Real stories and things nobody says out loud about data. Written by Emily Goh, Founder of Gohvio.
Metric definition governance means every key metric has a written definition, a named owner and a change record. Set it up before AI picks a meaning for you.
· 5 min readRevenue OperationsNew to a RevOps role? Learn the CRM later. First ask Sales, Finance and the CRM itself what Closed Won means, write every answer down and compare them.
· 5 min readSales OperationsA 30-minute pipeline review meeting agenda in four parts, plus the prep the day before that keeps the room focused on decisions, not blame.
· 5 min readAI & DataAI generated reports are only as accurate as the data and definitions behind them. Why every AI summary needs an owner, and how to check one this week.
· 5 min readSales OperationsWhen a deal slips, check the record before you blame the rep. A six-point pipeline hygiene checklist, and why it should run before the pipeline meeting.
· 5 min readRevenue OperationsSome backlog tickets are really arguments between teams. How to spot the two kinds, turn each into a decision with a written rule and an owner, then build.
· 6 min readFounder / LeadershipAccurate board reporting is not about better slides. It is about proving every number. What a number's receipt looks like and how to build one for your next board deck.
· 5 min readRevenue OperationsThe CRM can be learned in weeks. The judgment underneath it takes years. How to screen RevOps candidates for skills that travel, with interview questions and a short exercise.
· 6 min readRevenue OperationsMarketing reports $6.0M of sourced pipeline for $500K of spend. Finance sees $400K of new bookings. Same quarter, same company. Here is how to rebuild the number so you can answer one question: are marketing dollars working?
· 6 min readRevenue OperationsA use case, one question at a time: how to prepare QBR data two weeks out so the room spends its hour on decisions instead of arguing about numbers.
· 6 min readFounder / LeadershipCompanies rarely lack reports. They lack agreed definitions, data checked against them, and evidence they can trace. Here are five decisions that go wrong without it.
· 6 min readRevenue OperationsYour records move in a CRM migration. Their meaning doesn't. Write down each key definition, where it's enforced and who owns it, then test the new CRM against it.
· 6 min readSales OperationsMost forecast calls test the rep, not the evidence. Check the records first, write down what Commit means, and run a 30-minute call that ends with owners.
· 5 min readRevenue OperationsKey takeaways and a Q&A from the first Fireside with Leore Spira: fix the process before the tool, start with five metrics, and agree definitions and owners first.
· 4 min readRevenue OperationsA simple RevOps prioritization framework: two questions sort every backlog item into rebuild, work around or leave it. Run it in 10 minutes this week.
· 5 min readAI & DataAI on dirty CRM data turns stale records into confident answers. The fix is teaching it your guardrails. What building that really costs, and what to ask before you buy.
· 5 min readRevenue OperationsWhen the quarter misses, stop asking who and start asking what changed. Your CRM's deal history and a written pipeline definition already hold most of the answer.
· 5 min readRevenue OperationsNew to running RevOps? Fix the definitions and owners behind the numbers leadership already uses, then the fields that feed them. Build the dashboard last.
· 5 min readData QualityEvery number in your board deck has two trails behind it: who did what to the records, and which definition was used to count them. Lose either one and the number becomes an opinion.
· 6 min readRevenue OperationsEveryone assumes closed won is settled. Then Sales and Finance put the same deal in different quarters. Five questions your definition has to answer, and a check to run this week.
· 5 min readRevenue OperationsUrgency in a backlog is set by whoever filed the ticket. Rank by what it breaks instead: three questions for every item, and a six-step sort for this week.
· 5 min readData QualityEvery definition your numbers rely on has a person who approved it, and a record of who changed it and when. Different teams may mean different things. Nobody knowing is the problem.
· 6 min readRevenue OperationsForty minutes on whose number is right. Ten minutes on what to do. If that's your pipeline meeting, the problem isn't your people. It's that nobody walked in with evidence.
· 5 min readData QualityA 0 in Amount, a "0" in First Name, Closed Won with a close date next year. Every one passes a type check. Every one can wreck a forecast. Data quality has to move from type to meaning.
· 6 min readData QualityMarketing and Sales can each report a different number for the same quarter, both computed correctly. A guardrail is the written definition plus the flag when a record does not meet it.
