
The video editing app never took off, so 12 days later it pivoted to AI fortune-telling: Bansi, from the Writesonic founder, squeezed 12k installs into India's Top Grossing chart
What it is An Indian AI astrology app: enter your birth details, then ask about marriage, career, or money in Hindi, Hinglish, or English. It computes the birth chart with a rules engine first, then has AI explain it in plain language with the reasoning attached. You can type or call, with typed questions at ₹25 each and voice at ₹25 a minute.
Why it's taking off A serial founder who knows growth moved over the team and domain of a failed project. He copied the pay-per-question, top-up-your-wallet model Indians already know, started buying Google and Meta ads in late October, then added voice and a batch of SEO calculator pages. Daily installs went from 1 to about 760.
BHow you could follow it Don't build the 41th Hindi AI astrology app. Build the Tamil or Telugu version it doesn't have, or target overseas Indians. Copy the pricing as it is: two free questions, double first top-up, per-question charges. Optimize video ads on Meta and Google for paid events.
Est. monthly revenue (Google Play) under $1KWe estimated this from its rank on country top-grossing charts and real install counts, calibrated against real revenue; typical error is 2–3x. Read it as a tier, not an exact number.
- Data (updated daily) · first launch 2026-08-31 · 11-15 installs 13,259
Android · Total installs worldwide Data 2026-11-04 → 2026-11-15 (12 days, 8 observations). Google's detail page only refreshes install counts every other day, so a flat day doesn't mean nobody installed.
Bansi is an Indian AI Vedic astrology app. It first calculates your birth chart using classical rules, then has AI explain it to you in Hindi, Hinglish, or English. You can type your question or call. The person behind it is the founder of Writesonic, whose previous product shut down 12 days before this app launched. For its first 7 weeks almost no one installed it. In late October it started buying ads, added voice, and more than doubled its prices. By Nov 12 it had 12k cumulative installs. Over the last 7 days, installs rose 5,912, and it entered India's Lifestyle top-grossing chart. Few people know about it yet. Below: how it charges, where users come from, why its chart rank should be discounted, and what gaps remain in this already crowded category.
Aug 19, a shutdown post
On Aug 19, 2026, Samanyou Garg, founder of Writesonic (YC S21), announced on X that he was shutting down bansi, his AI long-form video editor. The product had launched in Apr 24, named after Krishna's flute. In the post he wrote: "Four months in, no product-market fit, and it cost more to run than it made." He took the blame: he had hired and built back-end infrastructure and aimed big before a weekend project had proven anyone wanted it. His new rule: "The next MVP gets the best usable version into one user's hands within a week or two."
12 days later, in Aug 31, "Bansi: AI Astrology & Kundli" appeared on Google Play, using the bansi.ai domain. The video editor bansi never took off, and an AI astrologer with the same name took its place.
On Oct 22, a five-star review appeared in the app's review section: "Amazing app for getting accurate astrology readings based on our traditional texts." It was signed by Samanyou Garg, and 29 people marked it helpful. By Nov 11 it was still among the top three reviews on the Play page.
He had already built Indian spiritual and devotional products: BhagavadGita.io in 2018, and an app rebuilt with an outsourced team in 2023. According to the outsourcing firm in the FlutterFlow community, that app passed 200k downloads.
What it sells: ₹25 per question, calculated first
The first line of the Play listing is "Call Bansi." Users enter their birth date, time, and place, then ask things like "Meri shaadi kab hogi?" (When will I get married?) or "Is it a good time to change jobs?" Marriage, career, money, health, moving abroad, buying property, and choosing auspicious dates are all covered. You can chat by text or hold a real-time voice call with the AI astrologer; on the website, the persona is called Vaani.
The pitch it repeats is "calculate first, then explain." The website says Bansi's own astrology engine does the interpretation, checking "thousands of rules, planetary combinations, exceptions and cancellations" under both the Parashari and KP systems. The AI only turns the result into plain language. Each answer cites the relevant planets, houses, and dasha periods, plus a classical source. The website example cites KP Reader IV, pages 15–17. When birth time is uncertain or the chart contradicts itself, it says so. The site also includes a comparison table of human astrologers, general-purpose AI, and Bansi, and the wording is restrained.
