Workflow · Account Scoring
Find the 50 accounts
ready for outreach this week.
A copy-paste Claude prompt that pulls HubSpot account data and LinkedIn ad-viewer signals, scores every account against six signal types, and returns a ranked top-50 list with role-by-role engagement and recommended next moves. Run it Monday morning, work the list all week. No third-party intent platform required.
- Multi-signal density scoring
- 6signalsMulti-signal density scoring
- Top-ranked weekly output
- 50accountsTop-ranked weekly output
- Pipeline + net-new accounts
- 2categoriesPipeline + net-new accounts
- First-run setup time
- 20minFirst-run setup time
01The Problem in 60 Seconds
Single-signal scoring is too noisy.
Six signals is the right unit.
A company that downloaded one whitepaper might be a buyer or might be a researcher who never returns. A company where three stakeholders engaged with three different signals in the same week is almost certainly in active evaluation. Multi-signal density is how you separate accounts that look interested from accounts that are committee-shaped buying.
Most B2B SaaS teams either rely on lead scoring (one contact at a time, behavioral only) or pay $30K–$100K per year for an ABM platform whose scoring is opaque enough that nobody trusts it. Neither approach answers the actual operating question: which 50 accounts should I work this week?
This workflow scores every account in your HubSpot and LinkedIn accounts against six signal types, weights them by predictive strength, and returns a ranked top-50 with the role-by-role engagement breakdown for each. Refresh weekly. Work the list. The signals you already own are usually enough. You just need a prompt that reads them well.
| 01Multi-stakeholder engagement (3+ roles) | 25% |
|---|---|
| 02HubSpot lifecycle to MQL/SAL | 20% |
| 03High-intent page visits | 18% |
| 04Sales touch history | 15% |
| 05LinkedIn ad viewer + ICP match | 12% |
| 06Content downloads | 10% |
02The Prompt
Copy this prompt into
Claude Desktop.
The gold variables (your ICP criteria, signal weights, output count) are the parts you edit. Default weights work well for most B2B SaaS but adjusting based on your conversion data improves scoring after 90 days.
RoleYou are scoring my B2B SaaS accounts on buying intent. The goal is a ranked top-N list of accounts ready for outreach this week, scored on six signal types from HubSpot and LinkedIn Ads. Multi-signal density matters more than any single signal, accounts firing 4+ signals are dramatically more likely to convert than accounts firing 1. ICP ParametersTarget industries: [B2B SaaS, Software, IT Services, FinTech] Target company size: [51–5,000 employees] Target seniority floor: [Manager and above] Geography: [US, Canada, UK, EU, Australia] Excluded segments: [Education, Government, Non-profit, Sub-50 employee companies] Signal Weights1. Multi-stakeholder engagement (3+ roles in same account engaging in last 30d): 25% 2. HubSpot lifecycle transition to MQL or SAL in last 60d: 20% 3. High-intent page visits (pricing, demo, integrations) in last 30d: 18% 4. Sales touch history (replied to outreach, took a meeting): 15% 5. LinkedIn ad viewer with ICP match in last 30d: 12% 6. Content downloads or form fills in last 30d: 10% Task1. Pull all active companies from growthspree-mcp hubspot connector: with engagement data, lifecycle stage, contact count by role, sales activity timestamps. 2. Pull LinkedIn ad viewer companies from growthspree-mcp linkedin_ads connector for the last 30 days. Include account-level frequency and demographic match data. 3. For each account, evaluate the six signals: Signal 01: Count distinct role types (researcher, manager, security, finance, decision-maker, influencer) with engagement in last 30d. 1 role = 0 points. 2 roles = 50 points. 3+ roles = 100 points. Signal 02: Did the account transition to MQL or SAL in last 60d? Yes = 100. Recent regression = 0. No transition = 0. Signal 03: Count pricing, demo, or integration page visits in last 30d. 0 = 0. 1–2 = 50. 3+ = 100. Signal 04: Did sales touch the account in last 60d AND get a reply? Yes = 100. Touch but no reply = 30. No touch = 0. Signal 05: LinkedIn ad views in last 30d AND ICP demographic match? Yes = 100. Views but mismatch = 0. No views = 0. Signal 06: Content downloads or non-demo form fills in last 30d? 1 = 50. 2+ = 100. 0 = 0. 4. Compute total weighted score for each account: sum(signal_score × weight). 5. Rank all accounts. Return top 50. 6. Split the top-50 into two categories: Pipeline accounts, already in HubSpot with sales activity Net-new accounts, LinkedIn ad viewers not yet in HubSpot, ICP-matched Output format1. Headline: number of accounts scoring above the "ready for outreach" threshold (75+), top score, biggest signal-density jump from last week if known 2. Top 50 ranked list: account name + category (pipeline / net-new) + active signals (as pills) + total score 3. Role engagement summary: how many top-50 accounts have 3+ roles engaged vs single-role 4. Three recommended actions: (a) which highest-scoring pipeline accounts to escalate to senior AE today, (b) which net-new accounts to add to LinkedIn matched audience for committee retargeting, (c) which mid-score accounts (50–75) to re-engage with sequenced creative // If LinkedIn Ads connector returns fewer than 50 ad-viewer companies, signal 05 becomes weak, surface this as a calibration recommendation. // If HubSpot has no defined "Manager+" seniority field, fall back to email domain + job title text matching. // Net-new accounts that score 75+ are the highest-leverage opportunities most teams miss, flag them prominently.
