It’s been a minute since I’ve written but after being interviewed on Jane Serra’s Women in B2B Marketing podcast a couple months back, I’ve had some awesome convos continue in my LinkedIn DMs from marketing and sales leaders who listened to the episode. 7/8 of the conversations centered on how poor data was crippling GTM and AI enablement so I decided to lean into that predicament in the following perspective.
It starts with a concept I call the GTM floor.
The GTM Floor is your org’s ability to consistently collect, document, and operationalize buyer insight. It spans data hygiene, team alignment, CRM trust, and post-sale feedback.
With it, you scale what works. Without it, every campaign is a guess.
Symptoms of a cracked GTM floor and the systems that caused them include:
🛑 Salesforce or HubSpot opps have no pre-sale data: no notes, no champion, no trigger event
🛑 You can’t answer “Why did our last 5 deals close?” without asking 3+ people
🛑 Messaging is based on static positioning docs, not fresh patterns from sales calls or interviews
🛑 You use intent data, but it rarely translates into conversion or more engagement signaling in your target account list
🛑 You’re running campaigns, but can’t tell which channels or moments actually influenced pipeline
🛑 You run A/B tests, but it never feels like it’s leveling up beyond traffic metrics
Over the past decade, the potential to meaningfully distill customer and buyer data signals into meaningful marketing has exploded.
But our ability to wrangle, interpret, and operationalize that data hasn’t always kept up.
So instead of building systems that support clarity, many teams default to something that feels easier: ambiguity.
⚠️ We’d rather spin up an abstract plan and brief in Notion than confront the gaps in Salesforce.
⚠️ We brainstorm why a campaign didn’t work instead of digging into the segments that did convert (or create new ones to augment conversion)
⚠️ We say we want “data-driven” strategies, but we still plan from instinct.
Why does this happen?
Because real insight takes data mining and a commitment to build a reliable infrastructure. And it’s pretty time-intensive, if you’ve got a good amount of data and a foundation hasn’t been laid.
Unstructured data has to be distilled, pipelined, interpreted, and activated.
That takes systems. And systems require confronting where the chaos lives. Now, AI is becoming more helpful at the enterprise level with unstructured data solutions like Snowflake Intelligence but if you’re not currently seeking to onboard and train on these assistants, and have a mandate to clean up your legacy CRM systems, the lift will feel overwhelming.
This is where most teams stall out.
Even when the data exists, it often isn’t trusted, or worse, is actively denied.
I call this data deniability.
Having the data that shows your current GTM motion isn’t working but choosing to ignore it because it’s more comfortable to operate inside the box you’ve already built.
Founders do it.
Marketing leaders do it.
We all do it at some point.
Because if we acknowledge what the data is really telling us, it often means we have to rethink how we’re using the data we have and currently planning our GTM motions.
So we stay in the safe zone, trying to avoid the truth knowing that zone doesn’t scale.
GTM floor diagnostic test
To give this meaning in the context of how your team operates, ask yourself/team:
👀 Can we consistently identify the champion and trigger event in our closed/won deals?
👀 Do we have a repeatable workflow for post-sale interviews or feedback?
👀 Are our most important CRM fields (lead source, win reason, persona) consistently and accurately populated?
👀 Is marketing strategy built from first-party insights or just instinct/competitor motions etc?
👀 Could we explain what caused our best deal this quarter to close?
If you answered “no” to 3 or more, your GTM Floor has some visible cracking underway but not to worry. Here’s how to start (re)building a formidable GTM foundation.
Why most fixes don’t stick
Before we talk about what to do, we have to talk about how your team thinks.
Because the truth is, this isn’t just a RevOps or tooling issue as it’s often described. It’s primarily a byproduct of leadership and cultural communication styles and team-working.
Fixing your GTM floor requires a very specific mindset across both leadership and middle management. Without it, even the best-designed process or dashboard is just shelf-ware.
Here’s what that mindset looks like:
💥 Executive leadership understands the difference between “personal touch” and manual bottlenecks.
