How does each run build on the last?
Two inputs shape the next run: your team’s fit ratings and the reasons accounts were rejected.
01Reject
Every reject keeps its reason.
An account that fails is rejected with the rule it failed and the evidence, such as company size or no buying signal in its window. On the latest runs each reason also carries a rule code, so a pattern across a run shows up as a count, not a hunch.
02Rate
Your team rates what came back.
Someone on your team marks accepted accounts as a good fit or not, at any point, including after a meeting. Each rating is tied to the account and the version of the profile that found it.
03Rerun
The next run starts from what the last one learned.
Your team’s ratings, with their reasons, go into the next research brief. When reject reasons show a pattern, SalesOne suggests a profile change, for example a size range moved from 200 to 250 employees, and a person confirms it. The change is saved as a new version, and the next research run in the same market uses it.
Why does your team’s judgment shape the next run?
AI does well on work within its reach: checking hundreds of accounts against the same rules, counting why they failed and drafting a change to the profile. Judging whether an account was worth the time, and whether the change is right, is your team’s call.
So SalesOne splits the work that way. Your team rates the accounts that came back and confirms each profile change. S1 agents count the reject reasons, carry the ratings into the next research brief and write the confirmed change as a new version. Each run starts closer to the companies with a problem you solve, and every version is kept.
Refine40%+higher human-rated quality when consultants used AI on tasks within its reach.Field experiment, 2023. 758 consultants.Source (opens in a new tab)What shapes the next run, and what does not?
Fit ratings
Your team’s good fit and not a fit ratings on accepted accounts, with their reasons. They go into the next research brief, and S1 agents apply them in the next run.
Reject reasons
The reason every rejected account failed, counted across the run. When a pattern shows, SalesOne suggests a profile change for a person to confirm.
Versions
Every change is saved as a new version, from v3 to v4, and rules your team set are labeled “Client decision”.
Replies and call outcomes
Captured in SalesOne on each account’s history, for your team to read. They do not change the profile on their own.
SalesOne product specification, October 2026.
Why does an ideal customer profile need refining?
An ideal customer profile is a guess until it meets the market. The first version is written from your best customers and your proof, and it is usually close. The gaps show up in the first run: a size range that lets in companies too small to buy, or a signal that fires often but rarely means anything.
Most teams fix this in a quarterly review, from memory. SalesOne fixes it from the record. Every rejected account keeps its reason, so you can see how many accounts in a run failed on size. Every rating from your team says which accepted accounts were worth the time, and goes into the next research brief. When the reasons show a pattern, SalesOne suggests the change; once someone on your team confirms it, it becomes the next version of the profile, and the next research run in the same market starts from it. Every version is kept.
Who does what?
S1 agents do the homework. Your team builds the relationship.
S1 agents
Record a reason for every reject, carry your team’s ratings into the next research brief and suggest profile changes.
A person on your team
Rates accepted accounts, confirms each profile change and sets the rules that are your team’s call.
Never
Changes the profile without a person’s confirmation, or counts a reply as a rating.
Where this fits in the loop
01
ProfileWho you are, what you sell, your proof
02
TargetYour ideal customer, written as rules
03
ResearchEvery account, from dated sources
04
ScoreFit, timing, reach and who decides
05
SequenceA plan and a week of steps per account
06
EngageOutreach your team approves
07
CloseMeetings booked, the brief attached
08
Refine
Each run builds on the last
Questions buyers ask
What is ICP refinement?
ICP refinement is updating your ideal customer profile from evidence instead of opinion. In SalesOne the evidence is your team’s fit ratings and the reasons accounts were rejected. Ratings go into the next research brief; a profile change is suggested from the evidence, confirmed by a person and saved as a new version.
Does SalesOne learn from replies?
No. Replies and call outcomes are captured in SalesOne on each account’s history, for your team to read. The profile itself is shaped by your team’s fit ratings and reject reasons, and a person confirms every change.
What is a reject reason?
The rule an account failed and the evidence for it, such as “Failed: employees” or “No buying signal in its window”. On the latest runs each reason leads with a rule code, so reasons can be counted across a run.
When does the profile change?
Between research runs, and only when a person confirms it. When reject reasons point to a change, SalesOne suggests it; once someone on your team confirms, S1 agents write the next version and the next run uses it. Earlier versions are kept.
Can my team still change the profile directly?
Yes. Your team can set or change any rule, and rules your team set are labeled “Client decision”, so it is always clear which rules came from you.
Related reading
See which companies have a problem you solve.
Bring your offer and one customer you would like more of. We will show you the accounts that fit, what changed at each one, who owns the problem, and the point of view your team would walk in with.