The pricing borrows from the human-astrologer platforms Indians already know, with one change:
- The first 2 questions are free. After that, each text question costs ₹25 (about $0.3), and follow-ups count as one question with no timer. The website's phrase is "never a running meter" (no meter running).
- Voice costs ₹25/minute. The price is shown before the call, and the account needs at least 3 minutes of balance.
- The first top-up is doubled: pay ₹100 and get ₹200 credited. There's no subscription.
- Separately sold: Kundli Milan (compatibility matching; one 11-03 review says "₹50 for milan is great value") and a 17-chapter full birth-chart PDF. In-app purchases on Play range from $0.19 to $39.99.
Prices have changed recently. Old pages still online say "first 3 questions free, then ₹10/question, voice ₹15/minute," and a review from Nov 3 still says "first 3 questions free." The current pricing is 2 free questions, ₹25/question, and ₹25/minute, which means unit prices rose to 1.7 to 2.5 times the original.
What a ₹25 question costs to serve: one answer uses a Claude-tier model, with a few thousand input tokens and a few hundred output tokens, roughly ₹2–3. Google Play takes 15%, about ₹4 (estimate). Text Q&A has thick margins. Voice is billed by the minute, and real-time voice models cost roughly ¥0.1 to ¥1 per minute, so ₹25/minute leaves room (estimate). The privacy policy says the language part uses Anthropic's Claude and also integrates OpenAI. It doesn't say which provider handles voice.
7 installs in the first 40 weeks, then 12k in one month
| Date | Total installs | Daily new installs at that time were roughly |
|---|---|---|
| 08-31 | Launched | — |
| 10-15 | 40 | 1 |
| 10-22 | 354 | 45 |
| 10-29 | 907 | 80 |
| 11-04 | 5,184 | First appeared on India's Lifestyle free chart, ranked #198 |
| 11-05 | 6,241 | 760 |
| 11-08 | 8,707 | Top Free #178 |
| 11-12 | 12,153 | 7 days +5,912; India Lifestyle Top Grossing #81 |
Source: Google Play install counts. Daily installs before 11-04 are estimates from the third-party tool AppstoreSpy.
The curve comes in two steps: around Oct 19 daily installs jumped from 1 to 45, and around Nov 2 they jumped again from 80 to 760. That pattern fits testing with a small budget and then scaling spend once it works. From Nov 3 to 9 it shipped 4 versions (0.4.2 to 0.4.7) and is still iterating fast. On Play it has 4.0 rating with 60 reviews (11-11). The iOS version says "coming soon."
Where users come from: ads, calculator pages, and self-written reviews
Google and Meta ads, confirmed. The privacy policy says: "If you installed Bansi from our ads on Google, Facebook or Instagram…" On Android it sends "first reading" and "payment amount" events back to Google and Meta, so optimization targets paying users rather than installs. The Facebook page "Bansi Astro" describes itself as "Astrology you can check. Vaani, our AI astrologer… First reading free 👇." Someone has seen a Hinglish ad creative that says "Dadi, type mat karo. Bolo." (Grandma, don't type, just speak), aimed at older relatives who don't type. We couldn't find the original post again, so this is unverified. How much was spent isn't visible from outside.
A batch of SEO pages, live in 10-30 day. Five free calculators: compatibility matching, Manglik (Mars dosha), Sade Sati (Saturn's seven-and-a-half-year transit), Darakaraka, and Navamsa. Plus three comparison articles: "Best Astrology App in India," "Best AI Astrology App," and "AI vs Human Astrologer," which name-check Astrotalk, InstaAstro, Astroyogi, AstroSage, Melooha, and Reva. This is Writesonic's home turf, the playbook of someone who works in AI search and SEO marketing. The second step up came right after these pages went live, but it's impossible to separate how much came from ads versus SEO.
Praise from the founder and his circle. Besides Samanyou Garg himself, a five-star review from "manoj garg" in Nov 4 says it "saved me the time, energy, and big money I used to spend chasing human astrologers." As of Oct 30 the app had only 10 ratings in total, so some of the early scores came from insiders.
The absences are worth noting too: no funding news, no press coverage, and the founder hasn't publicly promoted the app on X or LinkedIn (he posted build logs daily for the video version). It doesn't show up in TikTok or Reddit searches. This time he's building quietly.