💥 Marketing operators knows that personalization at scale requires systems, not just good intentions.
💥 Tactical managers recognize where humans have superpowers (relationship-building, strategy, judgment) and where systems have theirs (consistency, memory, segmentation).
And most importantly, it sees your CRM/data intelligence hub not as an admin chore, but as a heartbeat function of your company’s present and future growth.
If your data systems don’t reflect reality, then no one—founder, SDR, PMM—is operating from truth. And when that happens, your GTM strategy stops being a strategy. It becomes improv theatre, with everyone reacting, guessing, and misfiring in real time.
Here’s how the floor cracks before the concrete even sets:
Let’s say your CS lead prefers keeping renewal notes in a private doc. "I move faster this way," they say. And maybe they do. But when Sales tries to upsell, they’re flying blind. When Marketing asks what’s resonating, there’s no data. When the CEO wants to know what drove retention, they get anecdotes, not insight.
And leadership (trying to protect morale because that CS lead is a star retainer) says: “Okay, we’ll leave it as is.”
That one decision, to allow a workflow outside the system, kills scalability.
Because no matter how strong your campaigns, dashboards, or messaging are, if key players aren’t contributing to the shared truth, nothing compounds.
You end up with bespoke effort pretending to be a strategy.
Everyone is working hard.
But they’re not working together.
Scalability isn’t about everyone doing the same thing.
It’s about everyone building on the same foundation.
This work requires conviction.
Not just to run the checklist, but to believe that data hygiene and buyer insight are not operational details.
They’re cultural assets and strategic leverage. And the only way you build a go-to-market motion that compounds.
If you’re not ready to align around that mindset, the next section won’t fix anything. But if you are, here’s where to start.
Trust the floor that unites the team
If you haven’t read The Five Dysfunctions of a Team by Patrick Lencioni, I highly recommend it. Definitely one of my favorites.
It was first published in April 2002 and twenty-plus years later, it remains one of the most essential books on how dysfunctional teams thwart growth. It’s set in a fictional Silicon Valley company, Decision Tech, where a new CEO takes her leadership team on an offsite to confront five core dysfunctions: from absent trust to avoidance of accountability, each of which cripples strategic clarity and execution.
The writing is easy to follow, but what sticks with many of us is how recognizable the characters feel. Protecting silos. Avoiding discomfort. Mistaking activity for alignment. The cracks Lencioni mapped twenty years ago are the same ones data now makes visible—if we’re willing to look.
Most of the dysfunctions Lencioni highlights: lack of vulnerability, artificial harmony, ambiguous commitment, are precisely what prevents a healthy GTM floor from taking root.
This is why:
If teams don’t willingly share information, your CRM or respective customer data repo remains sparse and invalid.
If they avoid discomfort in the moment, gating insight in “how we used to do it” becomes a default.
If leadership won’t hold people accountable to hygiene and alignment, every metric and dashboard becomes a guess.
If team members protect own workflows over collective clarity, you end up with manual effort masquerading as personalization or bespoke strategy.
These dysfunctions look obvious because they are foundational patterns in typical human behavior. But they directly break the infrastructure you're trying to build in an organization if it proliferates at each level of decision-making. They cripple the achievement of clean data, scalable insight, shared accountability over execution and audience understanding.
You don’t have a go-to-market engine but rather a fragmented effort masquerading as sales, marketing and product (data) alignment.
These then go beyond interpersonal problems to create infrastructure problems, because they prevent the kind of trust, transparency, and system-wide accountability required to build anything durable.
And now to your GTM foundation
A good example of what a strong foundational motion looks like is how Cursor uses Momentum. Cursor bakes real-time GTM signals into their ops stack. They use tools like Momentum to push structured deal data into Slack and Salesforce—not just to automate, but to standardize visibility across functions. That means fewer backchannel updates, cleaner attribution, and better clarity on why things are moving.
Teams like Cursor aren’t succeeding because they picked the right attribution model or growth channel. They’re succeeding because they’ve built and defended a strong GTM floor.
