From 656 to 160k buyers: how TokoAI's buyer database grew
A week earlier, our buyer database held 656 importers. Now it is 167,000+ across 28 industries, with 438,000+ product-buyer match pairs. Not a one-time purchase - the result of continuous accumulation.
How the buyer database grew
TokoAI's buyer data was not bought - it was built step by step. It started from global public purchase-record data: a few hundred importers, manually verified contacts. Slow, accurate, small-scale - every record confirmed.
Then more public data sources were found. Not one by one, but batch-matched: cross domestic factories' product categories with overseas buyers' purchase records. A factory makes auto parts - we give importers currently buying auto parts.
From 656 to 167,000 is not a simple number doubling - it is a qualitative change in coverage:
- Industries from 26 to 28: niche categories once unreachable now have matching buyers.
- 438,000 match pairs: on average 20+ matching buyers per factory.
- Multi-country coverage: no longer limited to a single national market.
What more data means
For a factory, the point is not "the database got bigger", but:
Categories that could not find buyers before, now can.
A plush-toy factory might have had no matching importer in the database before. Now it can match a dozen overseas buyers actively purchasing that product - company name, location, purchase record, all clear.
Match precision changed.
Before: matched by broad industry ("you make toys, here are toy importers"). Now: product-level - down to HS-code level. You make plush toys, you get buyers purchasing plush toys, not all toys.
Emails went from "hinting" to "naming names".
Early templates said "we have some matching buyers" - believe it or not. Now we can write directly "4X4 REBUILDING LTD (UK) is purchasing your category" - a real, verifiable company name the recipient can look up themselves.
Expansion is not the finish line
167,000 is not a target number - it is the current accumulation node. Buyer data keeps expanding. Every week new sources connect and new match pairs generate.
Our logic is simple: each send produces data, data feeds back to make matching more accurate, better matching makes the next send more effective. A virtuous loop.
"Smarter the more you use it" is not a slogan - it is the system architecture.