· 6 min readRevenue OperationsRevOps unified Marketing, Sales, and CS. It worked. Now the same shift is happening across Finance, Product, and Customer Data. Here's what unified ops actually looks like -- and why it's the next step.
· 7 min readData QualityAuto-cleaning data sounds magical. It's actually dangerous. Here's why the right pattern is surface-and-recommend, not silently fix -- and why that's a design principle, not a limitation.
· 6 min readData QualityMost teams clean data at the dashboard. By then it's too late. The right place to filter is upstream -- before the data ever reaches a report. Here's what that looks like.
· 6 min readData QualityCleaning dirty data is janitorial work. Preventing dirty data is architectural work. Here's the shift from one to the other -- and why it changes what your ops team actually does.
· 6 min readFounder / LeadershipAutomating bad processes faster doesn't help -- it just produces the wrong outcome at scale. Here's why founders should pay for standardization first, automation second, and what it costs to skip the step.
· 7 min readRevenue OperationsPick any number on any dashboard. Can the person who owns it explain where it came from in under five seconds? If not, nobody owns it. And unowned data always breaks.
· 6 min readData QualityMarketing says $50k. Finance says $30k. You can buy the most expensive BI tool on earth and still get two different numbers. Here's why, and what actually solves it.
· 7 min readData QualityEveryone blames culture when teams don't talk. But the reason Marketing and Finance disagree isn't a relationship problem. It's a definitions problem. Here's the difference.
· 7 min readMarketing OperationsRevenue numbers that don't match usually come from two places: garbage data in the CRM and teams using different definitions of the same word. Fix one without the other and the numbers still don't agree. Here's the three-step fix.
· 6 min readMarketing OperationsYour data isn't broken. Your definitions are. 'MQL' means four things at your company, depending on who you ask. Until you fix that, every cleanup project is a band-aid.
· 7 min readRevenue OperationsEvery quarter, your definitions drift a little. By year three, the data tells a story nobody recognizes. Here's what the drift actually costs -- and why it gets bigger every quarter you ignore it.
· 6 min readRevenue OperationsMarketing has a dashboard. Sales has a dashboard. Finance has a dashboard. They all disagree -- and the era of accepting that as normal is ending. Here's what's coming next.
· 6 min readRevenue Operations'Data contracts' sound like engineering jargon. They're not. A data contract is one paragraph anyone can write, and it's the single most useful artifact ops teams can produce.
· 7 min readAI & DataMost AI tools race to give you an answer. The valuable ones slow down and ask you the right question first. Here's why -- and what to look for in any AI you let near your data.
· 6 min readAI & DataEvery tool says 'human in the loop.' Most still try to automate the human out. Here's what AI-enabled, human-first actually looks like -- and why it's the only sustainable pattern for data work.
· 6 min readFounder / LeadershipMost QBR decks show activity. The board wants to see decision-quality data. Here's the difference -- and why the gap usually comes down to the data underneath the slides, not the slides themselves.
· 7 min readFounder / LeadershipMarketing has their dashboard. Sales has theirs. Finance has 3. None of them show the same company. The org-wide view -- the one that connects pipeline to revenue to burn to EBITDA -- almost never exists. Here's what it looks like and why it's so hard to build.
· 7 min readFinance & AccountingMost finance dashboards are built for reporting. The one your CFO actually needs is built for early warning. There's a difference. Here are the 5 views and the data problems underneath them that nobody talks about.
· 6 min readSales OperationsSDRs open 6 tabs every morning. Their CRM. The sequencing tool. The LinkedIn tab. The shared spreadsheet. The Slack channel with lead updates. They trust one of them. Here's what a real SDR dashboard shows -- and why the others are noise.
· 6 min readMarketing OperationsMarketing teams average 7-12 dashboards. Most of them measure activity. The board needs results. Here are the 4 views that actually matter -- and why the other 9 exist only because the data in the first 4 isn't trusted.
· 7 min readRevenue OperationsEvery team built their own dashboard because they stopped trusting everyone else's. Now you have 14 versions of the truth. Here's why dashboard proliferation is a data problem, not a tool problem -- and what to cut.
· 6 min readMarketing OperationsTargets set on gut. Attribution that changed three times. A campaign live by Friday when the brief arrived Tuesday. Then everyone wonders why the leads didn't convert. We need to talk about what we're actually asking marketers to do.