Top Grossing #81, and last place on the chart
12k installs was enough to chart on Top Grossing, which sounds like strong conversion. Read it with caution, though: India's Google Play Lifestyle Top Grossing chart only ever lists 76 to 86 apps a day. On Nov 12 it held exactly 81 apps, and Bansi was last. The same day, another astrology app, KRSvedic, had only 8,192 installs and ranked #52. Drishti, which also pitches "accurate calculation plus AI interpretation," had 24k installs and ranked #78.
So the rank shows that people are paying, not that it's earning a lot. Also keep in mind that human-astrologer platforms like Astrotalk mostly use their own payment channels, so they never show up on Play's grossing charts. Don't compare rank to revenue. Outside observers can't see how much it earns per month or whether its ad spend pays back.
Who's competing
India's AI astrology is already a crowded niche. Searching Google Play for apps with astro or kundli in the name, over 40 have added 1k+ installs in the last 7 days, and 30k is the bar they're clearing.
| Competitors | Scale (Google Play, 11-12) | How it differs from Bansi |
|---|---|---|
| Astrotalk | 115M installs, +400k in 7 days | Human astrologers charge per minute; Bansi's pricing was set against them |
| AstroSage Kundli: AI Astrology | 82.4M, Lifestyle Top Grossing #48 | An established chart-calculation tool that launched an AI phone astrologer in 2025 October. Voice wasn't Bansi's invention |
| Reva (Mohalla Tech, parent of ShareChat) | Launched Sep 8, 8.17M, +520k over 7 days | Big-company resources, heavy spending; Bansi's comparison page files it under 'general LLM astrology' |
| Astro99 / Tarot99 | 16.8M / 10.4M | Large-scale paid acquisition, social-driven |
| AstroVaani | 183k, +57k in 7 days | Also lets you chat with an AI astrologer in Hindi and Hinglish, and growing much faster than Bansi |
| Aura AI, Drishti | 660k / 24k | Also says it works from your birth chart and dashas; you can chat or call. Drishti is the closest in scale |
| MyNaksh, Melooha, AskPandit | 2.05M / 1.46M / 3.16M | Earlier wave of AI astrology |
| ChatGPT, Claude asked directly | huge | Free, but the planetary positions may be recalled from memory; TikTok has tutorials on 'make Claude your Vedic astrologer' with one at 18k views |
What actually makes it different
Bansi's own stated differences are three: a rules engine calculates first and the AI only explains, so the same question gets the same conclusion; every answer can be expanded to show its basis, calculation steps and scripture sources; text is charged per question at a fixed price, with no tiers and no pushing of rituals.
The pitch is clear and restrained, and it holds its ground on the comparison page. But users can hardly actually perceive it: ordinary people can't tell whether an answer was computed by rules or made up by a model, Drishti and Aura AI also say they 'compute the chart first', and Bansi's own comparison page concedes that 'other AI apps also claim to compute your chart'. What's driving growth is plainer: it buys ads, offers two free questions plus double-on-first-top-up, and gives people who don't want to type a way to just talk. The differentiation is what holds paid conversion afterward and captures searches like 'which astrology app is good'.
Where it also falls short: the app only supports Hindi, Hinglish and English. Users in South India who speak Tamil or Telugu follow different chart conventions and matchmaking customs. The leading apps for those languages are calendar and almanac apps such as Nithra Tamil Calendar (17M installs) and Telugu Panchangam (3.77M). Among the AI astrology apps rising fast on Google Play, none is focused on Tamil or Telugu (11-12).
If it were you
B. The approach of 'AI calculates first, then explains, charge per question, and take phone calls' has already shown that people in India will pay for it, and Bansi itself has barely drawn attention. But the category is crowded, the top players have big-company resources and heavy spending, and Bansi itself grew by pairing an established team with paid acquisition. Copying it straight into an Indian version means you'd land behind more than 40 similar apps.
- Who it suits: small teams that can buy ads on Meta or Google optimized to paid events and can afford to burn a few thousand dollars testing creative; people who understand a specific Indian regional language and its astrology habits (or have such a partner); people who have built 'pay per use' service apps and know how to design a first purchase.
- Who it doesn't suit: anyone who expects to start from organic app store traffic or SEO (almost all traffic in this category is bought); anyone uneasy with astrology content who can't take a user's 'when will I get married' question seriously; anyone who wants subscriptions (Indian users are used to topping up a wallet and paying per use).
