First 1M people watched hurricane videos, then an earthquake app: how Ecuador's Radar Earth got pushed up by a massive quake

Radar Earth Terremoto is a Spanish-language earthquake app: a live quake map, alerts by magnitude and distance, plus an experimental feature that uses the accelerometers in users' phones for early warning. About 80% of features are free; the rest are subscription. It went live on Google Play in Jul 27, reached 87k installs by Nov 6, and more than half of those came in the three weeks after the Aug 10 M7.4 Colombia earthquake. The product didn't drive this alone: the team already had a 1M-follower TikTok disaster account and a 960k-follower Facebook page. Below: how it grew, where the traffic came from, how much money it makes, who it competes with, and what you can take away.

At 4:13 a.m., who sounded first

Early on 2026 Nov 3 4:13, an M5.6 earthquake struck Chaparral, Tolima Department, Colombia. That morning 11:15, the Facebook page "Alertas y curiosidades" posted a map with a one-line caption:

Nuestro sistema de Radar Earth emitió alerta temprana en varias localidades de Colombia en algunos puntos superando la velocidad de Google! (Our Radar Earth issued early warnings across several Colombian towns, in some places faster than Google!)

The post got 1,500 reactions, 186 comments and 89 shares on Facebook (11-06) within three days. The comments weren't all praise. One said "Depends on where you live. In Cali, Google sounded first"; another said "The notification came after the shaking"; a third said "The alarm went off 4:12, luckily I was in Boyacá and felt nothing."

These comments capture the product: it's racing a free, built-in system feature (Google's Android earthquake alerts); speed varies by location, and the team's own site says there is "no time guarantee"; but with each earthquake, a group of people discusses it under the same post. After that quake, installs on Play rose by 628 in Nov 3, then by another 726 in Nov 5 (Google Play install counts, 11-03, 11-05).

What it is

The Play store page (Spanish, 11-06) lists: a live quake map and event list; domestic and international quake alerts filtered by magnitude and distance; tsunami information; "community-assisted experimental early warning (EEW)"; and alerts that sound, vibrate, flash the light, and announce the magnitude and distance by voice. The store page itself is cautious: early warning "may not detect every earthquake, may give no lead time, is for reference only, and cannot replace official alerts."

The mechanism combines two things: if phones with the app detect shaking at the same time in the same area, it flags an earthquake; and it pulls data from official seismic networks like USGS and EMSC. The official technology page claims "145k users across the Americas," followed immediately by "this is a self-reported number, not an independent measurement" (radarearth.app/tecnologia, updated 11-04).

It's built by Ecuadorian John Alexander Villota Herrera. The Play developer name is "alerta inminente" (imminent alert), and this is the only app under it; he also makes MeteoPaint, a desktop weather visualization tool. The iOS version launched in Sep 10 under his real name. In-app purchases on Play are priced $0.99–9.99.

Users were specific about what they liked. In Sep 17, a user named Juan Pablo wrote on Play: "Today was a day with lots of earthquakes, seven or more low-magnitude ones, but it shook for a long time in Sipí. The app responded well and warned me in time: first a sound, then a voice telling me the magnitude and how far away it was." Sipí is in Chocó, the province hit by the August major quake. The negatives were specific too: a Lima user said a 3.8-magnitude quake "notified 6 minutes late" and that they kept getting woken at night by small quakes in distant provinces; another said they set alerts for 4.1+ magnitude but got alerted for a 2.3. The developer replied under each negative review (Google Play reviews, 8–September).

How it grew

Sources: before end of October, third-party estimates (single-day figures are averaged across observation gaps, so read them for trend only); from 10-29 on, daily installs from Google Play.

The shape is clear: this isn't a slow climb, it's a series of steps, each tied to an earthquake. The three weeks after the Aug 10 quake brought about 47k installs, more than the following two-plus months combined. The October Panama M7.7 quake was larger but brought only about two thousand installs. The people that the major quake brought in had already installed it the first time (inferred).

1,548 ratings, 1.8% of installs, which looks organic. On the charts it only shows up in small markets: Mexico weather Top Free #68 on Play, and weather Top Grossing #115–126 across 14 stores including Vietnam, Saudi Arabia, Indonesia, India and Mexico. On iOS it's Top Free #1 in Guatemala weather, #7 in Peru and #16 in Colombia, but the Colombia store has only 83 ratings (Google Play and App Store charts, 11-06).

Where the traffic came from: a disaster account built over two years

The creator's own TikTok @radarearth.pro. As of Nov 6 it had 1M followers, 953 videos and 11.1M likes, with a bio reading "earthquake early warning and severe weather alerts." The earliest videos go back to Oct 2024, mostly hurricanes Milton and Erick, cold fronts and cyclone visualizations, with the top video at 5.12M views; in Oct 2024 alone the account totaled about 21M views (TikTok, 11-06). So that audience was built on "a storm is coming" content, and is naturally people who care about sudden disasters. Two weeks before the app launched, in Jul 11, the account was already posting earthquake analysis videos, one with 327k views. By 2026, single videos were mostly tens of thousands of views, and the creator switched to live-streaming the real-time quake map on TikTok.

The Facebook page "Alertas y curiosidades". About 957k followers, run by the same person, who also promotes MeteoPaint. It posts live app demos (in Aug 27, a 5-minute "Bogotá and Ecuador live earthquakes" with the caption "This app is called Radar Earth Terremoto! 80% of content is free", 11k views), "we beat Google" reports after each quake, version update notices, and polls asking followers to rate Radar Earth and MeteoPaint videos. The most common comments are "I'm in X, my dot doesn't show on the map" and "Is there an iPhone version?" These are people who keep asking.