· 9 min readRevenue OperationsThe campaign hit its goal, but you can't tell which channel earned the lift. The renewal slipped, but the discrepancy was in a row no one read. The pipeline looked healthy at QBR, and then half of it never closed. You are not failing because the strategy is wrong. You are failing because the data underneath it is.
· 7 min readMarketing OperationsMQLs are up 43%. The budget meeting is Friday. Everyone's celebrating. Nobody validated where the lift actually came from. Here's what happens at the next QBR when pipeline doesn't convert -- and how to catch it before you report it.
· 7 min readRevenue OperationsLead routing feels like a process problem. It's a data problem. When the account record is wrong, the territory match is wrong, the rep is wrong, and the lead dies quietly. Here's the full breakdown.
· 7 min readSales OperationsDuplicate accounts don't just break your pipeline rollup. They break your headcount model, your territory design, and your quota targets. Here's the full cascade -- and how to stop it before Q3 planning.
· 7 min readCustomer SuccessYour customer data lives in 5 systems. By the time churn shows up in the board deck, the customer left 60 days ago. Here's how to see it coming.
· 8 min readOperationsLegal data isn't a board metric. But it feeds every board metric. Contract cycles, redlines, renewals -- they all matter. Here's why.
· 8 min readFinance & AccountingYour board deck looks polished. The numbers underneath? Stitched together from 4 systems. Let's fix that.
· 7 min readMarketing OperationsYour marketing board deck has 47 KPIs. The board needs 5. Here's which ones actually matter and why your current deck is missing the point.
· 7 min readMarketing OperationsMulti-touch attribution is the holy grail of marketing. Except your data is wrong, so your attribution is fiction. Here's why and what to do instead.
· 8 min readMarketing OperationsEveryone says MQLs are a vanity metric. They're not wrong. But the real problem isn't the metric -- it's that your MQL timestamps are wrong.
· 8 min readData QualityYou're not pulling reports. You're investigating them. You're asking Slack for confirmation. You're rebuilding numbers because something feels off. That's reverse-engineering.
· 9 min readFinance & AccountingYour contract says $100k. Your invoice says $95k. Your bank shows $90k. Finance teams call this normal. It's not.
· 8 min readFounder / LeadershipYour exit multiple is built on one number: EBITDA. If EBITDA is built on bad data, your company is worth what you think it is, not what it actually is.
· 9 min readData QualityA VP pulled the pipeline number. It looked light. She hired two reps. The number was wrong -- and the company spent six figures solving a problem that didn't exist.
· 6 min readSales OperationsYour sales board deck looks great. It's also hiding the stuff that actually matters. Here's what your board really needs to see.
· 8 min readData QualityBad data doesn't cost money when you clean it up. It costs money when you make decisions based on it, and nobody's tracking that.
· 7 min readData QualityEveryone knows the phrase. But when you ask ops teams how they actually catch bad data, the answers are scary, and always come too late.
· 8 min readAI & DataEveryone's building AI on their CRM data. But nobody's cleaning the data first. The model is confident. The predictions are wrong.
· 7 min readRevenue OperationsIf you had to bet $50k on your pipeline forecast accuracy, would you? Most teams would lose that bet. Here's why.
· 7 min readFinance & AccountingFinance teams say the same thing every month-end: the close is a nightmare. Not because people are slow. Because the data doesn't agree.
· 7 min readData QualityEvery method ops teams use to find bad data is reactive. The damage is already done by the time they discover it.
· 8 min readFounder / LeadershipI'm a marketing leader. I live in pipeline numbers, MQLs, and justifying every dollar I spend. All of that depends on trusting your data. Mine kept letting me down.
· 8 min readOperationsEveryone blames Excel. But Excel is doing exactly what you told it to do. The problem is what you're asking it to do.
· 7 min readAI & DataEveryone says they're 'adopting AI.' Most teams are just running ChatGPT, Claude, or Copilot. Real AI adoption is something different entirely.
· 7 min readData QualityEvery company wants one. Almost none have one. The problem isn't picking the right tool. The problem is the data flowing into it.
· 7 min readThirty minutes. We connect one source and show you what is in it.