- How to do it: don't build the 41th Hindi AI astrology app. Pick a piece Bansi hasn't covered: South Indian languages like Tamil and Telugu, with birth charts drawn in local styles, marriage matching following local customs, and auspicious days and hours from the almanac turned into a reason to open the app every day. Or target the Indian diaspora abroad, pricing in English with USD or AED, where the same question could be priced several times higher (inference). Copy Bansi's monetization: two questions free, a double-up on the first top-up, per-question pricing that doesn't jump in tiers, and voice billed by the minute. Attach a "why" to every answer; that's the line to lead with when comparing against human consultation platforms and general-purpose AI.
- How to acquire users: video ads on Meta and Google, sending purchase events back and optimizing for paid users, not installs. Creative in the spirit of 'Grandma, stop typing, just say it': speak in their own language to older users who don't type. At launch, also roll out a batch of free calculators (matchmaking, Mars dosha, Saturn's seven-and-a-half years) and 'best astrology app' comparison pages to capture search traffic.
- When to move: India's wedding season runs from November to February, when demand for matchmaking and auspicious-date checks peaks (inferred). To catch this wave, you need a paid-capable version and first creative ready by mid-December. Any later and the big companies may add regional languages too.
- What would show you're wrong: ① if leaders like Reva or AstroSage add AI voice astrology in Tamil or Telugu within a month or two, the empty space is gone; ② if Bansi's daily installs fall back below 100 in the next month or two and it drops off the grossing chart, that means its post-price-hike ad spend isn't paying back, and this pricing model can't sustain paid acquisition for a small team; ③ conversely, if Bansi launches on iOS and starts charting in more countries, the overseas-Indian route is wider than it looks, and you need to move before it does.
Data and sources
- Installs, charts, version updates: Google Play, 10-29 to 11-12; daily installs before 11-04 are AppstoreSpy estimates.
- Store listing copy, reviews, developer info: Google Play India storefront page, 11-11.
- Product, pricing, comparison table, privacy policy: bansi.ai homepage, comparison page, privacy page, November; old prices come from search-engine cache of earlier pages.
- Founder: Samanyou Garg's X post (2026-08-19), Product Hunt, LinkedIn (2023), FlutterFlow community post.
- AstroSage AI phone astrologer: press release reposted by Business Standard (2025-10).
- Costs are estimated from public API prices. Where we write 'estimated' or 'inferred', it's an estimate or inference. We confirmed the ads are running but did not find the scale of spend.
Ads it's running
Google has shown 2 ads, none running now, earliest 2026 Oct 18.
Google · video · ran 25 days · stopped Nov 12 · 3 versions · view original ↗
Call Bansi, your AI astrologer who knows your kundli. Talk any time.
These are the longest-running ads in this set. Long runs usually mean the creative is working. Data comes from the official Meta, Google and TikTok ad libraries and updates daily; open the official library to view the original creative.
Research notes: competitor table, acquisition evidence, full research report
- Competitors
Name Where Launched Scale How it differs from this one Astrotalk Android 2017-10 Play 115M installs, +400k in 7 days Human astrologers charge per minute; Bansi uses AI, with text priced per question, not timed AstroSage Kundli: AI Astrology Android 2011-09 Play 82.4M, +200k over 7 days, India LIFESTYLE Top Grossing #48 Established chart-calculation tool; 2025-10 has launched AI phone astrology. Bansi's voice wasn't first; the difference is its emphasis on the rules engine and the basis for answers Reva - Personal AI Astrologer(Mohalla Tech / ShareChat) Android 2026-09-08 Play 8.17M; 7-day +520k A conversational AI astrologer with big-company resources and heavy spending. Bansi is two orders of magnitude smaller and leads with 'calculate first, then explain' Astro99 / Tarot99(Primetrace) Android 2025-02 Play 16.8M / 10.4M Large-scale paid-acquisition AI astrology. Bansi goes for high prices and fewer users AstroVaani(Goblins Studios LLP) Android Unknown (10-29 already 50k) Play 183k, +57k in 7 days Also 'talk to AI astrologer Vaani in Hindi / Hinglish', with multiple personas; growing faster than Bansi Aura AI: AI