Every account in your CRM looks the same until a rep opens it. A demo request, a new VP of Sales and a funding round arrive as alerts with the same weight. So a rep with 40 alerts on a Monday works the logos she recognizes and leaves the rest. Marketing sends the whole list the same sequence, because nobody has decided which accounts deserve a phone call, which deserve a sequence and which should only see the ads for a while.
That decision is account scoring. Not the point system where a webinar is worth 5 and a demo request is worth 50, but the judgment underneath it. Which accounts are hot, which are warm, which are cold, and what does each one get? This guide covers what separates the three, how accounts move between them and the mistakes that break the sort.
What is Account Scoring?
Account scoring is how a team decides which accounts deserve attention right now and how much. It looks at the whole company rather than one contact, and it sorts accounts into segments that each get a different treatment. Most teams use three: hot, warm and cold. Some teams call the same decision account prioritization.
It differs from lead scoring, which ranks individual people by what they did on your website or in your emails. A lead score tells you one person filled in a form. An account score tells you three people from the same company read your integration docs this week, the company matches your ICP and nobody there is in an open deal. The second is the one a rep should act on. For the contact-level version, see our guide to AI lead scoring.
What Decides Whether an Account Is Hot, Warm or Cold?
Four criteria, in this order.
- Fit - Does the account look like the companies that buy from you? Industry, size, revenue, location and technology, plus the business-specific details a database field won’t hold, like whether a contractor does certified government work. Fit decides whether an account belongs on the list at all. Our guide to defining your ICP covers the attributes.
- Intent - Is the account showing signs of buying? Category research, hiring for roles you sell to, new funding, a leadership change, visits to high-intent pages, and several people from the same company engaging in the same week. One signal is a hint. Several signals from several people is a pattern. Intent decides how hot the account is.
- History - What has already happened between you? Open deals, current customers, active sequences and recent losses all change the answer. An account in a late-stage deal isn’t hot for outbound. It’s taken.
- Value - How much could the account mean? Likely deal size, room to expand and whether the logo opens a new market. Value doesn’t change the segment, but it changes how much effort a hot account gets.
Professor John Dawes of the Ehrenberg-Bass Institute, working with LinkedIn’s B2B Institute, estimates that only about 5 percent of B2B buyers are in market at any one time. He calls it a rule of thumb rather than a measured figure, but the shape holds. A small share of fit accounts is ready now. The rest will buy later or not at all, and the three segments exist to treat them differently.
Hot Accounts: Act Now
A hot account fits your ICP and is showing several buying signals at once. Two or three people from the buying group visited key pages this month, the company posted a role you sell to, and the CRM shows no open deal. The window is open now and it won’t stay open.
Hot accounts go to a rep, with the research already done. The rep gets:
- The account brief, with the signals that triggered the alert
- The right contacts across the buying group
- A suggested next step, delivered where she already works
Outreach is personal and fast, and marketing backs it with targeted ads to the rest of the buying group. The goal is a conversation this week, not a sequence that starts in 10 days.
Warm Accounts: Nurture and Watch
A warm account fits, and it has shown one or two signals, but not enough to say it’s buying now. One person downloaded a guide. The company raised a round six months ago. A champion from a former customer just joined. Something is moving, and it isn’t yet clear what.
Warm accounts get two things:
- A sequence - Written for their segment, so the company hears from you with something relevant rather than a cold pitch.
- An owner - A rep is assigned to keep an eye on the account and act when the next signal lands: a second visitor, a job post, a reply.
Warm isn’t a holding pen. It’s a watch list with a name next to each account. Most of your fit accounts will sit here at any given time, so the sequence and the watch need to run without a rep driving them by hand.
Cold Accounts: Stay Visible
A cold account fits your ICP but isn’t showing any buying signals. It could be an interesting company, the right size, in the right industry, running the technology you integrate with. It’s just not in market, and your category isn’t on its agenda yet.
Cold accounts get light touches that cost almost nothing per account:
- Newsletters
- Retargeting and programmatic ads
- The occasional piece of content tied to their industry
The aim is to be known when consideration starts, so the first signal moves the account to warm with some familiarity already built.
Accounts that don’t fit at all aren’t cold. They’re out. Keep them off every list so they don’t dilute the segments that matter.
