charts

Bottom line first

It hasn't really broken out yet. It's a traditional-Chinese "online temple" app from a small Shanghai studio (Shanghai Xiangxiang Network Technology Co., Ltd.): Jiaobei tossing, shaking the lot canister, a library of ten deities' lot poems, nine-slot altar incense offerings, chanting, vows and thanksgiving, plus BaZi charts and AI lot interpretation, earning money through subscriptions and in-app purchases ($0.99–$99.99). It launched on Play in Aug 29, first charted in Nov 5, and on the same day appeared on the lifestyle free charts of Taiwan, Hong Kong, Singapore and Malaysia. Its own privacy policy says it integrates Adjust to "measure ads from our own campaigns." So this wave of growth is most likely a small-budget paid test aimed at the Chinese-language folk belief market, and the Taiwan lifestyle free chart has a low bar. The card's "11/8 single-day jump from 48 to 605" is most likely a delayed Play install count caught up in one go, not a real single-day surge. The product is far more complete than its peers, but that isn't why it is growing. There is already a rival in this niche that makes money on iOS (FateTell reached #40 on the Taiwan lifestyle Top Grossing chart), so the demand is real.

1. What it is

  • Developer: Shanghai Xiangxiang Network Technology Co., Ltd. The contact details registered on Play are a personal gmail (pku249126752@…), a registered address in Baoshan District, Shanghai, and a +86 phone number (from the Play store page's "Developer info"), which points to a small team of one or two people. The library lists only this one app for this developer (googleplay library, developer).
  • Features (from the Play store description and the official site beijiao-site.pages.dev):
    • Ten deities: Guanyin, Guan Di, Mazu, the God of Wealth, the Buddha, the Matchmaker, Green Tara, the Yellow God of Wealth, Prince Nezha, and the Monkey King, each with a lot library with background stories and plain-language explanations, in both simplified and traditional Chinese;
    • 3D physical Jiaobei; charge the lot canister and a lot flies out; after drawing a lot, toss again to ask "is this the right lot?";
    • Nine-slot altar: three incense sticks, candles, flowers, oil lamps and fruit baskets, each with its own duration and multiplier; accumulate "incense merit" while offering, and tapping the wooden fish also adds merit; merit can be exchanged for new offerings, and every transaction is viewable;
    • Chanting hall: Heart Sutra, Great Compassion Mantra, Heavenly Holy Mother Sutra, Taishang Qingjing Jing, with audio tracks and timing, dedicated on completion;
    • Prayers and vow fulfilment, lot-book collection;
    • BaZi charts accurate to the minute (true solar time, late Zi hour); in-depth AI lot interpretation with multi-round follow-up questions; "Holy Blocks Revealed · Daily Fortune";
    • AI image recognition for offerings (photograph real offerings; subscription feature).
  • Tech and cost: AI runs on Volcano Engine Ark (ByteDance's large-model platform), with the backend on Cloudflare Workers. There is no account system; most data is stored locally, and cloud backup goes to the user's own Google Drive / iCloud (Android privacy policy beijiao-site.pages.dev/android-privacy).
  • Monetization: auto-renewing subscriptions (the membership tier alone unlocks multiple profiles, prayer records, cloud backup of lot draws, and image-recognition offerings), a one-time "称骨" purchase, and "随喜" (similar to incense money / donations); in-app purchase range $0.99–$99.99 (googleplay library app_snap). No ads on Play (ad_supported=False).
  • iOS: the official site has an iOS privacy policy (simplified Chinese, updated 2026-07-24), suggesting an iOS version exists or is in progress. But searching "杯筊天機 / 杯筊天机" in the Taiwan, Hong Kong, China and US iTunes stores finds nothing, and searching by name and Xiangxiang in our appstore library's app_name field also turns up nothing, so whether an iOS version is listed has not been verified.

II. Growth data

Source: googleplay library app_snap / chart_rank, 2026-11-05 to 11-10.

