
Most go-to-market teams are drowning in signals they can't act on. Every tool in the stack — the CRM, the marketing automation platform, the intent provider, the engagement tool — is trying to be the authoritative source on who to call and when. The result is not clarity. It's noise, fragmentation, and sales reps spending the majority of their day doing research instead of selling.
In a recent RevOps Co-op webinar, Matthew Volm sat down with Nico Druelle, Founder at The Revenue Architects, and Simon Travers, Head of RevOps at FullEnrich, to walk through how modern revenue teams can break out of that cycle — and what it looks like to build a centralized signal store that actually changes rep behavior.
The session combined Nico's framework for signal architecture with Simon's first-hand account of deploying it inside FullEnrich's own go-to-market motion — including the false starts, lessons learned, and the numbers that made it worth it.
Nico opened with an observation drawn from working with more than 20 sales and marketing teams: the fragmented tool stack isn't just inefficient — it's actively undermining pipeline capacity. Each tool in the stack tries to solve the signal problem in isolation, feeding different scoring logic and different recommendations to the same CRM or Slack channel, with no feedback loop to validate whether the recommendations were any good.
The consequence is measurable. As Nico noted, research suggests that somewhere between 60 and 80 percent of a seller's time is not spent selling or generating pipeline. For RevOps practitioners, that number should function as a north star metric — something to actively track and drive down through operational design rather than accept as a fixed cost of doing business.
The practical counterpoint to that dysfunction isn't a better tool. It's a different architecture. This connects directly to what the RevOps community has been grappling with in the broader conversation about AI readiness and data infrastructure — the insight work that has to happen before any automation layer can be trusted.
The concept Nico introduced is a signal store: a centralized data layer — typically a warehouse like BigQuery or Snowflake — where all signal data, first-party and third-party alike, is ingested, resolved at the account and contact level, and scored with a consistent methodology before being pushed to the teams who need to act on it.
This is a deliberate departure from the current architecture most teams run. Right now, intent platforms, CRM enrichment tools, and engagement platforms all pipe their own proprietary scores into the system of record. There's no shared entity resolution, no unified definition of what "hot" means, and critically, no loop back to those tools to inform whether the outcome they predicted was correct.
The signal store inverts that logic. Rather than asking each tool to be the decision engine, it treats those tools as data sources — and builds the decision logic internally, in a place (the warehouse) where it can be iterated on with SQL and dbt instead of hidden inside vendor black boxes. As Nico put it:
"If you do it downstream in your sales stack, you're gonna create a lot of technical debt, and you're gonna most likely not build something sophisticated enough to be statistically relevant." — Nico Druelle
The payoff for getting this right is illustrated by one of the session's most concrete examples: an AI-generated email built on top of a signal store. The email identified how many engineers at a prospect company were using a product that had hit a usage limit, flagged that another team at the same company was running a competing tool creating internal inconsistency, and offered a time-limited trial — all without a rep spending time on research. Reply rates on that kind of outreach, Nico noted, run roughly six times higher than generic cold email.
Simon built on Nico's architecture with a concrete implementation — the product-qualified account (PQA) framework FullEnrich uses to classify and prioritize signals across its entire revenue funnel. With roughly 10,000 inbound leads per month flowing into the business, the challenge is not finding demand. It's routing the right demand to the right motion at the right time.
FullEnrich organizes its signals into four categories, each representing a distinct phase of the revenue funnel:
PQA Convert captures sales-qualified accounts that have crossed specific usage thresholds — signals that meaningfully increase the probability of conversion over accounts with similar firmographic profiles but no product engagement. According to the data Simon shared, accounts that meet the usage threshold convert at a 60 percent higher rate.
PQA Expand fires when an existing paying customer crosses a usage trigger that signals readiness for a larger commitment. The example Simon used: a customer purchasing three top-ups within 30 days. That behavior pattern creates an opening for a conversation about resizing the plan — and 30 percent of FullEnrich's expansion revenue is now directly linked to expansion signals.