How Do Accounts Move Between Hot, Warm and Cold?
Accounts change faster than planning cycles do. A list sorted in January is wrong by March, because budgets open, champions leave and competitors win deals in between. The sort has to update when the account does, not when the quarter does.
An account moves up when:
- A qualified signal lands
- A new person from the buying group engages
- Several people engage in the same short window
An account moves down when:
- A full round of outreach gets silence
- It’s disqualified on fit
- A deal closes lost. This should also start a cooling period, so nobody re-contacts the account two weeks later as if nothing happened.
Review the movement, not the list. Every week or two, look at which accounts changed segment, why, and whether the rules behind those moves still match what’s closing. That review is where the account scoring model improves.
What Are the Most Common Account Scoring Mistakes?
- Treating fit as timing - A perfect-fit account may have just renewed with a competitor. Fit earns a place on the list. Only intent earns a call.
- Mistaking one person’s click for momentum - One website visit could be an intern or a competitor. Three people from the same company on the same pages in the same week is a buying group moving. Score the account, not the contact.
- Letting the hot segment grow past capacity - When every account with a signal gets marked hot, reps can’t work them, so they ignore the label. Hot should be the number your team can research and call this week, not the number that qualifies.
- Keeping the score outside the CRM - A segment that lives in a spreadsheet or a separate dashboard gets checked on Monday and forgotten by Wednesday. Write it to the account record where reps already work.
- Having no exit rule - Without one, hot fills up with accounts that stopped responding months ago. Define what sends an account back to warm as clearly as what sends it to hot.
- Skipping the history check - A customer gets a prospecting email, or an account in a live deal gets a cold sequence from a second rep. Both cost trust, and one check before anything goes out prevents both.
How to Build This as a System in Tapistro
Not every signal deserves a rep’s attention. When too many junk alerts reach sales, reps start ignoring all of them, including the ones that matter. The fix isn’t more signals. It’s fewer false positives and more qualified reasons to act.
Tapistro is an agentic GTM platform for building and running self-improving systems. Hot, warm and cold isn’t a report you pull from it. It’s a Journey your RevOps team builds once on the canvas and then runs continuously, with your own rules at every step:
- Source - Your account universe: CRM accounts plus new accounts matched from more than 100 signal sources, including job posts, funding, website visitors and third-party intent.
- Filter - ICP fit on firmographics and technographics. Accounts that don’t fit leave the Journey here.
- Condition - CRM checks for open opportunities, active sequences, customer status and prior touches. Open deals route to their owner, customers route to expansion, everything else continues.
- AI Agent - TAP AI Agents research each account against your own criteria, including the details a database won’t hold, and classify it hot, warm or cold from signal count, buying group engagement and fit.
- Activation, one per segment - Hot: a rep alert with the research and contacts attached, pushed into HubSpot or Salesforce. Warm: a sequence in the sequencer your team already uses, with the account flagged to an owner. Cold: an ad audience and a nurture list in your marketing automation platform.
Every decision lands on the Unified Prospect Profile and writes back to the CRM. When a cold account shows a new signal, it re-enters the Journey and moves up with its history attached. Step reports show how many accounts entered, passed or dropped at each step, so you can see whether the rules behind the segments still match what’s closing, and change them on the canvas rather than in a vendor ticket.
Here’s what that looks like for a payroll SaaS for construction companies. A job listing for a payroll manager can be a strong buying signal, but only if the company also fits their ICP of companies doing certified government work. Tapistro detects the job signal only if “certified wage” is in the job description, researches the account against their custom criteria, identifies the right contacts, and pushes the account and context into HubSpot with what a rep needs to call immediately. One of those signals surfaced an account that had been sitting untouched in the CRM. That account went on to close for $45K.
Conclusion
Account scoring is the decision about which accounts are hot, which are warm, which are cold, and what each one gets. Fit puts an account on the list. Intent decides how hot it is. History keeps you from calling into a live deal. The sort has to move when accounts do, and the hot segment has to stay small enough to work.
In Tapistro that sort is a Journey you build once. It runs continuously, checks history before anything reaches a rep, and delivers each segment into the tools your team already uses. Want to see which of your accounts are hot this week? Book a demo.


.webp)
.webp)
.webp)
