Pick your models.
Pay per day, not per listing.
Trofey watches a market worth UAH 897M in live listings and recalculates every model's fair price daily. When a listing drops below market — you hear about it first, with the exact saving in hryvnia.
Here is what sold below market yesterday
Real listings, real savings — we publish every lot we scored below market once it is more than a day old. Subscribers get the live ones at that very moment. The marketplace listing may already be gone — good deals go fast.
























Three steps and the market works for you
You pick positions
Model, storage, condition — in the bot, the mini app or the web cabinet: one account, shared settings. Every position shows its day price before you pay.
We price the market daily
A live market price for every configuration, from real listings over 14 days. Problem lots form separate markets and never distort the picture.
Deals reach you first
A below-market listing alerts everyone watching it at the same moment, with an honest competitor count on the card. Speed decides the rest.
What is your phone worth?
Selling, or just curious — a valuation from the live market: model, storage, condition, lock, battery. An honest range and median from real listings, no signup.
Three ways to hunt — an honest comparison
| Trofey | A marketplace saved search | Listing-feed bots | |
|---|---|---|---|
| What it sends | only listings below market | everything matching the filter | everything matching the filter |
| Knows the market price | yes, daily for every position | — | — |
| Counts the saving in hryvnia | yes, on every card | — | — |
| Sees a listing drop its price | yes — we track every listing | — | — |
| Screens out problem lots | separate markets + filters the marketplace has none of | — | partly |
| Shows the competition | an honest counter on the card | — | — |
| Payment model | per day of access — clicks and finds are free | free | per alert |
A free saved search is a fine tool, and everyone's speed is comparable. The difference is elsewhere: it shows you everything, while the money hides in «below market» — and that is our job.
We read the whole listing — and send you no junk
A marketplace lets you filter by model and price. Trofey pulls out of every listing what the standard filters simply do not have:
The saving against market
The filter nobody else has: how far below its configuration's live market price a listing sits — in hryvnia and in percent.
Unlock classes
iCloud-lock, carrier lock, rSIM — read from the listing text. Each class is compared with its own market instead of posing as a cheap clean unit.
Impossible configurations
«iPhone 17 Pro with 64 GB» does not exist — a storage size that model never had is flagged as suspect before any alert goes out.
Condition from the description
Battery health, «one catch with the back cover», what is in the box — pulled from text where the standard filters are blind.
The seller
Shop or private, how many active listings, how long on the platform — the signals sit right on the card.
Freshness
How many minutes old the listing really is, bumps accounted for — because an hour-old find is already too late.
Phones and tablets are live. More is coming
The mechanism is universal: a position means its own dimensions in every category. In the bot each upcoming category carries a «tell me when it opens» button.
85 positions · 3 brands · open →
Premium tier · soon
Premium tier · soon
Premium tier · soon
measuring demand
measuring demand
measuring demand
Honesty is a mechanism, not a slogan
- The competitor counter never lies. The number on the card comes from the same query as the seat limit. A number that lies once is dead forever — which is why it stays honest.
- Deal history is open to everyone — delayed by up to one day. Check that we really do find things before you subscribe, right here on this page.
- The saving is frozen at detection time and is never redrawn afterwards.
- Too little data — we honestly say nothing. Fewer than eight live listings means there is no market price, not an «approximate» one.
- Anonymity by default. Your trophies and filters are invisible to others — competitors never learn what you earn on.
- Refuted hypotheses get logged alongside the confirmed ones — examples in the methodology.
Data under the hood
34,800 listings a day · market price = a 14-day median · problem lots are separate markets · the official NBU rate · prices recomputed once a day.
The subscription pours credits. Your picks spend them daily
- A precise pick — up to ~10 positions
- Alerts when a day price changes
- Everything in Basic, no position limit
- The «Finds» feed — the bot searches everywhere, you set the bounds
- Everything in Premium, every category
- Position exclusivity — you are the only one who sees it
1 credit ≈ 1 hryvnia. Unspent subscription credits expire at month end — which is why each tier grants more than its face value. Credits you buy never expire, always at 1 UAH = 1 cr. There are no idle-account fees.
Questions before you start
Where does the data come from, and is this legal?
We read public marketplace listings — the same pages anyone sees in search. We publish aggregates (the market price, the saving) and links to the listings themselves. Seller contacts are never stored or published. Details are in the methodology.
What is a «market price»?
The middle of the prices of live listings of the same model, storage and condition over the last 14 days. We compute it only when there are at least eight such listings — otherwise we honestly say nothing. Problem lots (iCloud-locked and so on) are separate markets with their own prices and never spoil the general one.
What am I actually paying for?
Days of watching, not listings. 1 credit ≈ 1 hryvnia; every position costs its own sum per day, depending on how many good deals it really brings. Taking a find, skipping it, opening ten listings — none of that touches your wallet: the day is already paid for.
How many people use this already?
Honestly: a handful — early access has only just opened. That is not a weakness but your window: competition for finds inside the service is almost nil right now, and the number of people on each position is always on the card before you subscribe.
How many people see the same listing?
Everyone watching the position receives it at the same moment — and the card honestly states how many of you there are. Seats are limited (up to 5 per position), so you know the competition before you pay. The hunter ranking will be anonymous by default: your trophies never show to competitors.
How do you screen out junk and scams?
We analyse every listing deeper than the marketplace filters do: impossible configurations (storage the model never had), unlock classes, battery health and defects from the description, sellers with shop signals, implausibly cheap lots. The junk never reaches you.
When do paid tiers and new categories arrive?
Right now it is early access: the bot works, payments are not switched on, and tier numbers may still change before launch. The plan is on the Roadmap; in the bot every upcoming category has a «tell me when it opens» button.
How do I delete my data?
The /reset command in the bot wipes every filter. For full account deletion write to us on Telegram or at [email protected]; we keep only the minimum: your Telegram ID, your filters and the alert log.
A finance CEO × a data CTO
Numbers sit at Trofey's core, not hunches. The full story of the team and its principles is on the team page.

For years she answered one question for a business — «what does this actually cost». Now she answers it for the listings market, phones and beyond: the tiers, the unit economics and the «no number out of thin air» rule are hers.

Over 6 years in data — engineering, analytics, ML, infrastructure. Tens of thousands of listings a day flow through the pipelines he built; every threshold is measured and every mistake leaves a test behind it.
Not a person and not a language model. He holds the market price of every configuration and hands it to an assistant or another service — no keys, no sign-up. What the two above measure, he makes readable to machines.