Not an aged list. Not a form fill from six months ago that four shops already worked. A file cut from people who searched credit terms in the last 48 hours, filtered to the ones who can actually carry a monthly fee, and handed to one buyer. Plus the sending stack to work it without putting the channel that already pays you at risk.
Every number on this page came off a file on disk. Here is the session that produced it, start to finish, with the identifiers masked.
They are not one audience. Once your dispositions come back we can buy them separately, and weight the ones your closers actually convert.
They just got told no by the market. Searched bad-credit borrowing terms inside 48 hours, which means they have self-declared the problem you fix, in the same week they hit the wall.
sharpest urgency in the setSearched credit repair and credit report services directly. They are shopping the exact service you sell, which also means a competitor is being shopped in the same hour. Speed is the whole game on this one.
highest expected closeCredit terms sitting alongside a refinance or auto signal. There is a specific purchase in front of them and a score standing in the way. A deadline keeps people enrolled, so this is the retention play nobody sells.
best subscription fitWatching their score, aware, engaged, no acute pain yet. Largest volume, softest signal, and the one to test last rather than first.
largest volumeUnknown income is excluded rather than shipped as a hopeful blank, which is expensive on purpose: a third of the pool has no income band and all of it is dropped. That is the difference between a lead file and a list.
What survives is 42,501 people a day, against a floor of 26 closers working roughly a thousand a week. Supply is not your constraint and we are not going to pretend it is.
You said it yourself: if this one source breaks the other ten, that is the problem. A cold file blasted from the same numbers your team calls and nurtures from will pull opt-outs into the reputation your paid lane depends on. That is not a data problem, it is a plumbing problem, and it is the piece we would build with you.
Cold outreach goes out on its own registered sending identity and its own number pool, with nothing shared with the numbers your closers dial from or the ones your existing customers already know.
Opt-out handling sits on the cold stack and suppresses globally, so a stop from a cold message is honoured everywhere but the reputation hit never reaches your main lane.
The sending stack is provisioned and carrier-approved within three days of you saying go, so it is running before the first round has finished being worked rather than arriving weeks later once the damage is already done.
Which means you get both, this week: the file to work on day one, and the stack to work it from inside 72 hours. Every other seller of this data hands you the file and leaves the part that can hurt you as your problem. That part is the reason you have said no to all of them.
That is 978 people a day, not 42,501. The rest is a real but softer signal. Anyone selling you a file this size as "all searched yesterday" is selling you the tail as the head.
Two in five of the raw pool is 55 plus and one in five is 65 plus. That group churns off a monthly plan fastest and complains most, and a complaint rate is what costs people their ad account. We cut it rather than pad the count.
The contact data audits clean. That proves the people are real. It does not yet prove the intent signal converts, and only your closers on live calls will. That is exactly what round one is for.
Rounds of 1,000, about five days each, alternating, with the cut rebuilt between every one from what your closers report back. You never fund two in a row without watching us match the last one. You are not buying a list, you are funding a file that gets sharper six times.
If a round misses the buy-box you gave us, we cut it again on us until it lands. Six rounds is the plan, not the limit.
One buyer per record. We cannot promise nobody else independently sourced the same person, and we will not pretend otherwise.
The production rate gets set after the run, from what your team logs as answered, worked and closed. Neither of us should be guessing that number today.
States, the income floor you want, the age range, and how many your closers can absorb in five days. We cut the file to it while you watch the count move, and you keep sample records from your own filtered set either way.
Figures measured on a single intent pull, 8 to 9 August 2026, United States, top and mid tier, five consumer credit topics. The 1,000-record sample referenced in the integrity audit is the highest-ranked slice of the 85,002, not a random draw, and spans 50 states. Household income, net worth and homeowner status are commercial estimates, not verified financials. We place no calls and run no registry scrub, because that obligation belongs to whoever dials and cannot be transferred; every record carries its status so your team applies your own policy. We do not supply date of birth or social security number. A record is a qualified candidate, not a confirmed applicant, until your team has spoken to the person.