The Intelligence Layer: The Root of Continuous Improvement & Campaign Design

Every ABGtM campaign - new logo, expansion, or retention - is only as good as the inputs that designed it. Most fall short not because of poor execution but because the design was underclubbed: too much intuition, too little market signal, and almost no systematic use of what's already sitting in existing systems. Commonly this occurs in the interest of speed…but with modern system architectures & AI capabilities, the right system can actually function faster than the analog solution.

This collection of data sources - both Outside-in & Inside-out - comprise the system. The intelligence layer. Fourteen sources across two dimensions that, when consistently mined, produce better campaign design, sharper targeting, more resonant messaging, and offers that actually move buyers.

Outside-In: What the market, the customer, and the competition are telling you

These nine sources bring external signals into the building. Most companies are underweight here - a few rep-driven win/loss notes (terrible - see my post on how win/loss should actually work) and a quarterly NPS score don't constitute a listening infrastructure. Done well, Outside-In inputs tell you how the market sees you vs. competitors, the factors that actually drive decisions, and where the white space is. Functional ownership denoted in italics.

01 Customer & Prospect Interactions (Sales, CS/AM)Conversations across the full funnel - discovery, demos, onboarding, EBRs, renewals - on calls, over email, in shared Slack channels. The richest qualitative signal most companies never systematically capture. Source systems: Gong, Microsoft, Slack.

02 Win-Loss Reviews (Revenue/RevOps, PMM)Structured debrief of won and lost deals: messaging resonance, competitive dynamics, offer fit, timing, decision-maker behavior - conducted by a second or third party. Most companies have a win/loss process in name only - rep-driven, inconsistent, and biased toward "price." A formal, repeatable program changes that. Source systems: Clozd, Primary Intelligence.

03 CSAT / NPS (CS)Quantified satisfaction and loyalty data, plus sentiment analysis of verbatim responses. Segments accounts by health, flags at-risk relationships, and surfaces promoters most likely to expand or advocate. Source systems: Qualtrics, Gainsight, Delighted.

04 Market Surveys (PMM, Marketing)My favorite. Also the least leveraged. Structured research fielded to customers and prospects to validate positioning, test offer concepts, and understand evaluation criteria - benchmarked against competitors to track relative category perception over time. Source systems: GLG, NewtonX. More like a source capability than a specific system.

05 Social Scraping (PMM)Monitoring of G2, Capterra, Reddit, LinkedIn, and other category-specific channels for unfiltered product feedback, competitor comparisons, and emerging buyer language. Customers tell each other things they won't tell you. Source systems: Reddit, G2, Capterra, LinkedIn.

06 3P Signal Efficacy (Revenue/RevOps)Analysis of which third-party intent signals - job postings, funding rounds, tech installs, leadership changes, financial announcements - actually correlate with conversion or expansion in your market. Used to tune scoring models rather than treat all signals as equal. Source systems: Bombora, G2, ZoomInfo, LinkedIn, 6sense.

07 Customer Advisory Board (Product, PMM, CS)Structured input from a curated group of strategic customers on product direction, market trends, messaging, and competitive positioning. High-trust, high-signal…and chronically underutilized beyond product feedback. Source systems: Meeting notes.

08 Expert Interviews (PMM)Conversations with analysts, former practitioners, or category specialists to pressure-test strategy, benchmark against market norms, and identify blind spots before they show up in the field. Source systems: Guidepoint, GLG.

09 Competitive Intelligence (PMM, Marketing)Ongoing tracking of competitor moves - pricing, packaging, messaging, product releases, win/loss patterns - to inform differentiation strategy and campaign positioning. Agentic CI reports are increasingly viable here. Source systems: Klue, Crayon, Gong, Zapier.

Inside-Out: What your own systems, reps, and past campaigns already know

These five sources are frequently utilized. How systematically it’s being mined, synthesized, or routed back into campaign design is the question. Either way, Inside-Out inputs answer a different question than Outside-In: not what the market is saying, but what's actually happening inside your accounts and your own motions.

10 Usage & Engagement Telemetry (Product, Revenue/RevOps)First-party product data tracking feature adoption, session frequency, depth of use, and behavioral patterns. Engagement data for non-software businesses - website/portal engagement, frontline engagement, etc. Used to identify expansion readiness, churn risk, and product-led signals - which behaviors correlate with growth, and which with attrition. This data is some of the most predictive / impactful when it comes to NRR. Source systems: Mixpanel, Amplitude, Pendo.

11 Support Ticket Trending (Support, CS)Patterns in support volume, issue type, and resolution time that surface product gaps, onboarding failures, and accounts experiencing friction before it shows up in CSAT. One of the most actionable early-warning inputs available. Source systems: Zendesk, Intercom, Salesforce.

12 Rep Feedback (Sales, CS)Qualitative input from AEs, AMs, and CSMs on what's resonating in conversations, what objections are recurring, which offers land, and what customers are actually asking for. Regular roundtables with customer-facing personnel across the lifecycle are the mechanism - not ad hoc Slack messages. Source systems: Slack, Gong, roundtable notes.

13 Past Campaign Performance & Learnings (Marketing, Revenue/RevOps)Post-mortem analysis of previous campaigns: response rates, pipeline influenced, message performance, channel mix. For whatever reason, this is commonly ignored with surprising frequency. Predicates the need for a continuous listening/monitoring/improvement capability, which is immature in many organizations. Source systems: HubSpot, Marketo, Salesforce, BI/reporting.

14 Account Performance vs. x-ographics (Revenue/RevOps, Sales)Comparison of revenue performance, win rate, cycle time, growth rate, and churn against firmographic and technographic account attributes - to identify which account profiles over- or under-index, and sharpen ICP definition over time. Source systems: Salesforce, Looker, Tableau, data warehouse.

Why this matters across every motion

The instinct is to treat campaign inputs as a pre-launch checklist - do some research, build the campaign, run it. The intelligence layer isn't a checklist. It's an operating system.

New logo campaigns need Outside-In signal to know how the market perceives you versus alternatives, and Inside-Out data to know which account profiles convert. Expansion campaigns need telemetry and rep feedback to know which accounts are ready and what offer will land. Retention campaigns need CSAT, support trending, and usage data to catch risk before it's too late.

The fourteen sources above don't belong to one motion. They feed all of them - and the teams that systematize the loop between field input and campaign design are the ones who enjoy a superior return on their account-based investment dollars.