The archive · Consumer Apps · Strategic decision · 2017–2026
Co-Star: astrology-as-social-app bet grows to 4.9M MAU, then sells to Midjourney
Co-Star bet blunt, design-forward astrology would spread by screenshots with zero ad spend; ~4.9M MAU came, flat revenue led to a 2026 Midjourney sale.
Co-Star
What the business is
Co-Star is a mobile astrology app launched in 2017: users enter birth time and place to get NASA planetary data plus AI- and astrologer-written birth charts, blunt daily horoscopes, compatibility reads with friends and paid add-on readings.
How it started
Co-Star was founded in New York in 2017 by Banu Guler, Anna Kopp and Ben Weitzman and launched on 2017-10-17, per Cyzone (创业邦). Guler, an NYU psychology graduate and fashion-industry product and design director, traces the idea to a hand-made, book-length birth-chart gift for a friend's baby that made her see astrology could leave the niche; the team translated astrological rules into computational models and layered NASA Jet Propulsion Laboratory planetary data over content written by AI plus professional astrologers.
What happened
The product grew with almost no paid marketing: blunt daily notifications and magazine-style screenshots spread on Instagram and Reddit, and Guler told Reddit that 20% of young Americans had downloaded the app. By the 2026 deal Co-Star had about 4.3-4.9M monthly active users and roughly 200K downloads a month, but revenue sat around $400-500K per month after store fees (BaijIng, citing DianDian), peaking only in December 2025 near $1M when an $11.99 'Your Year Ahead' product launched - because most money came from one-off purchases such as $2.99 chart reads and compatibility checks, while the Pro Star subscription stayed under 10% of revenue. Midjourney, in its first acquisition, bought Co-Star's parent in a deal that closed in spring 2026 and was announced on 2026-07-24 with terms undisclosed; Guler became Midjourney's chief design officer, the roughly 24-person team joined, and Co-Star keeps operating under her direction.
How it ended up
Acquired: Midjourney announced the purchase on 2026-07-24 (it closed in spring 2026 per Bloomberg reporting); Co-Star's ~24-person team joined Midjourney, founder Banu Guler became chief design officer, and the app continues to operate under her control.
Background
Co-Star is a mobile astrology app founded in New York in 2017 by Banu Guler, Anna Kopp and Ben Weitzman and launched on 2017-10-17. Users enter birth time and place; the app pulls planetary positions from NASA Jet Propulsion Laboratory data and layers on birth charts, daily horoscopes and compatibility reads written by AI plus human astrologers. Guler, previously a fashion-industry product and design director, has said the idea began with a hand-made, book-length birth-chart gift that convinced her astrology could move beyond a niche audience.
The bet was that astrology could sell as an aesthetic, social product: a magazine-like black-and-white UI, one blunt notification a day ('do not text your ex'), and no attempt to buy attention. That design was the growth engine - users screenshotted the sharp daily horoscopes and posted them on Instagram and Reddit - and Co-Star reached ~4.3-4.9M monthly active users and about 200K downloads a month on a near-zero ad budget.
The weak point was monetization. BaijIng, citing DianDian data, put monthly post-store revenue near $400-500K for most of the year, with a one-off December 2025 spike toward $1M from an $11.99 annual forecast product; the Pro Star subscription contributed under 10% of revenue, and one-time purchases like $2.99 chart reads dominated. Midjourney, in its first acquisition, bought Co-Star's parent in a deal that closed in spring 2026 and was announced on 2026-07-24; terms were undisclosed, Guler became chief design officer, and the ~24-person team joined.
As of 2026-09-02 Co-Star still operates as an app inside Midjourney under Guler's direction, while she also builds the design, product and front-end team for Midjourney's first standalone apps. The week after the announcement, 39 of 46 new App Store reviews were one-star, most citing AI, Midjourney or data concerns, while downloads rose about 16% and revenue only 6.4% (kchuhai) - a reminder that the audience that loved Co-Star's mystique did not all welcome its new owner.
What has to be true
- Co-Star proved a consumer hit can come from design and tone alone: ~4.9M MAU with a near-zero ad budget and no paid creatives in the 90 days before the deal.
- The stall is precisely diagnosed with numbers: ~$400-500K monthly post-store revenue for a year, a one-off December 2025 spike, and subscriptions under 10% of sales (BaijIng/DianDian).
- The outcome is unambiguous and dated: Midjourney announced the acquisition on 2026-07-24 (closed spring 2026), Guler became chief design officer, and the app continues.
- It is a clean contrast for founders: the same traits that generated attention - sharp tone, magazine UI - produced one-shot purchases rather than recurring value.
What can be applied
Aesthetic virality with zero ad spend can win users, but one-off 'curiosity' payments stall out; the design engine that could not grow Co-Star's revenue was exactly what Midjourney paid for.
Aftermath
As of 2026-09-02 Co-Star is an operating app inside Midjourney: founder Banu Guler is Midjourney's chief design officer building a cross-functional design, product and front-end team for its first standalone apps, and CEO David Holz has said Co-Star runs 'completely under Banu's control'. Review data right after the announcement was harsh - 39 of 46 new App Store reviews were one-star, most citing AI, Midjourney or data concerns - while downloads rose ~16% and revenue only ~6.4% (kchuhai). The acquisition price was never disclosed.
Sources
- Midjourney acquired the astrology app Co-Star
- Just now, Midjourney bought a top-3 AI astrology app? A teardown of Co-Star's UI-driven growth and stalled monetization
- Zero funding, $500M annual revenue: the AI image king's first acquisition is a 24-person AI astrology company
spotted an error? The archive wants to know.
Your turn
You just read one. Describe what you are building, and see who is betting on the same thing.
Free account · 3 free questions · no card