Draw a plot on a map, pay 559 lira to see if there's gold underground: six weeks with a Turkish treasure-hunting app

A Turkish individual developer used public satellite data and Gemini to build an app that "looks at your land from space." Outside Turkey almost no one has heard of it. In 6 weeks it got 15k installs and ranked 30 on Turkey's utility Top Grossing chart. The people actually using it are the large number of treasure-hunting hobbyists in Turkey. Below: what it sells, where the money comes from, why such a niche product gets paid, and whether you could copy it in another country.

"We've wondered for years what's really in our land"

Oct 24, a user on the Turkey App Store gave it five stars: "Senelerdir arazimizde neler var merak ediyorduk. Bütün analizlerimizi yaptırdık." (We've wondered for years what's on our land; we had all our analyses done.)

Four days later, on Oct 28, another user gave it one star with the title "Tamamen para tuzağı", a pure money trap: "We ran land analyses, the map marked precious metals, structures and water, but on site we found none of it."

Both reviews are about the same app: Arazi Tarama: Land Scan AI, listed on Google Play in Sep 21 and on iOS in Oct 22, with the iOS seller Kasim Yilmaz, an individual developer. The usage is simple: draw a plot's boundary on a map, and after about a minute the map shows numbered markers for "precious metal traces," "groundwater," "buried structure traces" and "void / cave traces," each rated in three tiers: VAR (present), OLASI (possible), ZAYIF SİNYAL (weak signal). Next to it is a Turkish-language AI interpretation you can keep questioning.

Who it actually sells to

The website says "land analysis" and includes quite respectable features like crop suitability, distance to the nearest active fault and parcel numbers (ada/parsel). But looking at its other features shows who the real users are:

• İşaret Çözücü (symbol interpreter): photograph symbols carved into a rock, and the AI tells you their "traditional meaning" and suggests scanning the surroundings from satellite. The Play store listing is more direct: it interprets the symbol's meaning in "folk belief (define tradition)."
• Coin and artifact identification: photograph coins, seals or pottery, and it tells you which civilization and era they're from.
• Metal detector: the phone's magnetometer used as a detector, and it can walk a field cell by cell to draw a heatmap. The website honestly states the range: small items a few centimeters, large objects 10–30 centimeters.
• Navigate to discovery points, expert opinions, community: community posts show location only down to the province.

In Turkish, "define" refers to buried treasure and also to the act of hunting for it. Turkey's treasure-hunting scene is large: on TikTok, dealers selling underground imaging devices have videos with up to 1.04M views (TikTok, @mena.dedektor34). "Having someone look at your land by satellite (uydudan yer baktırma)" has long been a business, with people taking orders via DMs on TikTok and Facebook, and some selling Surfer, a Windows automated analysis tool, listed in YouTube descriptions at 20,000 TL per year. These services generally have a bad reputation, and TikTok has dedicated anti-scam accounts making videos that say "satellite treasure hunting is a scam."

Arazi Tarama didn't need to educate anyone. The demand, the habit of paying and the scene's jargon were all ready-made; it simply swapped "pay someone to look at the land" for "look at it yourself on your phone."

The numbers: few users, but a high share pay

The most telling is the gap between the two rankings: around #30 on Top Grossing versus around #120 on Top Free. Not many people install it, but the share who pay is far higher than their neighbors on that chart. On that list, its neighbors are VPNs, cleaner apps and "who viewed my profile"; it's the only vertical niche tool.

Another number: analyses run (14,000+) and registered users (13,500+) are about the same. That means most people used only the one free analysis, and a minority pay, but those who do buy several plots' worth of inferences. How much revenue this brings in is not visible from outside, and the developer has never disclosed it.

Where the money comes from: one lock

Pricing is listed on the website. There is no subscription; credits are sold per use and never expire:

(Play Store in-app purchase range is $2.99–$99.99; 559 TL is roughly $13 USD, estimate.)

The key is one line in the official FAQ: "In the free analysis, the discovery points show as locked markers; one analysis credit unlocks them all." So the first time you outline your own land, it tells you "there's a VAR-grade precious metal signal here" but not where. For someone who's been curious for years, that's nearly impossible not to pay for.

Other details are also designed around payment: re-scanning the same plot costs no credits (to build trust), 3- and 10-scan bundles have a lower unit price (treasure hunters often have several plots, or want to scan the surrounding area), and the phone metal detector has a free trial that charges after it expires.

559 TL isn't cheap, and a one-star review says "free first analysis sounds nice, but this price is too much." Still, compared with what this niche used to pay, human satellite-analysis DMs quoting prices, Surfer software at 20,000 TL a year, and ground-penetrating imaging equipment costing tens of thousands to hundreds of thousands of TL, it's the cheapest entry point.

Where users come from: most likely search

Its own social presence is weak and can't explain 15k installs: the official TikTok account @arazitarama has 78 followers and 10 videos, with the top video at 728 views (TikTok, 11-06). Its content features "satellite analysis" of popular sites like Göbekli Tepe and the "Noah's Ark remains" on Mount Ararat. The Facebook page has about 140 followers but shows over 500 people "talking about this," with engagement far above follower count, usually a sign of boosted posts, though I couldn't find specific ads, unverified.

The more likely main channel is inferred from store search. The app name "Arazi Tarama" is a generic Turkish term meaning "area scanning / land scanning"; searching it on TikTok, nearly all top results are treasure-hunting equipment videos, showing it's a high-frequency term in the community. The name itself is the keyword, and there are few similar apps, so searching this term on Play likely ranks it near the top. There's no discussion of it on X, Reddit, or in the media.

