Account Selection Criteria: How to Choose Target Accounts

What account selection criteria actually are

Account selection criteria are the weighted, scored rules a sales team uses to pick the 50 to 75 logos a rep will work in one cycle. They translate an ideal customer profile into a ranked list with a play assigned to each band. Skip them and a rep ends the quarter with 400 untouched accounts and a flat pipeline.

Direct answer. Account selection criteria are a weighted score across six lenses: firmographic fit, technographic fit, buying signals, capacity, access, and strategic value. The Gangly 6-Lens Account Fit Score buckets each account into A, B, C, or D and assigns a play per band. A score above 75 earns a multi-thread motion. Anything below 35 is parked until the next signal refresh.

The list is the second-most leveraged artifact a B2B rep owns, after the deal-stage definition. Get the criteria right and one hour of prospecting earns five conversations. Get them wrong and ten hours of prospecting earn none. This guide ships a six-lens scoring model, a build sheet, and the five mistakes that quietly kill account lists. For the full account-based selling playbook, see the cluster pillar.

Why most target account lists fail in the first 30 days

Most target account lists fail because they confuse the universe with the list. A reasonable ICP filter still returns thousands of logos. A rep who tries to work 400 accounts books fewer meetings than a rep who works 60, because depth of research is what triggers a reply.

The pattern repeats: lists that win are short, ranked, and refreshed on signal. Lists that lose are long, flat, and frozen for a quarter. The fix is a scoring rubric that forces a number on every account and a band that forces a play.

The 6-Lens Account Fit Score: the Gangly framework

The 6-Lens Account Fit Score is a 100-point rubric that grades every account on six dimensions, applies a weight, and lands a band. The score is recomputed every two weeks because signal moves faster than firmographic data.

Apply the weights, sum to 100, and bucket. The score is meaningless without a play attached to each band, so the second table fixes the play.

Verdict. The 6-Lens Score works because it forces a number on every account and a play on every band. Reps stop debating whether a logo is worth working. The rubric decides. Pipeline coverage improves because the list reflects what actually closes, not what looks good on a slide.

How to set the firmographic floor and ceiling

The firmographic floor is the minimum company size, revenue, and industry your product can serve. The ceiling is the upper bound where deal complexity or procurement overhead destroys margin. Together they define the universe the other five lenses score inside.

Pull closed-won data for the last four quarters. Plot company size against win rate and average contract value. The floor is where win rate crosses 15%. The ceiling is where sales cycle exceeds your average plus 50%. Cut everything outside. The HubSpot 2024 sales benchmark found teams that set explicit firmographic ceilings closed 18% faster than teams that took every inbound (HubSpot, 2024).

Industry filters need a "served" list and an "exclude" list. The served list captures the verticals where your product has at least three case studies. The exclude list captures industries where you have lost twice in the last year for product or compliance reasons. Excluding losses is harder than adding wins and matters more.

For a deeper read on capacity benchmarks, see the sales pipeline management guide and the sales pipeline definition. Both anchor the math on what coverage a healthy firmographic floor produces.

Technographic and tech-stack fit signals that predict pipeline

Technographic fit predicts pipeline because the tools a buyer already runs reveal both the problem they admit and the budget they release. A company running HubSpot Marketing and Outreach is buying outbound. A company running Salesforce and nothing else is buying CRM data quality.

Grade technographic fit on three sub-signals. Stack overlap with your integrations earns full points because deployment is fast. Presence of a rival earns half points because the buyer knows the category but the cycle is competitive. No relevant tool earns a quarter point because education is required. BuiltWith reports stack data on roughly 712 million sites as of 2026 (BuiltWith, 2026), so coverage on mid-market and enterprise logos is high.

Reps who skip the technographic lens are flying blind. The Gong 2025 revenue intelligence report found AEs that opened first calls with a referenced stack item earned a follow-up meeting 1.7x more often than AEs that did not (Gong, 2025). The stack is the cheapest research signal you can use.

• Stack overlap (full points). Buyer runs at least one tool your product integrates with natively.
• Rival present (half points). Buyer runs a direct competitor; the cycle is winnable but competitive.
• Stack gap (quarter points). Buyer runs neither; education burns three extra touches.
• Anti-fit signal (zero points). Buyer runs a tool incompatible with yours; cut the account.

How to weight buying signals inside the score

Buying signals weight the score most heavily because they predict timing. A perfectly fit account that has shown no signal in 90 days is a slower deal than a B-fit account that just raised a Series B and posted three sales engineering roles last week.

Inside the score, weight signals at 25% by default. Higher for signal-led teams, lower for teams with no intent feed. The trick is to use the surge delta, not the raw value. Bombora intent topics on a 0 to 100 scale are noisy. A jump from 40 to 78 in one week is what predicts a 30-day cycle. A flat 85 predicts nothing.

1. 1

   Funding rounds (last 90 days)

   Series A through D events open budget and force vendor selection within two quarters.

2. 2

   Leadership change in the buying role (last 60 days)

   New VP Sales, CRO, or RevOps Director typically reviews the stack inside 90 days of joining.

3. 3

   Hiring posts that match the problem (last 30 days)

   Three or more sales-related job posts signal pipeline pressure and a budget you can ride.

4. 4

   Intent surge on category terms (last 14 days)

   A 30-point delta on Bombora or G2 buyer intent inside two weeks is the most predictive single input.

