Português

A Day in the Life of a RevOps Manager: What the JD Won't Tell You

The job description for your RevOps Manager role probably opens with something like "own forecast accuracy, partner with sales leadership, drive cross-functional alignment." Real, technically true, useless. None of that prepares you for the Slack DM at 7:42am from your CRO that reads: "Hey, can you check why ACV dropped 18% in the dashboard? Asking before standup."

That DM is the actual job. Translating a panic into a number, a number into a root cause, a root cause into a one-sentence answer your CRO can repeat in standup without sounding like he's flailing, all in the seventeen minutes before the standup starts. The JD won't tell you about that DM. So here's a Tuesday I had recently, hour by hour, with the names changed and the data quality issues real.

If you're a candidate evaluating the role, this is what you're signing up for. If you're six months in and wondering if it's normal, it is. If you're a hiring manager scoping the job, read this and revise the JD.

7:42am: The Slack DM Before Coffee

Phone buzzes. CRO. ACV question. I open Looker on my phone, glance at the cohort, and immediately I can tell it's not a real ACV drop. It's a definition drop. Someone changed how Annual Contract Value rolls up in the dbt model two days ago, and now multi-year deals are getting averaged across years instead of taking the first-year contract value. The number didn't move; the calculation did.

I type back: "Not real. dbt model change Friday, multi-year deals are averaging. I'll have it patched by 9. Use last week's number in standup." Crisis defused in one message. Time to actually start the day.

8:00am: Pipeline Scan

Coffee, second monitor, two tabs: Salesforce and Clari. I'm not looking at the pipeline. I'm looking at what moved in the pipeline overnight.

This is the muscle no one tells you to build. Pipeline at rest is meaningless. Pipeline in motion (what stage-changed, what close-date-pushed, what amount-edited) is where the truth lives. I run a saved report: Opportunities modified in last 24 hours, group by Stage Change. Sixty-three rows.

Most are noise. Three aren't.

The first is a $180K deal that just pushed close date for the third time. Q3 to Q4, Q4 to Q1, Q1 to Q2. That's the third-time slip pattern, and in my experience it means the deal is dead and the AE doesn't know it yet, or knows it and doesn't want to take the hit. I flag it in my notebook to raise on the forecast call.

The second is a deal that quietly regressed from Negotiation to Proposal. Stage regressions are almost always either a real signal (champion change, budget refresh) or a CRM hygiene problem (an AE who hit the wrong dropdown). I check the activity log. Champion left the company on Monday. Real regression. AE hasn't updated the deal description, so this would have been invisible on the forecast call. I add it to my list.

The third is a $40K deal where the AE bumped the amount up 20% with no activity in the account in eleven days. Suspicious. Amount edits without activity are a sandbagging-or-stretching tell. I'll ask about it at 1:1 tomorrow, not on the forecast call. Calling out an AE in front of the CRO is how you lose AEs.

9:30am: Forecast Call Prep

The 10am forecast call is where my job is most visible and most fragile. I have thirty minutes to reconcile three numbers that should match but never do:

  1. What the CRO will say, based on his read of the AEs in his 1:1s yesterday.
  2. What Clari shows, based on rep-entered commit and best case, weighted by the AI model.
  3. What the AE-by-AE roll-up shows in Salesforce, based on what's actually in the CRM right now.

These three numbers are typically off by 8-15% from each other. The CRO's number is usually highest (recency bias from the 1:1s). Clari's is usually middle. Salesforce roll-up is usually lowest because half the AEs haven't updated their deals since last Thursday.

My job in those thirty minutes is to identify the three deals the CRO will get questioned on, pre-write the answers, and decide whether to flag the underlying data quality issue today or wait. Today the issue is that a Zapier flow somebody set up in 2024 is silently rewriting Stage values on inbound leads. There are forty-two opportunities currently in Stage = Discovery that should be in Stage = Qualification because the Zap is firing on the wrong trigger. I find this every couple of months. I log it, fix it after the call, and don't mention it on the call. The CRO doesn't need a Zapier diagnosis at 10am.

