Sales & Revenue path
Free starter lessonChatGPT16 min

Map your selling hours and lock your ICP

By the end you will have logged where your selling hours actually go, named the one task to systematize first, and written a filter-ready ICP anchored in your real closed-won accounts.

Every downstream tool in this path, the Apollo list, the Clay enrichment, the sequence, inherits whatever you define here; a sloppy ICP means you enrich and email the wrong companies at scale. Sixteen minutes spent naming your highest-leverage task and writing a concrete ICP is the cheapest quality gate in the whole pipeline, because a filter is only as good as the profile behind it.

This is the free first module of urfired.ai's Sales & Revenue path. No signup, no card. The steps note what ChatGPT access you need.

The workflow

8 steps, start to finish

Set up a reusable ChatGPT project for this

In the left sidebar, create a new project and name it something like "Pipeline Engine." Open its settings and add project instructions: one paragraph on what you sell, your rough average deal size, and the market you cover, so every chat in the project starts with that context instead of a cold prompt. Projects, custom instructions, and file uploads are available on the free tier.

If you do not see Projects in your account, open one dedicated chat and paste the same context as the first message.

Dump your last five selling days into a time audit

In a new chat inside the project, list how you actually spent your working hours across your last five days: rough blocks and minutes, in plain language, no formatting needed. Ask ChatGPT to sort those blocks into categories like prospecting and list-building, writing outreach, live calls and meetings, CRM and admin, and forecasting, then total the hours in each.

You now have a real picture of the week, not a guess.

Name the single task to systematize first

Ask ChatGPT to rank your categories by total hours, then mark which are low-value and repetitive versus high-value and human. Tell it which task you personally dread most and have it cross-reference that against the hours.

Commit to one task, most likely list-building or writing outreach, as the single thing you will systematize first, and write that one sentence down.

Feed it your ten best closed-won accounts

In the same chat, list ten real accounts you have closed that you would happily clone: company name, rough headcount, industry, and what was going on when they bought, such as a funding round, a new leadership hire, or a tool they already ran. Ask ChatGPT to find the firmographic and behavioral patterns those winners share.

This is how you build an ICP from evidence instead of from a wish list.

Draft the ICP across the five fields that become filters

Ask ChatGPT to write your ICP as five specific fields: buyer title, headcount range, industry, funding stage, and tech stack. Force specificity; "B2B SaaS, Series B to D, 100 to 2,000 employees, running HubSpot" is usable, "tech companies" is not.

These five fields map directly to the filters you will set in Apollo.io in the next module, so keep every one of them concrete and searchable.

Add your disqualifiers

Ask ChatGPT to list negative indicators: the attributes that should exclude an account even when it otherwise fits, such as under 100 employees, no CRM in place, or a business model you cannot serve. A clean exclude list keeps your target list from filling with accounts you will never close.

Keep it to three to five hard rules you can actually apply as filters.

Pressure-test against a deal you lost

Give ChatGPT two or three real closed-lost accounts and ask whether they look different from your closed-won list on the five fields. If lost and won look identical, your ICP is missing a layer, so add a technographic or behavioral signal until the two groups separate.

Tighten the ranges based on what actually distinguishes the wins.

Lock the one-pager into the project

Ask ChatGPT to output the final artifact as one page: the time audit with the single task named at the top, then the ICP five fields, then the disqualifiers. Paste that page into your project instructions or save it as a project file so every downstream chat reuses it.

This one page is what drives every filter in the modules that follow.

Do this now

Using your own numbers, run the full pass inside one ChatGPT project: log your last five selling days, name the one task to systematize first, feed it ten real closed-won accounts, and have it draft an ICP across buyer title, headcount range, industry, funding stage, and tech stack, plus a short disqualifier list. Real accounts and real hours only, no hypotheticals.

What you keep

A one-page artifact held in a ChatGPT project: a time audit that names the single task to systematize first, plus a written ICP across buyer title, headcount range, industry, funding stage, and tech stack, with a short disqualifier list, ready to drive your Apollo.io filters in the next module.

What good looks like
  • Each ICP field is specific enough to paste straight into a list-building filter: ranges and named tools, not "mid-market tech."
  • Your closed-won and closed-lost examples look measurably different on at least one field.
  • The time audit points to exactly one task to systematize first, chosen from logged hours rather than a hunch.
  • The one-pager lives in the project, so your next chats start from it instead of a blank prompt.
Common traps
Writing a vague ICP like "tech companies that need our product."
Force every field into a filterable value: a headcount range, named funding stages, and specific tools in the stack, so it maps cleanly onto Apollo filters.
Picking the task to systematize by feel instead of from the audit.
Rank the categories by real logged hours first, then choose; the biggest low-value block wins, not the one that annoyed you today.
Building the ICP from who you wish you sold to.
Anchor it in ten real closed-won accounts and pressure-test against closed-lost, so the profile reflects who actually buys.
Saved on this device
Next in this path

Module 02: Build a targeted account list in Apollo.io (Apollo.io, 22 min)

Free gets you this starter lesson today. Pro unlocks the full 14-module path when checkout opens, no card and no charge until then.

New here? The free 2-minute assessment reads your role and gaps, then matches you to the right path. Get your score