5. 5

   Product launch or M&A event (last 60 days)

   Both events expose stack gaps and create urgency around new sales motions.

For the deeper signal taxonomy, see the signal-based selling guide and the signal-based selling glossary entry.

Capacity, intent, and access: the three filters reps forget

Capacity, intent, and access are the three filters reps skip most often and the three that quietly destroy pipeline quality. Capacity asks whether the deal is large enough to matter. Intent asks whether the account will buy in this fiscal year. Access asks whether a rep has any warm path at all.

Capacity math is the easiest of the three. Multiply your average contract value per seat by the company headcount that maps to your buyer role, then divide by your target ACV. If the number is less than one, cut. Outreach the company found in 2024 that AEs spent 22% of their sequenced touches on accounts that could never reach minimum ACV (Outreach, 2024). Capacity scoring fixes that.

Set capacity at 10% and access at 15% by default. Reps with strong networks should weight access at 20%. New hires should weight access at 5% until the network builds.

How to build the scoring sheet in 45 minutes

A scoring sheet ships in 45 minutes with a spreadsheet, a CRM export, and one signal feed. The version that runs forever takes a quarter. Ship the 45-minute version first.

1. 1

   Pull the firmographic floor and ceiling

   Filter the universe to industries, revenue bands, geographies, and headcount ranges that match your closed-won data from the last four quarters. Cut anything outside.

2. 2

   Score technographic fit

   Run BuiltWith or Clearbit Reveal across the filtered list. Mark accounts that run a rival, accounts that run a complementary tool, and accounts that show no stack at all.

3. 3

   Pull live buying signals

   Bring in hiring posts, funding events, leadership moves, product launches, and intent surges from the last 60 days. Tag each account with the strongest signal.

4. 4

   Add capacity, access, and strategic value

   Annotate team size, your warm paths, and any strategic angle. Reps often skip access. Do not.

5. 5

   Apply the weighted score

   Multiply each lens by its weight, sum, and bucket into A, B, C, D. Rank inside each band by the strongest signal so the top of A is obvious.

6. 6

   Lock the list and ship the play

   Freeze the top 50 to 75 for the cycle. Assign a play per band: A gets multi-thread, B gets a sequence, C gets one shot, D gets parked.

The sheet needs eight columns: account name, firmographic score, technographic score, signal score, capacity score, access score, strategic-value score, and weighted total. A ninth column captures the play assigned. Rank-sort by weighted total. Lock the top 75 for the cycle. Re-score every two weeks, but only let the bottom of B and the top of C move. Rotating A or D weekly destroys focus.

For teams running the workflow inside Gangly, the same score appears against every account in the Signal Detection view and refreshes nightly. Reps stop maintaining a side spreadsheet inside the second week.

The five mistakes that wreck an account list

Five mistakes account for most failed lists. Each is fixable in a single working session. None require new tools.

1. 1

   Picking the universe instead of the list

   A reasonable filter still returns 5,000 logos. That is a universe. Cut to a target list of 50 to 75 you will work with intent inside one cycle.

2. 2

   Weighting firmographics at 50% or more

   Firmographic match is necessary, not sufficient. Closed-won data from Gong shows fit-only lists convert at 1.2% while fit-plus-signal lists convert at 4.8% (Gong, 2025).

3. 3

   Treating intent as a binary

   Bombora intent topics range across a 100-point scale. Use the surge delta, not the raw score. A jump from 40 to 78 in a week is more predictive than a static 85.

4. 4

   Ignoring capacity math

   A 12-person company will not buy 40 seats of a $299 product. Run the seat math before the list locks or pipeline will be qualified out by AEs in week one.

5. 5

   Refusing to demote

   A B account that ghosts five touches is now a D account. Lists that never demote bloat to 800 logos by quarter end and reps stop trusting them.

The pattern across all five mistakes is the same: refusing to commit a number. A list without numbers cannot demote, cannot rank, and cannot answer the question "why this 50 and not those 50". Force the number even when the data is thin. Refine in cycle two.

For the team-level view on how this connects to coverage and quota, see the account-based selling metrics guide and the sales pipeline management playbook.

How Gangly fits account selection

Gangly runs the 6-Lens Account Fit Score as a live workflow, not a spreadsheet. Signals refresh nightly. The score recomputes. The band shifts. The rep gets a queue of accounts ranked by today's score, with the play preloaded and the research preassembled. The Sales Workflow System keeps every input visible so the score is auditable and the rep trusts the rank.

• Signal Detection — Pulls funding, hiring, leadership, product, and intent signals against every named account, recomputes the score nightly, and surfaces accounts that crossed a band boundary.
• Workflow Sequencer — Assigns the right play per band so A accounts get multi-thread, B accounts get a sequence, and C accounts get a one-shot test.
• Call Prep Engine — Bundles the research dossier for every A account so the rep walks into a discovery call with the stack, signals, committee map, and three open questions.
• CRM Hygiene — Writes every score input back to the CRM so RevOps can audit weights against closed-won data and tune the model quarterly.

Teams running the loop end to end cut list-build time from 4 hours per week to 25 minutes, and lift A-band conversion by 28% in the first two cycles (Gangly customer benchmark, 2026). The score is only as good as the team that trusts it. The product makes trusting it cheap.

Frequently asked questions

Common questions reps and managers ask when first setting account selection criteria. The FAQ accordion above answers each in 60 to 120 words.