DateTotal installsTierTaiwan lifestyle freeHong KongSingaporeMalaysiaVersion
11-054110+5667691781.0.6 (released 11-04)
11-064810+5059691681.0.7
11-074810+5664109—1.0.8
11-08605500+4447180—1.0.8
11-09642500+4445187—1.0.8
11-10817500+4561179—1.0.8

My take on these numbers:

  1. First chart appearance was 11-05 (app_profile.first_chart_date). We have been tracking the Taiwan lifestyle free chart since 10-29, and from 10-29 to 11-04 it wasn't on it. In other words, for over two months after launch there was no movement, and in early November it charted in four regions at once.
  2. The +557 on 11-08 is most likely a backfilled count, not a single-day surge. From 11-05 to 11-07 it was already ranking around #50 on the Taiwan lifestyle chart, while the neighbours at #40–50 on the Taiwan lifestyle free chart each added only a few hundred a week (on 11-10: #48 "旅電 PLUG-INN" +7 over 405 days, #38 "夢時代" +461, #50 "小掌櫃" +1,368). A cumulative count of 41–48 could not hold that rank, which suggests Play's install counter was lagging on those days. The jump from 11-08's 10+ tier to 500+ while the rank only moved from 56 to 44 also fits a backfilled count with no real change in pace.
  3. A rough estimate of the actual pace is a few dozen to over a hundred installs a day, mostly from Taiwan and Hong Kong. By Play's numbers, from 11-05 to 11-10 it gained +776 in total.
  4. The store page shows 4.8 stars and 23 reviews (Play Taiwan page, scraped 2026-11-09), with the two most recent reviews from 11-07 saying "還行" and "讚!!!". The review count is roughly 3% of installs, which is high and may reflect an in-app rating prompt; not verified.
  5. Not on any region's Top Grossing chart (best_grossing_rank is empty).

III. Competitors

There are many online "Jiaobei / lot-drawing" apps in Chinese folk belief, but most are single-temple apps from the temples themselves, or small tools that only toss. The ones below are the closest to it:

ProductPlatformLaunchedScaleHow it differs from Bei Jiao Tian Ji
FateTell: Chinese Divination Blocks & Daily Horoscope (AURAAI LIMITED)iOS2025-10-31Taiwan 1,063 ratings, Hong Kong 315; Taiwan's best Top Grossing rank is #40, Hong Kong #36, Malaysia #21 (appstore library app_profile)Closest in direction: Jiaobei divination + AI daily fortune, and it has already been proven to earn money on iOS. It has no altar / merit economy.
Holy Cup Jiaobei - AI Divination Fortune Telling (Wuhan Hei Bo Cai)iOS2024-04-21Taiwan 1,511 ratings at 4.8 stars, China store 1,821 ratings, Taiwan Entertainment chart 115 (iTunes lookup 11-09)Lots draws plus "AI emotional companionship / role-play," taking the AI companion route. Also a mainland team targeting the Taiwan market.
武財神 / 迎財神 (CMoney)Android + iOS武財神 is the earlier one; 迎財神 was still being updated in 2024Wu Cai Shen: 50k+ Play downloads, 5,130 reviews at 4.9 stars. Ying Cai Shen: 10k+ downloads, 991 reviews (Play page).A major Taiwanese company (CMoney is a finance media brand): it focuses on the God of Wealth, with lamp lighting, lot draws, and lottery blessings, monetized through ads and in-app purchases. It has already shown that "online lamp lighting" as a incense-money model works.
3D擲筊 (BiBiJiJi Studio)AndroidUnclear; 2026-07 still being updatedPlay 100k+ downloads, 1,370 reviews at 4.8 starsOnly does 3-day physical Jiaobei tossing as a tool, with no lot library, altar, or AI.
溫府千歲 - 擲筊、求籤 (e68Club)Android / iOSFrom ¥2011Play 50k+ downloads, 539 reviewsOne temple, one deity, free
Donggang Haigong Temple - Online Oracle LotsAndroid / iOS2019Play 72,846 installs (googleplay library); 322 iOS ratings; Marie Claire Taiwan included it in its 2026 "Threads Recommended Apps" listReleased by a temple, backed by faith, free, and not reliant on AI or in-app purchases.
心誠則靈 (Utoappia)AndroidUnknownPlay 10k+ downloadsOnline worship with ads plus in-app purchases.
觀運: 3D real Jiaobei, lot drawing, BaZi / Zi Wei / tarot (HoolyHi)Android2026Play 50+The feature combination is nearly identical (3-day Jiaobei + lot draws + BaZi / Zi Wei / tarot), but it hasn't gained traction yet.
A batch that emerged in the past 2026 years.iOS2026-07 to NovemberXin Nian – Incense Jiaobei & Lot Drawing (09-15), Wen Bei · Jiaobei (10-01), Ling Guang Jiaobei (11-06), Jiaobei · 3D Jiao Bei (07-25, by Shanghai Wangyue Huishi), Kaucim BaZi Lots (08-25), Bai Gong Ma Digital Ancestral Tablet (08-23), BuaBuei Holy Blocks (07-14). Ratings range from 0 to 6.They entered the market around the same time; none copied it. It shows that indie developers from both mainland China and Taiwan are crowding into this direction.
Web toolsWebAlready aroundtw.piliapp.com Jiaobei, and online Jiaobei on temple websites (e.g., Liuxing Temple).Free, one-time use, with no retention design.
big techBig-company features—Found no built-in Jiaobei feature in Google, LINE, or WeChat.—

