# A good prospect has no address

The prospect is right, the research is right, and no source published an address. Connect will not generate one — not from a name and a domain, not from a pattern that works for most companies. The record is kept as a researched prospect with a next action that is not email, and that is a deliberate trade rather than a missing feature.

- **Status:** Available
- **Audience:** both
- **In the app:** #/prospects, #/follow-ups
- **Last verified:** 2026-09-10
- **Canonical:** https://connectbyjbrh.com/docs/troubleshooting/prospect-no-email/

## What the record actually says

Qualified, evidenced, and not contactable by email. Those three states coexist, which is exactly why the counts above the list report qualified and ready-to-send separately. The gap between them is this population, and it is a list rather than an error.

Nothing on the record is provisional. The claims carry their sources, the fit reason stands, and the score reflects real evidence. The only absent field is the one nobody published.

## Why the gap was not filled

A generated address is a guess wearing the clothes of a fact. The usual pattern — a first name, a full stop, a surname, the company's domain — is right often enough to be tempting and wrong often enough to be expensive. Three costs follow from being wrong, and only the first is obvious.

1. The message does not arrive, and the prospect is recorded as contacted when they were not. That is a false entry in the one record you rely on to know who has heard from you.
2. Invalid addresses bounce, bounces feed suppression, and enough of them harm the sending reputation of the mailbox every other conversation depends on. The damage is not confined to prospecting.
3. Some guesses land — on a shared inbox, on a former employee's forwarder, on somebody with a similar name at the same company. A stranger reading a personalised approach written for somebody else is the worst outcome of the three and the least visible.

Against those, the benefit is volume: more addresses, more sends, more replies in absolute terms. That is a real benefit and this page does not pretend otherwise. Connect declines it because the record's honesty is worth more than the extra sends — a prospect list you can act on without checking is worth more than a longer one you cannot.

> **Careful** This applies everywhere, not only to discovery. An address is never generated during an import, during a re-research pass, or when an outreach sequence finds itself without a recipient.

## What Connect completed, and what it did not

**Completed** — Discovery, research, evidence capture, scoring and the written fit reason. The prospect is fully worked.
**Completed** — A contactability state that says, precisely, what is missing — rather than a blank field that could mean anything.
**Not completed** — No address was found, so no address exists on the record.
**Not completed** — No outreach was drafted or sent, and none is queued. The sequence has nothing to aim at and does not pretend to.
**Not completed** — No pattern was tested against a mail server to see whether a guess would land. Probing is not done here either.

## What to do instead

1. Look for a published contact route that is not a personal address — a switchboard number, an enquiry form, a general address the business publishes itself.
   - Result: Any of those is a real, sourced contact method, and a call or a form submission from a person is a legitimate next action.
2. Record what you find on the prospect, as a supplied value.
   - Result: It is attributed to you and does not masquerade as research, which is exactly right — you did find it.
3. Create a follow-up on the `task` channel for the person who will do it.
   - Result: No drain ever sends a `task` follow-up; it appears as work for a person on the follow-ups screen, which is the honest shape of this particular job.
4. Leave the prospect on the list.
   - Result: Businesses publish addresses they did not previously publish, and a later research pass will find one if it appears. Deleting the record throws away the research and guarantees nothing.

An administrator can also review whether the discovery brief is aimed at a kind of business that rarely publishes individual addresses. If most of a run lands in this state, the brief is selecting for it, and a phone-first approach to that segment will outperform any amount of address hunting.

## Questions

### Can I turn address guessing on?

There is no such setting. It is not a default that can be changed but a property of how contacts are handled — no code path produces a pattern-generated address, so there is nothing to enable.

### What if I supply the address myself?

Then it is used, and it is recorded as supplied by you on that date. Connect will not verify it for you, and if it bounces the resulting suppression works the same way it would for any other address.

### Does this prospect count against my allowance?

The research it consumed was booked when it ran. It uses no send allowance, because nothing is sent, and it will not quietly begin consuming one later.

## Related

- [Connect does not guess email addresses](https://connectbyjbrh.com/docs/prospects/no-guessed-email/)
- [Contactability](https://connectbyjbrh.com/docs/prospects/contactability/)
- [Keeping a good prospect you cannot email](https://connectbyjbrh.com/docs/workflows/keeping-a-prospect-without-email/)
- [Evidence-first prospecting without guessed emails](https://connectbyjbrh.com/research/evidence-first-prospecting/)
- [Finding the right person](https://connectbyjbrh.com/docs/prospects/decision-makers/)
- [Follow-ups in Connect](https://connectbyjbrh.com/docs/follow-ups/)

## What this page is based on

- `docs-source/sources/CHANNELS.md` §4 — the no-guessed-address rule
- `docs-source/sources/GENERAL.md` §7 — the `task` follow-up channel
- `backend/app/prospect_contacts.py` (capability registry entry `no_guessed_email`)
- `FORBIDDEN_CLAIMS` in `docs-source/facts.py`