Astrology & Kundli Android 2026-05-25 660k on Play Also says it computes from your kundli and dashas, and you can chat or call. Bansi adds scripture-source display and per-question billing Drishti: AI Astrology & Kundli Android Unknown Play 24k, India LIFESTYLE Top Grossing #78 Also precise chart calculation with Swiss Ephemeris plus AI interpretation. Similar in scale; the closest rival MyNaksh / Melooha / AskPandit Android 2023–2025 Play 2.05M / 1.46M / 3.16M Earlier wave of AI astrology; Bansi's comparison page classes them as 'LLM interpretation' InstaAstro / Astroyogi / AstroLokal Android 2016–2021 Play 12.4M / 20.8M / 7.41M Human platform, first consultation ₹1 or free the first time ChatGPT / Claude asked directly Big-company features From ¥2024 huge Free, but planetary positions may be recalled from memory; Bansi's pitch is precisely 'calculate with ephemeris first' grahai.com / astroma.co / kundligpt.com Web 2025–2026 Not verified Free web AI astrology that also emphasizes showing its basis - Where users come from
- Confirmed Google / Meta paid acquisition: the privacy policy states users 'install Bansi from one of our own ads on Google, Facebook or Instagram', with first readings and purchase events sent back to Firebase / Meta; the Facebook Page 'Bansi Astro' (61594192376358) reads 'Vaani… First reading free'; creative seen during recall included 'Dadi, type mat karo. Bolo.'; daily installs rose in stepwise jumps from 1 to 45 to 760. Our ad library didn't track it, so scale is unknown
- Confirmed SEO content and free calculators: the sitemap lists 10-30 free calculators launched 5 (Kundli Milan, Manglik, Sade Sati, Darakaraka, Navamsa) and 3 comparison articles naming Astrotalk, InstaAstro, AstroSage, Melooha and Reva; their contribution can't be separated out
- Confirmed Founder / friends' positive reviews: on Play, Samanyou Garg himself left a five-star review 10-22 (29 people marked it helpful), along with a five-star from 'manoj garg'; at 10-30 the app had only 10 ratings in total
- Inferred Founder's personal influence (X / LinkedIn): he has publicly posted build logs for the video version of bansi, but x-radar, Exa and Brave found no posts of his promoting Bansi Astro
- Confirmed Free quota and double-on-first-top-up: per the website, the first 2 questions are free (earlier it was 3 questions); a first top-up of ₹100 credits ₹200
Bansi: AI Astrology & Kundli (ai.bansi.astro): how did it take off?
Verdict
Strictly speaking it hasn't taken off yet; it has only just started gaining traction. It is the pivot product launched 12 days after Writesonic founder Samanyou Garg announced the shutdown of bansi, his AI video-editing project at 2026-08-19, and it reuses the same domain bansi.ai and brand. The first 7 weeks after launch saw almost no installs (10-15 cumulative 40). Paid buys on Google and Meta began in late October, and in early November it added voice calls and raised the price 2.5×, with daily installs jumping from about 80 to about 760. At 12k cumulative installs it reached #81 on India's LIFESTYLE Top Grossing chart, but that chart holds only about 81 spots and this placement is the last one, so the bar is low.
The product pitches 'a proprietary engine computes the chart from classical rules first, then AI explains it in plain language, with the basis shown'. That's a common claim among India's AI astrology apps in 2026, not a unique one. AstroSage already offered voice calls in 2025-10. It took off because of this: a serial founder who understands growth and previously pushed a Bhagavad Gita app to 200k downloads turned 'AI astrology' into a pay-per-use wallet model, then added ads and a set of SEO pages. Few people know about it yet.
---
1. What it is
- An AI Vedic astrology app for the Indian market. Users ask questions in Hindi, Hinglish or English, either by text or in a live voice call with the AI astrologer 'Bansi' (the persona on the website and Facebook has been renamed 'Vaani'). Answers come with relevant planets, houses and dashas, plus cited scripture sources (Play store page, captured 2026-11-11).
- The website says: 'Bansi's proprietary astrology engine does the interpretation, and the AI only explains it', running 'thousands of rules, combinations, exceptions and cancellations' under the Parashari and KP systems (bansi.ai). The privacy policy says the language layer uses Anthropic's Claude model and also connects to OpenAI; payments go through Google Play, with RevenueCat managing the wallet (bansi.ai/privacy).
