The score and the reasons
The first thing to read on a row is why it scored what it did. Here is one, ranked 14th of the 5,813 investors scored in this radius:number
0 to 1, best first.
gate times the weighted sum of the reason values, rounded to four decimals. Compare scores
only within one response.string
The sentence to show a user. The API already writes it for a person: no ids, no internal names.
price_fit is only 0.46. The
endpoint page has every factor, and says how the API rescales the weights
when it cannot score one factor.
The proof and the contact
object
What they did inside your radius, after your filters. The
investor block beside it is their whole-market
profile, so it is normal when the two numbers differ.array
Up to five of their nearest deals here. Proof to show the user.
array
A person paired with an address, best first: “send these to a skip-trace vendor in this order”. This host does not
serve it: the whole block is null and
contact_redacted is true.integer
5,813 here: the whole radius, not the page.
Things you can change
- Leave out the subject facts and the ranking is proximity, recency and documented wholesale buying only.
- Send
subject_arv=560000instead of the asking price, or with it. The API uses each flipper’s own buy-to-resale ratio to convert an after-repair value into the price that flipper is likely to pay. - Tune the weights per request with
w_price_fit=0.4. The effective weights come back inmeta.weights. - Narrow the field:
active_within_months=12drops anyone with no purchase in the last year.buys_wholesale=truekeeps only investors with a documented purchase from a wholesaler.investor_kind_exclude=WHOLESALERremoves the competition.
w_activity=1 multiplies every score by the probability that the buyer buys again soon. It is off by default. See
Same buyer, cash right now.