Skip to main content
You do not fill in a search form. You ask AI Fuel in plain English.
“Industrial buyers over 2million"∗∗"Whowantsretailinthenorthvalleyunder2 million"* *"Who wants retail in the north valley under 5M?” “Cash buyers for a 7 cap”
It turns that into the right criteria and searches your book.
On a seller’s listing there is a button that copies the listing as a ready-made prompt you can paste straight into AI Fuel. Faster than describing the property yourself, and it will not miss the asking price.

​
It searches your tagged buyer book only

Matching reads the people you have tagged as buyers. Nobody else appears, however well their data seems to fit.
Owning something is not evidence of wanting to buy. In the data, an office owner and an office buyer both look like “Office”. If matching went on criteria alone, every owner of the right asset type would surface as a buyer for it.The tag is the difference, and it means something specific: you recorded that this person told you they are looking. That is a fact only you have, and it is why the book has to be tagged rather than inferred.
To be clear: owners make excellent buyers. Most serious investors own several buildings and are always looking for the next one. An owner who tells you they are buying gets qualified and tagged like anyone else, and matches normally from then on.The same is true of sellers. A client selling a building to trade up into a larger one is a seller and a buyer at the same time, and both sides of that should be recorded. See Qualifying a Buyer.
One thing to watch: matching does not know which buildings a person already owns.If a client is selling a property and is also tagged as a buyer whose price range covers it, they can appear in the match results for their own listing. Nothing filters that out today.Scan the names before you start dialling. It is obvious the moment you see it, and it is exactly the situation the trade-up client creates.
The honest consequence of the tag gate: a thin book returns thin results, and that is the correct answer rather than a bug. If AI Fuel says you have two office buyers, you have two office buyers. See Buyer Book. When the book is too thin, it can widen the search to your old buying leads, but it will offer first and label those results as prospects, never as confirmed buyers.

​
Exact matches and partial matches

You get two numbers, and they mean different things. A blank field is a data gap, not a “no”. A buyer who wants industrial but has no price recorded is not disqualified from a $2M industrial listing, you just do not know yet. Those buyers come back as partial matches, with a note saying which fields are missing.
Treat the partial list as two things at once: people worth calling, and a cleanup worklist. Filling in those blank fields is exactly the work that makes the book sharper next time.

​
Turning matches into a call list

Once you have matches, you can build a call list from the full match set, not just the names shown on screen. The list re-runs the same match, so the count you were told is the count you get.

​
The reverse: matching a buyer to your listings

You can also go the other way. Give AI Fuel a buyer and ask who is selling what they want, and it scores your captured listings and rolls them up to the seller who owns them.
Two honest limits on seller matching.There is no location filter yet, because listings store a street address rather than a submarket. It matches on asset type, asking price and cap rate, so if you name an area it will tell you it could not filter on that.It only sees listings you have actually captured. A thin result usually means listings need their asking price and asset type recorded, not that there is no inventory.