- Pricing: the first 2 questions are free; after that text costs ₹25 per question (follow-ups also count as one question, not timed), voice costs ₹25 per minute, and you need at least 3 minutes of balance to place a call; the first top-up is doubled (₹100 credits ₹200); there are no subscriptions. Separately sold are Kundli Milan (matchmaking, reviews mention ₹50) and a 17-chapter Brihat Kundli PDF. In-app purchases on Play range from $0.19–$39.99.
- Pricing has changed: the older page cached by search engines listed 'first 3 questions free, then ₹10 per question' and '₹15 per minute' (Exa and Brave snippets); now it's '2 questions free, ₹25 per question, ₹25 per minute'. Both increases happened after 10-30 (the comparison page published that day already showed ₹25).
- Operator: the website footer reads 'Bansi Astro is operated by Tridev, C-604, IRWO, Sector-47, Gurugram'. The Play developer name is Tridev Labs, but the account is a personal one under Priti Garg ([email protected], New Delhi address). Support email is [email protected].
2. Who's behind it: a pivot built on resources
- Samanyou Garg is the founder and CEO of Writesonic (YC S21). 2026-04-24 He launched bansi, an AI long-video editor (X: @bansi_ai, Product Hunt), 2026-08-19 then posted that he was shutting it down, citing "four months with no PMF, and it costs more to run than it earns," and set a new rule: "Next MVP goes into one user's hands within a week or two" (X post).
- 12 days later (2026-08-31) Bansi Astro went live on Play, still on the same domain bansi.ai. webclaw still lists bansi.ai as "Bansi AI by Writesonic" (webclaw).
- He had a base in spiritual products for Indian audiences: in 2018 he built BhagavadGita.io, and in 2023 he rebuilt it as an app with an outsourced team, which claimed 200k+ downloads (LinkedIn 2023, FlutterFlow community post).
- The third review on the Play page is a five-star from Samanyou Garg himself on 2026-10-22 ("Amazing app for getting accurate astrology readings based on our traditional texts," marked helpful by 29 people). Another five-star from "manoj garg" (5 stars, Play page, 11-11) also appears. This looks like the founder seeding reviews, so part of the early rating was propped up by insiders.
3. Growth data
| Date | Total installs | Notes | Source |
|---|---|---|---|
| 2026-08-31 | — | Launched | googleplay library app.released |
| 09-22 | 17 | 1–2 per day | AppstoreSpy estimate (app_day_ext) |
| 10-15 | 40 | Still almost zero installs 6 weeks after launch | Same as above |
| 10-22 | 354 | Daily installs jumped to about 45 | Same as above |
| 10-29 | 907 | About 80 daily installs | Same as above |
| 10-30 | — | The comparison page claims "500+ installs, 10 ratings" | bansi.ai/compare/best-astrology-apps-india |
| 11-04 | 5,184 | India LIFESTYLE Top Free, #198, first time charted | googleplay library |
| 11-05 | 6,241 | About 760 daily installs | AppstoreSpy estimate |
| 11-08 | 8,707 | Top Free #178 | googleplay library |
| 11-12 | 12,153 | +5,912 over 7 days; India LIFESTYLE Top Grossing, #81 | googleplay library |
- Between versions 0.4.2 and 0.4.7,11-03 and 11-09, it shipped 4 updates (googleplay library, app_day), still iterating fast.
- The Play page shows 4.0 stars and 60 reviews (11-11).
- Take #81 on Top Grossing with a grain of salt. Our India LIFESTYLE Top Grossing chart only has 76–86 entries a day, and on 11-12 there were 81 entries in total, with Bansi last. KRSvedic reached #8,192 on the same chart with only 52 installs, and Drishti reached #24 with 78k installs (googleplay library 11-12). So charting means someone is paying, not that it's earning much. Also, real-human astrology platforms like Astrotalk mostly use their own payment flows and aren't on the Play grossing chart, so don't compare these ranks to their revenue.