4. Growth levers (evidence)

  1. Paid acquisition: Strong evidence that it runs ads, but no direct evidence of channels or budget.
    • The Android privacy policy (updated October 2026) states: "To measure the effectiveness of our own ads, we read the Google Advertising ID and send install and specified conversion events to Adjust." The conversion events sent back are: completing a bond, first lot draw, viewing the membership page, starting a trial, completing a subscription, one-time purchase of Chenggu fortune reading, and Suixi tipping. This setup is designed for event-based bidding campaigns such as Google App Campaigns, with trial and subscription as the optimization targets.
    • The iOS privacy policy (2026-07-24), by contrast, states "no cross-app tracking" and "no collection of advertising identifiers," so paid acquisition is only enabled on Android.
    • The timing lines up: zero charting for over two months after launch, then 11-04 released 1.0.6,11-05, charting in four Chinese-language storefronts at once, with two more versions shipped within two days.
    • Our Meta / Google / TikTok ad libraries do not track it (not found ≠ not advertising).
  2. Content and creators: none found. A TikHub search for "杯筊天機" and "擲筊app" did not return it. x-radar search for "杯筊天機", "擲筊", and "beijiaotianji" gave 60 results over 0 days. A Brave search with quotes for "杯筊天機" returned 0 results. Exa only found the official site and third-party store data pages.
  3. Store search: a little, but not a main channel. Searching "擲筊" in the Play Taiwan store, it ranks #11, behind 3D擲筊, 聖筊, CMoney Wu Cai Shen / Ying Cai Shen, and others (Play search page 11-09).
  4. Low threshold on the niche chart: this acts as an amplifier. Taiwan's Lifestyle free chart rank 45, requiring only a few hundred installs per week, makes it visible to some users browsing the chart.