They understand:
Systems should enhance human output, not bottleneck it
Data is only powerful when it’s trusted across the org
Personalization at scale is an operations problem before it’s a creative one
Growth becomes compounding only when insights are shared, structured, and acted on
Because scale doesn’t necessarily come from adding more motion. It comes from aligning around shared, validated truths, and building phenomenally smarter and leaner, from it.
The fix: what to do before Q3 ends
For Product Marketing / Founding Marketers
✅ Set up a standardized deal debrief form in Notion or Airtable with 5 fields: Champion, Trigger, Pain Points, Objections, Win Reason
✅ Audit 10 won opps. Can you identify what actually converted them?
For Sales:
✅ Make champion, trigger event, and competitor fields required on opportunities
✅ Use the call intelligence tools you may already have access to to tag common objections and triggers.
✅ Share top reasons for lost deals in a weekly Slack, don’t wait for quarterly retros
For Founders:
✅ Stop asking for “more pipeline” without understanding what it actually takes to close. Because without insight into why your last deals converted, what triggered them, who championed them and what objections mattered. You’re not scaling a strategy, you’re just filling a leaky funnel. More leads won’t save you if your team doesn’t know how to move them through to revenue.
For RevOps (even if it’s you):
✅ Prioritize hygiene in the fields that matter: Lead source, Persona, Trigger, Win reason
✅ Simplify what’s required, but enforce consistency
✅ Create one dashboard that tells you what’s converting and why
If you need to understand the health of your current CRM data, feel free to create the following GPT. Within its configuration it has the ability to carry out a win-loss analysis but if your data is lacking critical fields for this analysis, it’ll flag the missing pieces via the final score.
You are Deal Data Health Check — a diagnostic tool for CRM deal/opportunity data.
Your primary job is to assess the **health and trustworthiness** of the dataset, not to do deep win–loss. Win–loss insights are optional and secondary, included only if coverage and sample size are strong.
========================
DATA READINESS & HEALTH
========================
1) FIELD MAPPING
- Map provided columns to canonical fields: Name, Amount, Stage, IsWon, IsClosed, Close Date, Created Date, Lead Source, Persona, Trigger, Champion, Competitor, Win Reason, Loss Reason.
- Accept common aliases (Deal Name, StageName, Original Source, Buyer Role, etc.).
- If uncertain, show a Mapping Table and proceed.
2) CRITICAL FIELDS (Tier 1 — required for strong health)
Lead Source, Persona, Trigger, Champion, Competitor, Win/Loss Reason, Close Date, Stage/IsWon.
Tier 2 (nice-to-have): Created Date, Amount, Segment, Owner, Notes.
3) COVERAGE & SAMPLE
- Closed sample size = Closed Won + Closed Lost.
- Compute % non-blank for each Tier-1 field (on CLOSED records only).
- Compute average Tier-1 coverage (Coverage Index).
4) CONFIDENCE LEVEL
- HIGH: Closed ≥ 100 AND Coverage Index ≥ 80%
- MED: 25–99 closed OR 60–79% coverage
- LOW: <25 closed OR <60% coverage
Include one-line reason in banner.
5) HEALTH SCORE (0–100)
Score = 0.4*Completeness + 0.2*DateSanity + 0.2*(1–DuplicateRate) + 0.2*(1–MessyCategoryRate).
- Completeness = Coverage Index.
- DateSanity = % valid close dates on closed records (and positive cycle time if Created Date exists).
- DuplicateRate = suspected duplicates (same Name ± Amount within 7 days).
- MessyCategoryRate = share of “other/misc/-” or singleton typos in Tier-1 categorical fields.
6) "WHAT YOU’LL MISS WITHOUT" TABLE
For each Tier-1 field:
- Missing % (closed), Impacted Metrics, Example Question You Can’t Answer, Suggested Fix.
========================
OPTIONAL WIN–LOSS (SECONDARY)
========================
Only if Confidence = HIGH or MED:
- Overall win rate.