On 10-22 it launched on iOS, and on 10-31 it shipped a 3D terrain update, both announced officially, but Play's daily growth showed no clear jump afterward.

Who it's competing with

Similar apps exist, but they all emerged in the last few months and are very small. The combination of "Turkish language, self-service, satellite plus AI conclusions, pay per use" is held by it alone on the charts. I also haven't seen anyone copying it.

What it got right

1. It turned a poorly regarded paid service into a "legitimate tool". The official site has a whole section called "Transparency," opening with "Arazi Tarama is not a magic wand," then listing its data sources: ESA Sentinel-2, Copernicus elevation models, Esri imagery, USGS earthquake catalog, geological maps, Google Gemini; conclusions are stated only as probabilities; the footer cites Turkey's Law No. 2863 on cultural heritage and the Mining Administration MAPEG, stating that digging requires permits and the app cannot replace them. Adding agriculture and earthquake features lets it list and pass review as a "tool" and makes users feel it isn't a scam.
2. The pricing targets this community's wallet: free first time, key locations locked, a few hundred lira per use, no subscription.
3. The entire treasure-hunting process lives in one app: outline the plot → 3D terrain view → navigate to discovery points → phone on-site scanning → photograph symbols and coins → ask an expert → post to community. Each step is a reason to come back and buy again.
4. Fast. One person, 6 weeks: Android, iOS, 3D terrain, metal detection, and community all shipped; the Play version went out in two versions within 10-25 to 10-31. The data is all public, interpretation is handed to Gemini, no heavy lifting.

Conversely, this is also its weakness: no moat. Anyone can download the data and tune the models, and VeldScan has already shown that switching languages makes a new app.

The risks are in the reviews

On iOS there are 8 reviews, 3 of them one-star, all saying the same thing: it's marked on the map, but nothing was found on the ground; too expensive. That's an unavoidable problem for this kind of product: it sells "maybe," users want "certain." Two other risks: the stores might treat it as a misleading feature, and Turkey's illegal digging is under constant enforcement, so such apps could be implicated. Neither has happened yet, but followers should factor them in.

If it were you

One-line verdict: worth following, but don't compete with it in Turkey; instead, take the same playbook to another market where "treasure-hunting culture is strong and people already pay to have land looked at," before it or VeldScan goes multilingual.

Who it suits: 1–2 people, one who can build an app (maps, satellite data processing, and calling LLMs are all not hard), and one who is a native speaker of the target market, ideally already in the local treasure-hunting, mineral-prospecting, or water-finding circles, knowing the slang and symbols.

Who it doesn't suit: those who don't understand local culture and only want an English global version (VeldScan is already doing that, and English-speaking global markets are scattered without a concentrated community like the one define has); those who can't get past "selling probabilities" will always have one-star reviews.

How to build it: launch Android first, iOS later. Outline the plot → within a minute, a map with numbered markers + native-language AI interpretation → first use free, locations locked → sell credits per use, no subscription, no expiry. Learn its "transparency" approach on the website and store page: list data sources, give only probabilities, cite local heritage law as a disclaimer, then add legitimate features like agriculture and groundwater. Don't skip companion features like the phone magnetometer, symbol interpretation, and coin recognition; they're the reason for repeat purchases. Markets worth considering: Arabic-speaking regions (Gulf, Egypt, Iraq), Iran, the Balkans, Central Asia, South Africa, Latin America. Pick which market to test first by checking how many accounts on TikTok sell treasure-hunting gear or "satellite land viewing," and how high their view counts are.

How to get users: use the generic local-language term for "land scanning / treasure scanning" as the app name, which is likely its main traffic source; then find local treasure-hunting equipment dealers and treasure-hunting creators on TikTok / Facebook to partner with. Their single videos reach hundreds of thousands of views, much faster than building an account from scratch. Using historic sites as "satellite analysis" demos can work as content, but by its own experience such content gets small views, so don't expect much.

Deadline: it finished both ends plus 6D in 3 weeks, so followers should also launch within 4–6 weeks; if by 2027 January it or VeldScan has shipped Arabic or Persian versions, the market you choose will need to change.

Signs you've misjudged: ① it drops out of the top 100 of the Turkish tools top-grossing chart while installs keep rising, meaning payment can't hold up and only one-time curiosity remains; ② it gets removed from Play or the App Store, or the stores start a sweep against similar apps; ③ a month after your launch in the chosen market, your post-free-analysis payment rate is clearly below its (its top-grossing rank should be 80 or more places higher than its top-free rank, and yours should also rank higher on top-grossing than top-free); ④ "money trap" one-star reviews make up more than half of your reviews.

Data and sources

• Installs, daily figures, charts: Google Play Turkey store, 2026-10-29 to 11-06.
• Pricing, features, self-reported user count, locked free-analysis locations, data sources and legal notes: arazitarama.com (11-06), Google Play store page.
• iOS seller, rating, original review text: App Store Turkey store (11-06).
• Social numbers: TikTok @arazitarama, @mena.dedektor34, @xicatprojeleri, etc. (11-06); Facebook @Arazitaramax (search snippet).
• Surfer analysis price: YouTube video description from Osmanlı Baykuşu.
• Places marked "inferred," "estimate," or "unverified" are inferences and estimates; revenue has no public figures.