4. Competitors
| Product | Where | Launched | Scale (googleplay library 11-12) | How it differs from Bansi |
|---|---|---|---|---|
| Astrotalk | Android / iOS | 2017-10 | 115M installs, +400k in 7 days | Human astrologers charge by the minute; Bansi is AI, and its price is set against that |
| AstroSage Kundli: AI Astrology | Android | 2011-09 | 82.4M total, +7k over 200 days, India LIFESTYLE Top Grossing #48 | An old-school chart-casting tool, 2025-10 has launched the "world's first phone AI astrologer" (Business Standard); Bansi's voice feature is not the first of its kind |
| Reva - Personal AI Astrologer(Askk App,Mohalla Tech / ShareChat) | Android | 2026-09-08 | 8.17M total, +7k over 520 days, India LIFESTYLE Top Free #3 | AI astrology backed by a big company with heavy paid acquisition; Bansi's comparison page names it as "general LLM astrology" |
| Astro99 / Tarot99(Primetrace) | Android | 2025-02 | 16.8M / 10.4M | AI astrology with large-scale paid acquisition and a social angle |
| AstroVaani(Goblins Studios LLP) | Android | Unknown (10-29 already has 50k) | 183k, +57k in 7 days | Sells the same idea of "talk to an AI astrologer in Hindi and Hinglish." Its persona name "Vaani" clashes with Bansi's current "Vaani", and its site lists 15+ AI personas. It is growing much faster than Bansi. |
| Aura AI: AI Astrology & Kundli | Android | 2026-05-25 | 660k | Same pitch: "every answer is computed from your kundli and current dasha," with chat or calls. |
| MyNaksh / Melooha / AskPandit | Android | 2023–2025 | 2.05M / 1.46M / 3.16M | A longtime AI astrology player; Bansi's comparison page names Melooha |
| Drishti: AI Astrology & Kundli | Android | Unknown | 24k, India LIFESTYLE Top Grossing #78 | Pitches "precise Swiss Ephemeris charts + AI interpretation," at about Bansi's scale. It is the closest rival. |
| Kismo(ParakeetAI) | Android | 2026-08-26 | 92k | "Chat astrology" billed by the minute |
| InstaAstro / Astroyogi / AstroLokal | Android | 2016–2021 | 12.4M / 20.8M / 7.41M | A human-astrologer platform, first session ₹1 or a free first consultation |
| ChatGPT / Claude doing the chart calculation themselves | Big-company features | From ¥2024 | Huge. TikTok has tutorials on "making Claude your Vedic astrologer" (@badar.munirofficial, 18k views). | Free, but planetary positions may be made up from memory. Bansi's selling point is "calculate first, then interpret." |
| grahai.com / astroma.co / kundligpt.com | Web | 2025–2026 | Not verified | Free web-based AI astrology that also emphasizes "showing planetary and scriptural basis" |
Competitive read: the category is crowded. In our library alone, there are over forty Play apps with astro or kundli in the title that gained 7k+ installs in 30 days. Drishti, Aura AI, and Astroma all use the line "engine calculation + AI explanation + cited basis." Bansi's own comparison page concedes: "Other AI apps also say they calculate charts."
5. Where users come from
- Paid ads on Google and Meta (confirmed). The privacy policy states: "When you install Bansi from one of our own ads on Google, Facebook or Instagram…" On Android, first readings and purchase amounts are sent back to Google (Firebase) and Meta (purchase events via RevenueCat) to optimize ad delivery. The Facebook page Bansi Astro (id 61594192376358) has the bio "Astrology you can check. Vaani, our AI astrologer… First reading free 👇" (Brave search snippet). A Hinglish creative, "Dadi, type mat karo. Bolo." (speak to grandparents who can't type), also appeared, but we couldn't re-fetch it. Our metaads and googleads libraries don't track this Page or advertiser, so spend is unknown. Daily installs stepped up twice, in late October and early November (1→45→760), which fits a pattern of testing with a small budget and then scaling.
- SEO content site (confirmed to exist, contribution unknown). The sitemap went live on 10-30 with five free calculators (Kundli Milan, Manglik, Sade Sati, Darakaraka, Navamsa) and three comparison articles ("best astrology app / best AI astrology app / AI or a real astrologer") that name Astrotalk, InstaAstro, Astroyogi, Guruji, AstroSage, Melooha, and Reva. This is Writesonic's core playbook (AI search and SEO marketing). When searching "bansi.ai AI astrologer phone call," its page ranks above AstroSage's press release.
- Founder and friends' reviews (confirmed). Samanyou Garg himself and "manoj garg" left five-star reviews. At 10-30, there were only 10 ratings in total.
- Founder's personal reach (not found). He publicly shared build logs for the video version of bansi on X and LinkedIn, but we found no posts from him promoting Bansi Astro in the x-radar corpus, Exa, or Brave. No related content on TikTok or Reddit either. This launch was quiet.