V. Why now, why users come, why they stay or pay

  • The demand has always been there: Chinese communities in Taiwan, Hong Kong, Malaysia, and Singapore have the habit of daily worship, Jiaobei tossing, and lot drawing. Media keep reporting that "young people use their phones to draw lots" (Sanlih 2024 Lunar New Year report; SETN covering the Seventh Prince lot App; Marie Claire listing the Zhenhai Temple lot App in 2026 Threads Recommended Apps).
  • Someone has already proven people will pay: FateTell had 1,063 Taiwan ratings in a year and reached Taiwan's Lifestyle Top Grossing chart at #40. CMoney reached 50k+ downloads with "light a lamp to add fortune." Incense money, lamp lighting, and offerings are already what believers willingly spend on. Moving them online turns them into consumable in-app purchases.
  • Its retention hooks are stronger than its peers: offerings are time-limited and have multipliers, with a push alert "offerings burned out" when they expire. There are incense reminders on the 1st and 15th of the lunar month, daily auspicious/inauspicious guidance, and streak counters for practice and lot collections. These are gamified reasons to return. Whether they actually raise retention and payments is not known from the data.
  • AI costs are low: domestic models via Volcano Ark handle lot interpretation and BaZi readings, and each call is cheap, which suits a subscription model.

VI. Differentiation

  • Compared with single-temple apps (Zhenhai Temple, Wenfu Qiansui): it isn't tied to one temple. It covers ten deities together, organized by purpose (marriage and relationships with the Matchmaker God, career with Guan Gong, wealth with the God of Wealth), serving a broader range of needs.
  • Compared with tool-style apps (3D Jiaobei, web-based Jiaobei): it has a complete ritual flow (shaking lots → Jiaobei confirmation → interpretation), plus a long-term economy of altars and merit.
  • Compared with AI fortune-telling apps (FateTell, Holy Cup AI): BaZi is calculated down to the minute, and lot interpretations combined with the individual's chart allow multi-round follow-up questions. There is also "AI photo offerings": you photograph a physical offering, and the AI suggests an elegant name and generates a blessing. I haven't seen this in the category.
  • Unusually clean privacy: no accounts, data stored on the device. This is a plus for faith-based users, but it's hard to use as a selling point.
  • Are these differentiators why it gained traction? No. The traction timing matches when attribution-ready paid acquisition was set up, with no trace of organic spread. The differentiators likely affect retention and payment, which determines whether paid acquisition pays back.

7. Visibility and copycats

  • Few people know about it yet. No media coverage, 0 discussion on X, TikTok, or Threads, no public revenue, and it isn't on the Top Grossing chart. Only a third-party store data site (chrome-stats) indexes its page.
  • Copycats: none. In 2026 seven or eight new Jiaobei / lot-drawing apps launched at the same time. They entered the market together rather than copying it.

8. What's worth borrowing

  1. Market choice: The Chinese-speaking niche markets (Taiwan, Hong Kong, Singapore, Malaysia) have a low barrier on lifestyle charts, and user willingness to pay has been proven by FateTell and CMoney. A mainland team built it in with Traditional Chinese and local deity systems (Mazu, Third Prince). Launching it makes it a global storefront.
  2. Monetization design: split "incense money" into three layers: time-limited offerings (consumable), membership subscription (cloud backup, multiple profiles, photo offerings), and one-time purchases (Chenggu fortune reading), plus a "Suixi" tipping entry.
  3. Paid acquisition setup: integrate Adjust only on Android, send trial and subscription events back, and bid on Play events. No tracking on iOS, which avoids the ATT prompt.
  4. Cost structure: Cloudflare Workers plus a domestic model API, with no account system. One person can run it.
  5. Directions that can be copied overseas: online offerings with consumable items, such as Vietnamese Mother Goddess worship, Thai Buddha amulets, Indian puja, and Japanese omamori / shrine visits.

9. What I'm still unsure about

  • Paid acquisition channel (presumed to be Google App Campaigns), budget, and CPI have no direct evidence;
  • Whether the 11-08 557 installs are backfilled counts or true single-day peaks can only be inferred from rankings;
  • Subscription prices, trial conversion rates, and retention are unknown; whether the 23 reviews at 4.8 stars include rating prompts is unknown;
  • Whether an iOS version exists and what it is called;
  • Actual revenue: it has not entered the Top Grossing chart and has no public figures.

Sources

This article was written by AI from public data and the data we collect daily. Numbers are sourced, and inferences are labeled.