Similar product matching
The query is stemmed into Polish product words and matched against product titles across the index — the product type up to the first preposition, so “case for iPhone” finds cases, not iPhones.
Product Research finds Allegro products similar to the one you want to sell, measures the market around them — revenue, sellers, prices, competition — and says how hard entry looks. In plain words, with the numbers next to the answer.
Product Research says whether the market pays for a newcomer — and at what price — before you order samples, pay for photos and lock money in stock. Ten reports cost less than one wrong SKU.
One field. Describe the product the way a buyer would search for it. Marzi turns the phrase into Polish product stems and finds the Allegro product cards whose titles match — the product type, not a category guess.
The similar products are read from the index: how much they turn over, how many sellers share them, where the price sits and how crowded the head is. Each measure is scored against the distribution of the whole index, so “high” means high for Allegro, not for this list.
Marzi's assessment states whether the opportunity is worth pursuing and why — with a score, entry accessibility, and how confident the answer is. If the list includes products that are not yours, remove them and the verdict is recomputed.
Every figure on the screen has the page it was taken from and the method printed next to it. These are the parts.
The query is stemmed into Polish product words and matched against product titles across the index — the product type up to the first preposition, so “case for iPhone” finds cases, not iPhones.
Revenue, buyers, sellers and price are measured on the matched products, not on a category that happens to contain them.
Every measure is a percentile of all 10,217 niches, so a score of 71 means the same thing in pet supplies and in power tools.
A written verdict grounded in the figures on the page. Every number in the text is checked against the measurement it came from.
Remove products that do not belong, keep the ones that do, and the market is re-measured for the set you actually mean.
Reports stay in your account. Reopen last month's question and see whether the answer moved.
Marzi says which is which on the screen itself. Here is the short version for Product Research.
The first question, answered before the first purchase order: is there room in this market for one more seller, and at what price.
Check the next product on the list against the market it enters, and compare last month's report with this month's.
Screen a catalogue: run the products one by one and keep the reports side by side.
The product, in plain words: “wireless earbuds”, “etui na telefon”, “garden hose 30 m”. Marzi works from the product type, so the shorter and more concrete the phrase, the better the match.
The phrase is turned into Polish product stems and matched against product titles in the index. The match keys on the product type — the words before the first preposition — so accessories are found as accessories, not as the thing they attach to.
A score, the accessibility of the entry, a confidence level and a short written assessment. Under it, the facts: revenue of the set, sellers and offers, median price, competition and the share held by the head. Every figure in the text is checked against the measurement behind it.
Yes. Open Adjust, untick the products that are not what you meant, and the market is re-measured for what is left.
As current as the last weekly pass of the index — the report states the window it was measured over.
Yes. The cabinet runs in English with euros or in Polish with złoty; the query can be in either language.
7 days, no card required. The whole index, your trackers and the MCP endpoint are live from your first sign-in.