- Conversion design: 2 first question free, first top-up doubled, text charged per question rather than per minute. It targets Astrotalk's pain point of per-minute billing with a balance that keeps running. Voice is billed per minute to match how human astrologers charge, which is why the average order value is higher.
6. Why it broke out
- Founder resources: as CEO of Writesonic, he already has an AI product engineering and growth team. After the video project was shut down, the domain, brand, and team carried over directly, and it went live within 12 days (X post 08-19, Play 08-31).
- Category timing: from 2026 through 9–October, AI astrology in India took off. Reva reached 8.17M installs in two months, and Astro99 / Tarot99 combined reached 27M. Users had already been trained to "ask an AI astrologer" (googleplay library).
- Monetization: no ads, no subscription. It copies the "wallet top-up + pay-per-question / per-minute" model Indian users already know, and raised the unit price from ₹10 to ₹25. 12k installs got it onto the grossing chart (last place, but it's there), which suggests paying intent holds up under paid acquisition.
- Voice entry point: the first line of the Play description is "Call Bansi." For middle-aged and older Indian users who don't type comfortably, "just speak" lowers the barrier (the "Dadi, type mat karo. Bolo." creative targets exactly this group). But AstroSage had AI phone astrology by 2025-10, so this isn't exclusive.
- Paid ads plus SEO for acquisition: the privacy policy confirms ads. The SEO pages that went live on 10-30 were followed by a second surge in installs. The timing lines up, but the two channels' contributions can't be separated.
7. Differentiation
Bansi leads with three claims:
- a deterministic engine calculates first, and the LLM only explains, so the same question gets the same conclusion;
- every answer comes with supporting and opposing evidence, the calculation steps, and scriptural sources (e.g., KP Reader IV, pages 15–17);
- fixed price per question, no timer, no upselling of remedies.
The official comparison table spells this out clearly and with restraint. It even says outright "astrology has no scientific basis" and "Bansi has things other apps do better."
But this differentiation is hard for users to perceive directly. Ordinary users can't tell whether an answer came from a rules engine or was made up by an LLM, and Drishti, Aura AI, and Astroma all say they "calculate the chart first." So differentiation isn't the main reason it's gaining traction. Growth comes mainly from ads, low-price trials, and the voice entry point. Differentiation mostly helps protect paid conversion and word of mouth, and claims a spot in SEO and comparison searches.
8. Visibility and copycats
- Awareness: still low. No funding news, no press coverage, and the founder hasn't publicly promoted Bansi Astro on X or LinkedIn. In the x-radar corpus, "bansi" returns only 1 unrelated results, and nothing on Reddit or TikTok. Revenue isn't public; the only signal is the last place on the Play grossing chart.
- Copycats: none clearly identified. AstroVaani's AI persona is also called Vaani and also speaks Hindi and Hinglish, but who used the name first can't be determined. AstroVaani already had 50k installs by 10-29, so it probably predates Bansi's persona name, or Bansi borrowed it.
9. What others can learn
- A failed project's domain, brand, and team can be moved quickly onto a validated category: live in 12 days, spend nothing for the first 6 weeks, and buy ads only after the product stabilizes.
- For AI services in India, make the billing feel like a human service (wallet, per-minute), then build a contrast point around "fixed price per text question, no timer." Doubled first top-up and 2 free questions drive the first purchase.
- Voice first, with creatives aimed at grandparents who can't type.
- Turn "we're not a ChatGPT wrapper" into verifiable claims: show the basis and sources, and be candid about limits.
- Build SEO assets at launch: free calculators and "best app" articles that compare competitors, to capture comparison searches.
- Send purchase events back to Meta and Google, and optimize on first reading and purchase rather than installs.
10. Still unclear
- Ad spend and creatives: neither the metaads nor the googleads library tracks it, and the Meta Ad Library page wasn't captured.
- Actual revenue and ROI: the last place on the grossing chart shows revenue exists but not how much. The effect of the ₹10 → ₹25 price increase on conversion is also unknown.
- How Tridev and Writesonic are related (a side project of Samanyou personally or his family, or an internal Writesonic project). The Play developer account uses the personal name Priti Garg.
- When the voice-call feature launched (likely early November, around the second surge), and which vendor powers the voice (the privacy policy doesn't say).
- User retention: too few reviews (60) and no review sample in the library.