Two connectors required. This prompt uses hubspot and linkedin_ads via the free Growthspree MCP. If both are connected, the full scoring run takes 90–120 seconds for accounts with 500+ active companies in HubSpot.
03What Claude Returns
A ranked list. A signal density map.
Sample weekly output for a B2B SaaS company spending $15K/mo on LinkedIn with 800 active accounts in HubSpot. Real structure, illustrative names. Notice the mix of pipeline accounts and net-new ad-viewer accounts.
Claude · via Growthspree MCP
Pulling account data and computing signal density now.
growthspree.hubspot.get_active_companies { engagement_window: 30d, include_contacts: true } growthspree.hubspot.get_company_activity { window: 60d, dim: [lifecycle, page_views, sales_touches] } growthspree.linkedin_ads.get_audience_reach { window: 30d, dim: company_domain } Scored 800 accounts. 73 are above the ready-for-outreach threshold (score 75+). The top 50 are ranked below, split between pipeline accounts already in HubSpot and net-new ad-viewer accounts your CRM doesn't yet know about.
| Account | Active Signals | Score |
|---|---|---|
| 1Acme CorpPipeline · MQL · 4 roles | 3+ roles MQL pricing visit SDR reply linkedin viewer | 94 |
| 2Northbridge HealthPipeline · SAL · 3 roles | 3+ roles SAL demo visit 2 downloads | 87 |
| 3Vertex Capital SoftwareNet-new · LinkedIn | 3+ roles linkedin viewer pricing visit 1 download | 82 |
| 4Helix LabsPipeline · MQL · 2 roles | 2 roles MQL pricing visit SDR reply | 81 |
| 5Quartile SystemsNet-new · LinkedIn | 2 roles linkedin viewer demo visit | 79 |
| 6Pellion IndustriesPipeline · MQL · 3 roles | 3+ roles MQL 2 downloads | 78 |
| 7Lattice NetworksNet-new · LinkedIn | 2 roles linkedin viewer pricing visit | 77 |
| 8Meridian RoboticsPipeline · SAL · 2 roles | 2 roles SAL SDR reply | 76 |
| 9Praxis LogicPipeline · MQL · 3 roles | 3+ roles MQL linkedin viewer | 76 |
| 10Cinder MaterialsNet-new · LinkedIn | 2 roles linkedin viewer demo visit | 75 |
Pipeline vs Net-New split. Of the top 50, 31 are pipeline accounts (already in HubSpot) and 19 are net-new from LinkedIn ad views. The net-new accounts are the highest-leverage discovery, most teams never know these companies are interested because they never reached the CRM. Adding them to a coordinated multi-channel motion this week is the single biggest upside in the report.
Role density. 22 of the top 50 have 3+ roles engaged, the strongest committee-formation signal. These are the accounts where a buying conversation is already happening internally. The other 28 are 1–2 role accounts where you're still landing on a single stakeholder.
- 01
Escalate the top 5 pipeline accounts to a senior AE today. Acme Corp, Northbridge Health, Helix Labs, Pellion Industries, Praxis Logic, all scoring 78+ with multi-role engagement and active sales motion. The committee is already forming; the right next step is a senior conversation, not another email sequence.
- 02
Build a LinkedIn matched audience from all 19 net-new accounts. These companies viewed your ads, match ICP, and are not yet in HubSpot. Run sequenced retargeting with three role-targeted creative variants for the next two weeks. Trigger SDR research on the highest-scoring net-new accounts (Vertex Capital Software, Quartile Systems, Lattice Networks).
- 03
Re-engage the 23 mid-score accounts (50–74) with content sequencing. These accounts have 1–2 signals firing, interested but not yet committee-formed. Add to a 4-week LinkedIn nurture with case-study creative aimed at expanding internal champions. Half typically move to score 75+ within 30 days with the right cadence.
04Setup
Four steps. Twenty minutes first time.
First run only. Every weekly run after that takes under 2 minutes, paste the prompt, get the list.
-
Install · 4 min
Install the Growthspree MCP
Head to
Install nowgrowthspreeofficial.com/mcp. Authorize both HubSpot and LinkedIn Ads through the OAuth flow. Read-only on both, the workflow only reads data, never writes. -
Verify · 1 min
Confirm both connectors are green
Open Claude Desktop. Click the tools icon. You should see
growthspree-mcpwith bothhubspotandlinkedin_adsshowing green. If either is red, re-run that platform's OAuth flow. -
Configure · 5 min
Set ICP criteria and signal weights
Copy the prompt from section 02. Edit the gold variables to match your ICP: target industries, company size, seniority floor, geography, exclusions. Default signal weights work for most B2B SaaS, but if you have 90+ days of conversion data, calibrating weights against your closed-won pattern is recommended. The biggest tune is signal 01 (multi-stakeholder), companies with longer sales cycles and bigger committees should weight this higher.