Streamers and small creators, with referral codes. TikTok live-streamer @waldoamarilla (159k followers) repeatedly posts Radar Earth alert screen recordings, with captions like "@RADAREARTH.PRO gives 5 days of full membership free" and "code WALDOZO." @pz_cam posted "An earthquake hit during the livestream, and @RADAREARTH.PRO's app went off" in Sep 19, 26k views; @dariensanchezz's Aug 18 Venezuela quake clip "App: Radar Earth" got 306k views. Clips of "the app suddenly alerted mid-livestream" spread on their own.

Store search: unclear. After the August major quake, Colombian media were telling people to turn on Google's Android earthquake alerts and listing "the best earthquake apps" (Infobae, AS Colombia, around 08-10). Radar Earth isn't in those lists. How much it gets from searches for "sismo" or "terremoto" can't be seen from outside.

Paid acquisition: none found. The ad libraries show no ads from it; the review rate and the quake-driven curve both look more like organic traffic. No media coverage, and no discussion found on X or Reddit. It spreads only within the creator's own fan base and a few Latin American streamers.

How much money

Not much. About 80% of features are free; advanced alerts and full features require a subscription, with in-app purchases at $0.99–9.99 on Play. It reaches #115–126 in weather Top Grossing across a dozen-plus stores. The bar for these small-market category charts is low, so it shows some people are paying, but not at scale (inferred; revenue isn't public). Earthquake apps have a built-in problem: nobody thinks about them between quakes, and people who install them don't open them, so the urge to pay only shows up in the few days after a quake.

Who it's competing with

The core mechanism has been done before, and Google ships a free big-company version. What Radar Earth does differently is three things: fully Spanish-language, focused on South and Central America (Colombia, Ecuador, Peru, Central America), and alerts built to be loud: sound, voice, flashing light, wavefront animation, so they record well and stream well. The creator also makes weather visualizations himself. None of these are hard to copy. What nobody can copy is the 1M followers.

If it were you

Verdict: You can learn a playbook from it, but copying it step by step isn't worth much. Its real fuel is a disaster account the creator spent two years building to a million followers, plus a big earthquake that hit two weeks after launch. Nobody following it gets either of those. The product itself sits in a crowded category with a free Google version, growth comes in bursts with each disaster, is cold between them, and the revenue is small.

Who it fits: Creators who already run a small-language account covering weather or disasters in a quake- or storm-prone region, with a few hundred thousand followers, and want a paid tool for that audience. For them, the "content account → alert app" path is almost ready-made.

Who it doesn't fit: Developers with no existing audience who expect store search to bring in users. Earthquake apps are dominated by older products with tens of millions of installs, and Google gives its alerts away in the OS. Also developers aiming for steady subscription income, since the urge to pay only lasts a few days after a quake. And anyone who can't take on safety liability: once an alert app says "warning," a missed or false alarm carries more reputational and regulatory risk than a typical utility.

How to build it: Don't start with the app. Start with the content account. Pick a region with frequent quakes or typhoons and few local-language alert apps (Turkey, Indonesia, the Philippines, Chile, Nepal; search Play in the local language for "earthquake" and "alert" first to see how many localized ones exist). Use short videos of disaster weather and quake visualizations to build an audience first. For the app, start with a map plus adjustable alerts, pulling directly from USGS, EMSC and local official seismic networks. Crowdsourced early warning can come later, or never. Design the alert screen as an ad: it should record well and stream well. The recurring complaints in negative reviews are "too many alerts, too noisy," "settings don't take effect," and "notifications arrive late." Get those three right from day one.

How to acquire users: Post everyday content on your own account. When a quake or typhoon hits, go live with a real-time map. Send creators who stream radar and map content a personal promo code plus a few days of free trial, so clips of "the app alerted during my live stream" spread for you. After each event, post a map showing "where we alerted first." Just don't claim you beat Google the way Radar Earth does. Commenters will pick that apart on the spot.

How long you have: No fixed deadline. The window follows the disasters, so it can't be scheduled. The real time cost is building an audience. Based on @radarearth.pro's experience, that takes one and a half to two years. If you have no account today, count on at least 2028 years.

What would show this read was wrong: ① Over the next few months, Latin America has no major quakes, yet Radar Earth still installs steadily at over a thousand a day. That would mean it has organic traffic independent of disasters, and the category is worth more than I've judged. ② It climbs into the top 30 of the weather charts in Colombia or Peru, which would mean earthquake alert subscriptions can support a business. ③ The reverse: Google makes Android earthquake alerts more complete and faster across Latin America, or Radar Earth stumbles publicly over a missed or false alert in the comments. Either would suggest this path doesn't go far even for someone with a million followers.

Data and sources

• Google Play store page com.radarearth.app (Spanish, read 11-06): description, in-app purchase prices, rating count, reviews and developer replies. Install counts from 10-29 onward are collected daily; earlier figures are third-party estimates. Chart ranks 11-06.
• App Store: launch date of iOS id6809333133, developer name, weather chart ranks by store, and rating count (11-06).
• radarearth.app/tecnologia (updated 11-04): 145k users is self-reported by the developer, and alert timing is not guaranteed.
• Facebook "Alertas y curiosidades": 11-03 alert-performance posts and comments, 08-27 demo videos, read 11-06.
• TikTok: profile and posts of @radarearth.pro, and videos by @waldoamarilla, @pz_cam, @dariensanchezz, read 11-06.
• Earthquake sources: Spanish Wikipedia "2026 Colombia earthquake" and "2026 Panama earthquake," ReliefWeb, CNN Spanish (08-10); El Tiempo, Semana (11-03 Tolima M5.6).
• Competitor installs: Google Play, 11-06.
• "Faster than Google" and "20 / 34 seconds early" are the team's own claims, not independently verified. Revenue is not public. Where I wrote "inferred" or "estimated," those are my inferences and calculations.