PQA Migrate identifies self-serve customers who have grown organically to a point where introducing a sales rep creates incremental value. Simon noted that when accounts hit the PQA migrate threshold, deal generation doubles compared to self-serve accounts without that trigger.
PQA Recover is distinct from the other three in that it flows to the customer success team rather than sales. It's a churn prevention signal, fired when a customer's health score drops below a defined threshold, with a clear SLA for the CSM to intervene.
"As you can see here it's really important because a big part of our acquisition and conversion is linked to inbound. So really important for us to have a well-defined PQA." — Simon Travers
The PQA framework is worth sitting with because it does something most scoring models don't: it ties each signal category to a specific GTM motion, a specific team, and a specific action — not just a vague heat score. This kind of operational specificity is what separates a signal store from a glorified intent dashboard. It's also an illustration of the broader principle that lead and account qualification only works when definitions are precise and consistently enforced.
Not every company comes with FullEnrich's natural advantage — a B2B data business sitting on top of its own product usage data. For teams without that first-party foundation, or teams still building the data pipelines that will eventually supply it, third-party signals remain the accessible starting point.
Nico was direct about the state of that market: third-party signals have become a commodity. New hire alerts, job postings, funding rounds, champion job changes, product launches — most tools in the market now surface these in some form. The differentiation has shifted entirely to the interpretation layer.
"The only alpha you have left is your ability to parse out intelligence or context from — I'll call them signals, but they're sort of like little footprints that your buyers are leaving online." — Nico Druelle
The framework Nico recommends for getting that interpretation right starts with the sellers themselves. What do your top reps systematically look at when they research an account? Those patterns are the input — what has empirically worked becomes the starting template for signal selection. The goal is to operationalize instinct: if every strong rep checks LinkedIn for recent hires before reaching out, that behavior should be automated into the signal pipeline.
From there, the work is in data labeling — taking a raw payload like a job description and identifying the specific snippets of text that carry meaningful context about the company's current state, then converting those snippets into structured, queryable data points inside the warehouse. Nico was explicit that this is where most of the complexity lives: roughly 80 percent of the work in building a functioning signal store is in this labeling and enrichment step, not in the ingestion or scoring that follows.
FullEnrich's own stack for external signals, as Simon described it, routes through two primary providers — one for a broad bank of configurable signals (LinkedIn engagement, champion job changes, competitor engagement), and a second for custom signals built in Clay and pushed into BigQuery via Workflow.io. Those external signals are joined with internal signals from Stripe payment data and backend usage data in a unified fact signals table, which is then matched against a total addressable market (TAM) table containing all ICP companies and target personas.
The signal delivery layer — HubSpot tasks, Slack notifications with direct links to contact and company records — is designed to match the existing inbound workflow the sales team already operates in. That design decision turns out to be more important than it appears. Teams that try to introduce signal-driven outreach as an entirely separate motion face an adoption hill. Teams that thread it into existing processes face a much smaller one. This maps closely to the change management challenges that consistently undermine RevOps rollouts of otherwise well-designed systems.
Scoring and routing the right accounts requires a consistent framework for prioritization — and Nico introduced a simple but effective two-by-two for that purpose.
The framework plots accounts on two axes: fit (how closely does this account match the ideal customer profile, and what is its LTV potential?) and intent (how timely is it to reach out right now, based on observed signals?). The resulting quadrants give sales and marketing leadership a clear action model.
High fit, low intent accounts are the classic nurture play — companies that look exactly like ideal customers but haven't yet exhibited meaningful buying signals. These should be in ABM programs, marketing retargeting, and early-stage rep nurture, not on a cold call list. High fit, high intent is the priority bucket: the five to ten percent of the pipeline that represents genuine immediate opportunity, where delay is the primary risk. As Nico noted, if any of these accounts are not being actively sequenced, the revenue leak is significant.
The bottom right quadrant — lower fit, high intent — is where programmatic and agentic outreach earns its place. These accounts aren't the flagship logos, but they are real opportunities for teams with the operational capacity to pursue them at scale without diverting senior rep attention from the high-fit accounts that require more considered outreach.
