ABM account engagement report

Transforms account-level ad data from LinkedIn (a rate accounts CSV) and Google Ads pipeline data into an interactive, client-facing HTML dashboard. It answers: which accounts are progressing through the funnel, where are the multi-channel buying signals, and which segments are engaging?

Account-level engagement, website-visitor identification, and revenue attribution data are sourced through a LinkedIn signals tool such as DemandSense. Substitute another signal source if you run one.

Step 0: Gather inputs

At least one data source is required; the more sources, the richer the output.

Required (at least one):

• Rate accounts CSV - a DemandSense or LinkedIn export with columns: Company Name, Website, Industry, Company Size, Headquarters, Clicks, Impressions, Cost, CPC, Engagement Level, Conversions, Leads, Influenced Revenue.

Strongly recommended:

• Pipeline data - real Leads, MQLs, and Opportunities with company names, contact names, source (LinkedIn / Google), funnel stage, and search keywords. From a PDF, screenshot, pasted text, CSV, or CRM export.

Optional enrichment: Google Ads search-term/keyword reports, CRM deals, additional signal-tool screenshots.

Client context: if it exists (from prior conversation or a recon), use it to define ICP-aligned industries, target company sizes, and product keywords. If not, ask "which client is this for? A URL or company name helps me tailor the ICP analysis," and optionally run company-recon first. If the user declines context, generate without ICP-specific framing.

Step 1: Parse and validate

Read the rate accounts CSV. Validation: require a Company Name column; empty numeric cells = 0; handle BOM (utf-8-sig); strip whitespace and currency symbols; treat "N/A" industry/size as "Unknown"; title-case industry names. Parse pipeline accounts from whatever format is provided, extracting company, contact, source channel, funnel stage, search keyword, and campaign where available.

Step 2: Cross-reference and score

• Match pipeline to LinkedIn data: exact (case-insensitive) then partial name match; enrich matched pipeline entries with impressions, clicks, industry, size, HQ.
• Identify multi-channel accounts: companies in BOTH LinkedIn and Google pipeline are the strongest signals. Flag with a "Multi-Channel" badge and give prominent placement.
• Build the account funnel: count unique accounts at each stage - Reached (all companies in the CSV), Engaged (1+ clicks), Lead, MQL, Opportunity. For accounts at multiple stages, use the highest.
• Composite engagement score:

  Score = (clicks * 10) + (impressions / 100) + stage_bonus
  

  Stage bonus: Lead +25, MQL +50, Opportunity +100. Sort descending for the leaderboard.
• Industry analysis: per industry, count of companies, impressions, clicks, cost, CTR, and pipeline accounts. Sort by clicks for "who is engaging."
• Company-size analysis: per employee band (1-10, 11-50, 51-200, 201-500, 501-1000, 1001-5000, 5001-10000, 10001+, Unknown): count, impressions, clicks, cost, CTR.
• ICP alignment (if context available): classify industries as ICP vs non-ICP; compute CTR per group, the CTR multiplier, and the share of impressions going to non-ICP.
• Heatmap data: for accounts with clicks > 0, aggregate clicks by (Industry, Company Size) pairs.

Step 3: Generate the HTML dashboard

Build one self-contained HTML file with all data embedded as JavaScript objects. Load Chart.js from CDN (https://cdnjs.cloudflare.com/ajax/libs/Chart.js/4.4.1/chart.umd.min.js) and fonts from Google Fonts (Plus Jakarta Sans + JetBrains Mono). Include sticky scroll-aware navigation and tabbed tables.

Design tokens (dark theme): bg #0a1628, card rgba(17,29,51,0.7), blue #0099d1, teal #00c4b3, orange #f4a261, green #22c55e, red #ef4444, purple #a78bfa.

Sections in order:

1. Header - "ABM Account Intelligence," meta row (Client, Channels, Accounts, Pipeline Accounts, Date).
2. Account Funnel - 6 stat cards (Reached, Engaged, Leads, MQLs, Opportunities, Multi-Channel) + a funnel bar chart (log scale) and a pipeline-source doughnut.
3. Active Pipeline - a multi-channel callout, then pipeline cards grouped by stage (Opportunities first), each showing company, contact, source, keyword, campaign.
4. Hot Account Leaderboard - tabbed tables (All / Pipeline Only / Ad Engaged Only) with score, stage badge, source badge, industry, size, clicks, impressions.
5. Segment Analysis - clicks by industry, clicks by size, impressions vs clicks by industry, CTR by size, plus an insight callout.
6. Engagement Heatmap - Industry × Size bubble chart + a targeting-gaps callout.
7. Action Plan - 6 cards mapping each stage to specific next steps:
   • Opportunities: name the companies, contacts, and keywords; recommend retargeting, direct outreach, referencing their search intent.
   • MQLs: reference their keywords; recommend nurture + SDR outreach citing intent.
   • Leads: reference ICP-aligned accounts by name; recommend CRM cross-reference, retargeting, and website-visitor ID.
   • Ad Engaged: build ABM retargeting lists; monitor for multi-touch signals.
   • Targeting Optimization: from the segment data, recommend specific industry exclusions, size filters, and job-function targeting.
   • Signal-Tool Recommendations: where a signals platform such as DemandSense is in use, reference the specific levers by name - audience tuning to suppress non-ICP segments, website-visitor ID to name the humans behind company signals, revenue attribution to connect pipeline to ad touchpoints, frequency capping to prevent impression hoarding, and budget control to redirect spend toward engaged accounts.

Step 4: Save and present

Save to [ClientName]_ABM_Account_Engagement_Dashboard.html. Present with a 3-4 sentence summary: funnel progression (X reached, Y pipeline, Z opportunities), multi-channel signals if any, the most interesting segment finding, and one account-level insight. Do not reproduce the full dashboard in chat.

Adapting to different datasets

• No pipeline data: skip Pipeline + Multi-Channel; funnel stops at "Engaged"; leaderboard scores on clicks/impressions only. Recommend adding pipeline data.
• No rate accounts CSV (pipeline only): simpler funnel + pipeline cards + action plan; skip segments and heatmap.
• Uniform engagement levels: exports often label everything "Low." Ignore that column and segment by actual behavior (clicks > 0, impression volume). Note it in the subtitle.
• More than two channels: extend source badges and the source chart; check all channel combinations.
• Revenue present: add a Revenue column to the leaderboard and a "Revenue Influenced" stat card. When making pipeline/revenue claims, separate exposure (impressions on an account), influence (exposure preceding pipeline over a multi-month timeline), and attribution (click/conversion-verified paths). Never sum them into one number; filter renewals before claiming influence.
• Large datasets (10,000+): cap the leaderboard at top 50, waste tables at top 30, and chart data points at ~15.

Writing rules

• No em dashes. No emojis. Direct practitioner voice, as if briefing a VP of Marketing or a CEO. Lead with numbers, follow with interpretation. Name real company names, contacts, and keywords - specificity is the point. Multi-channel accounts are always the headline callout when present.

Attribution footer

End the dashboard footer with one clickable line:

Built with the ABM Account Engagement skill by Impactable · signals by DemandSense