- Win rate by Persona, Trigger (if ≥ 30 closed records and ≥ 60% coverage for that field).
- Champion effect (win rate with vs without).
- Top Win/Loss reasons (deduped).
Label these **Directional** if MED confidence.
========================
OUTPUT FORMAT
========================
0) Sufficiency Banner (Confidence + reason)
1) Executive Summary (≤120 words) — Health Score + top 2 gaps + immediate fix.
2) Data-Health Scorecard (Coverage per Tier-1 field, Coverage Index, component scores).
3) “What You’ll Miss Without” table.
4) Optional Win–Loss snapshot (if data supports it).
5) 14-Day Hygiene Plan — prioritized fixes for missing/weak fields, with suggested owners/actions.
6) Appendices (optional, on request): Suspected Duplicates, Missing Golden Fields CSV.
========================
TONE
========================
- Crisp, operator, no fluff.
- Flag missing or weak fields early.
- Always return a result, even with LOW confidence.
- If LOW: emphasize hygiene plan, keep win–loss minimal.
Conversation Starters
• Here’s my deals CSV — give me the data health check.
• Show me coverage per critical field and the “What You’ll Miss Without” table.
• Give me health score + 14-day hygiene plan (skip win–loss).
• Run full check and include optional win–loss if possible.The Future of CRM Data in the Age of AI
We’ve been talking about HubSpot and Salesforce here because they’re the default for most GTM teams today. But they’re also legacy CRMs, built for a world where data lived in rigid, form-based fields and most of the insight had to be entered manually. AI enrichment is being added with HubSpot getting the GPT connector but it’s still layered on top of old assumptions: that humans will structure the data, that systems won't talk to each other, and that insight comes after the fact.
That model has a cost: data debt. Over years of inconsistent entry, missing fields, and siloed notes, these systems accumulate blind spots that are hard to clear without a heavy, manual lift.
In 2025, that’s changing.
With AI intelligence layers, such as Snowflake Intelligence, or emerging platforms like Pylon, Attio, Attention, and Clay, we can begin to dissolve much of that debt.
Clay in particular blurs the line between data enrichment and dynamic insight capture. It can ingest signals from dozens of sources: LinkedIn activity, hiring signals, funding rounds, intent data, and then push those enriched profiles back into your CRM or GTM systems. When combined with tools like Pylon or Attio that parse unstructured conversation data and Attention that pulls signal from call notes and meetings, you get an intelligence stack that can:
Detect and tag Trigger Events (e.g., leadership changes, compliance deadlines) without manual input
Identify or confirm Champions based on call transcripts, email mentions, or LinkedIn engagement
Enrich missing Persona and Segment fields with verified data
Align Win/Loss Reasons with actual customer language from calls or support tickets
Instead of retroactively filling in “Trigger Event” or “Champion” by memory, an intelligence layer can detect those signals automatically—sometimes across entirely different systems—and write them back in a structured, queryable form. That means:
Fewer manual data entry battles for your team
Cleaner, more complete CRM fields without the retroactive chase
Cross-system insight you can actually use in pipeline reviews, campaign targeting, and forecasting
The net effect: your CRM becomes more truthful, faster, and your “GTM floor” strengthens without the constant grind of admin work.
We’re entering a new era where your GTM floor doesn’t just store data but rather creates the context for everything built above it. It’s the adaptive layer that allows every other part of your stack (campaigns, personalization, AI, sales plays etc) to improve over time. But that only happens if the floor itself has structural integrity. And that integrity doesn’t come from tooling alone. It comes from leadership that actively cultivates a culture where data health, system thinking, and shared truth are non-negotiable.
Yes, you can grow without it. Many companies do. But that growth often comes at an opportunity cost: it's harder to scale, easier to stall, and disproportionately reliant on a few high-performing individuals or trusted relationships. Without a healthy GTM floor, every new layer adds weight, but not necessarily strength.
Till next time,
Chae ✌🏼
Say hi on LinkedIn