This kind of tiered prioritization model is a natural extension of the account-based marketing frameworks many revenue teams already have in place — the signal store just gives those frameworks a data foundation with real teeth.
Simon was candid about the gap between how they expected the project to unfold and how it actually did. The original assumption was that building a signal store was a one-month project: set it up, push signals to the sales team, build some automation, done. That assumption underestimated almost everything that actually mattered.
The first surprise was adoption. The benefits of a signal-driven motion aren't self-evident to a sales team that has never experienced it. Early on, the signals landed in Slack and in HubSpot but weren't being used — not because the reps didn't want better information, but because they didn't yet trust it and didn't have the context to know what to do with it.
The second problem was that the initial rollout of PQA migrate signals gave reps only one lever to pull: discount. That created a dynamic where the signal was sound but the prescribed action undermined the business. The fix was to build out a fuller range of responses — feature gating, usage education, commitment-based pricing conversations — so that reaching out on a migrate signal felt like genuine consultative selling rather than a defensive price negotiation.
The lessons Simon distilled from that year of iteration map to three non-negotiables: clear ownership and territory routing for each signal (so there's never ambiguity about who should act), SLAs that are matched to the urgency of the signal type (competitor engagement signals decay in days; champion job changes can wait weeks), and tracking of signal usage and outcomes — not as a monitoring exercise but as the data input that makes signal improvement possible.
FullEnrich now runs a regular review cadence — roughly every two weeks when active iteration is happening — to assess which signals are being used, which are generating false positives, and which should be killed or scaled. The practical implication for RevOps teams is that signal governance is ongoing work, not a launch activity — and the teams that treat it as a one-off project will get one-off results.
"Not treat this as a one-off project, but as a continue work and part of the run." — Simon Travers
The logical endpoint of the signal store architecture is a consolidated operating environment for every GTM team — what Nico calls a Sales OS. The idea is to eventually eliminate the tab-switching, the manual research, and the competing inboxes of signals from disconnected tools, and replace them with a single interface that surfaces the right accounts, the right context, and the right recommended actions.
The important clarification Nico offered is about sequencing. The Sales OS is not where you start. Building a custom UI before you have adoption of the underlying signal infrastructure is exactly backwards — you get a beautiful empty vessel that no one uses because the data feeding it hasn't yet earned trust.
The right path is to prove out each individual workflow before consolidating them. Start with the BDR team, where the complexity of account context is lower and any incremental lift in research quality is immediately appreciated. Build signal-driven workflows into the channels the team already uses — Slack notifications, HubSpot tasks — matching the format of their existing inbound processes exactly. Demonstrate that this workflow is replacing 25 percent of the manual research time, then move to the next block of time.
"You're rolling out features for each of those chunks of work, and you are trying to make this frictionless in their workflow." — Nico Druelle
Once enough of those individual workflows are proven and adopted, the case for consolidation makes itself. The Sales OS becomes the natural container for capabilities the team is already using rather than a new thing they're being asked to adopt. And the FOMO effect is real: when SDRs are visibly benefiting, AEs and CS teams want in. This matches the pattern many teams have experienced when rolling out enablement changes — adoption doesn't follow mandate, it follows demonstrated value.
The aspiration Nico laid out is broader than a single team: a sales OS for the sales team, a growth OS for marketing, a CS OS for customer success — each one pulling from the same underlying signal infrastructure, each one purpose-built for the specific actions those teams need to take.
The through-line across this session is straightforward but easy to miss in the rush to build: the signal store is not a technology project, it's a trust project. The warehouse architecture, the data labeling, the scoring models — all of it is in service of a single goal: getting sales reps to the point where they open their computer on Monday morning, look at the accounts the system surfaces, and act on them without question. That trust is built in increments, signal by signal, workflow by workflow, until the system has replaced the manual research behavior it was designed to eliminate.
Learn more about how FullEnrich helps sales teams and revenue operators access best-in-class B2B contact data across 20-plus providers — built for the kind of signal-driven prospecting and account enrichment this session covered.
Check out our blog, join our community and subscribe to our YouTube Channel for more insights.