-
Run weekly · 10 min
Make this the Monday morning ritual
Run the prompt every Monday morning. The output is the focus list for the week: top 5 pipeline accounts get senior AE attention, net-new accounts get added to LinkedIn matched audiences, mid-score accounts get nurture sequencing. Save each week's output as a markdown file to track week-over-week movement and which accounts are climbing or falling.
05Prompt Variations
Three ways to cut the same data.
Same six-signal foundation, different framing. Pick the one that matches what you're trying to decide this week.
-
01 / Top 10 instead of top 50
Tighten to the highest-leverage 10
For small teams or weekly all-hands stand-ups, the top 10 list is more actionable. Same scoring logic, smaller surface. Easier to review per-account in a 30-minute meeting.
TweakChange the output count parameter from 50 to 10. Append: "Include a one-line account summary per account with the most likely buying motivation based on engagement pattern."
-
02 / Stalled accounts overlay
Surface accounts that were hot but went quiet
Add a parallel "stalled" list, accounts that scored 75+ in any of the last 4 weeks but dropped below 60 this week. Often the highest-ROI re-engagement targets because the committee was forming and may need one more push to reactivate.
TweakAppend: "Also output a 'stalled accounts' section: accounts that scored 75+ in any of the last 4 weeks but are below 60 this week. Show their highest score, current score, and what changed."
-
03 / Vertical-segmented scoring
Run separate top-50s per vertical
If you sell into multiple verticals (e.g., HR Tech, FinTech, DevTools), the right cadence is one top-50 per vertical. Different signal patterns, different velocity, different committees.
TweakAdd: "Segment the output by industry. Return a top-25 per vertical instead of one top-50. Compare velocity and signal density across verticals."
FAQ
Quick answers on the scoring engine.
What does this account intent scoring engine do?
It pulls account data from HubSpot and LinkedIn Ads, scores every active account against six signal types (LinkedIn ad viewers, high-intent page visits, content downloads, lifecycle transitions, multi-stakeholder engagement, sales touch history), weights them based on B2B SaaS conversion data, and returns a ranked top-50 list of accounts ready for outreach this week. Each account in the output includes signal density, which roles inside have engaged, and the recommended next move.
Why score accounts on six signals instead of a single signal?
Single-signal scoring is too noisy. A company that downloaded one whitepaper might be a buyer or might be a researcher who never returns. A company where three stakeholders engaged with three different signals (ad view, pricing visit, content download) is almost certainly in active evaluation. Multi-signal density is how you separate accounts that look interested from accounts that are committee-shaped buying. The six signals each capture a different intent dimension, and accounts in the top decile typically fire 4–5 of them.
How does this differ from generic ABM platforms like Demandbase or 6sense?
Generic ABM platforms cost $30K–$100K per year and rely heavily on third-party intent data, companies you don't have any first-party relationship with, scored on opaque algorithms. This workflow scores only accounts that have firsthand engaged with you (visited your site, viewed your ads, filled a form, took a sales call). The signal is cleaner because the engagement is real. The accounts that score high here are accounts where a buying conversation has actually started, which is what most teams want from an intent platform but rarely get.
What signal weights does the prompt use by default?
Default weights ranked by predictive strength based on B2B SaaS conversion data: multi-stakeholder engagement (3+ roles in same account engaging) carries the highest weight at 25%, followed by HubSpot lifecycle transition to MQL/SAL at 20%, high-intent page visits (pricing, demo, integrations) at 18%, sales touch history at 15%, LinkedIn ad viewer + ICP match at 12%, and content downloads at 10%. The prompt accepts custom weights, adjusting based on your conversion data is recommended once you have 90 days of pipeline history to calibrate against.
How often should I run this scoring engine?
Weekly is the right cadence. Account intent moves fast in B2B SaaS, a company that wasn't on the list last week can fire 3+ signals this week and jump into the top 50. Monthly is too slow; daily creates noise that overwhelms outreach capacity. The Monday morning weekly run produces a fresh top-50 that becomes the focus list for the week's coordinated paid + SDR outreach.
What if my HubSpot lifecycle stages aren't well-defined?
The scoring works with whatever lifecycle definitions you have, but the output quality depends on those definitions being meaningful. If your HubSpot has a single 'lead' stage with no progression beyond MQL, the lifecycle signal becomes weak and other signals end up doing more of the work. The prompt handles this gracefully by reweighting toward the available signals, but the recommendation is to clean up lifecycle stages first if they haven't been defined recently. The B2B SaaS pipeline stages framework is a good starting point.
Does this workflow work for accounts I don't yet have in HubSpot?
Partially. The LinkedIn Ads viewer signal surfaces companies that engaged with your ads but never reached HubSpot. For those companies, the score is based on LinkedIn-only signals: ad views, demographic match to ICP, account-level frequency. They show up in the output as 'Net-new accounts' separated from 'Pipeline accounts,' so SDR outreach can prioritize differently. This is one of the highest-value parts of the workflow, most teams have hundreds of net-new ICP-matched companies showing intent that never made it into the CRM.