11:00am: The System Fix Request

Forecast call wrapped. Inbox open. Four Slacks. The one that matters is from an AE: "Outreach isn't logging tasks back to SF for the new sequence. Is this a bug?"

It's not a bug. It's never a bug. It's that Marketing renamed the campaign two weeks ago, the campaign-to-sequence mapping in Outreach broke silently, and now task logging is failing for any deal sourced from that campaign. Forty minutes to diagnose (read Outreach API logs, cross-reference SF campaign history, find the rename). Ten minutes to fix (update the mapping, push the change, validate three test tasks log correctly). Zero thanks, because to the AE it was just "Outreach being broken again."

This is the part of the job that compounds. Every system you run is one Marketing rename, one Salesloft sequence change, one Salesforce field deprecation away from breaking. The good RevOps Managers I know all keep a private wiki: which integrations break when which thing changes upstream. Mine has 31 entries. It's the most valuable doc I own.

1:00pm: The Async Fight About Data Quality

Lunch at my desk. Slack thread with three SDRs and two AEs about Lead Source going blank. This is the recurring fight, and I've learned the only way to win it is to make the cost their problem, not mine.

I post in the channel: "Pulled the data: last quarter, 31% of closed-won deals had no source attribution. That's why Marketing's getting credit cut on the comp plan. If you want comp to weight inbound correctly, we need source filled on every lead. I'm not asking you to like the field. I'm telling you it's why your spiff math is off."

Three replies in five minutes. Two AEs argue. One SDR says "fair, I'll fix it." That's the win. You don't need everyone to agree; you need the math to be visible enough that nobody can pretend it doesn't exist. Data quality fights are won with comp impact, not lectures.

3:00pm: Three Stakeholders, One Dataset

Mid-afternoon is when the requests stack up. Today:

  • CFO wants net revenue retention by segment for the board deck. Needs it Friday.
  • CMO wants attribution by channel for a campaign post-mortem. Wants it tomorrow.
  • CRO wants pipeline coverage by rep for a 1:1 with the COO. Wants it in two hours.

Same source data. Three different cuts. Three different deadlines. I can ship one today.

I ship the CRO's. Not because he's most important, but because his deadline is shortest and the cut is simplest. Pipeline coverage by rep is a 90-second Looker explore filter. I send the CFO a holding email: "Working on NRR by segment. I'll have a draft for you Thursday end of day so we have Friday for revisions." I Slack the CMO: "Channel attribution post-mortem coming Wednesday afternoon. Want to scope the channels with me first or should I default to your last brief?"

The trick to not getting buried is converting every request into either a delivered artifact or a scoped commitment with a date. Open requests rot. Scoped ones don't.

5:30pm: Commit vs. Best Case Reconciliation

Most of the team logs off. I open Salesforce and pull the day's pipeline movement into a one-tab Google Sheet. Three columns: AE, change in Commit, change in Best Case.

I'm looking for two patterns. AEs whose Commit barely moves but whose Best Case keeps swelling are over-committing: they're cushioning for the call but stuffing optimistic deals into Best Case to look good on coverage. AEs whose Commit moves down but Best Case stays flat are sandbagging, quietly lowering the floor while keeping the ceiling unchanged. Both patterns lie in different directions. Both matter for next month's quota planning.

Today there's one of each. I write a two-line summary into my notes: "Sara: over-committing on Best Case, three deals look stretched. Marcus: Commit dropped $90K with no activity to justify, sandbag candidate." That's it. Those two lines become the agenda for tomorrow's forecast call.

9:00pm: Dinner. Then the dbt Model.

Wife asks how my day was. I say "normal." She knows what that means. I make it through dinner.

9:00pm I'm back at my desk because the Looker dashboard the CRO opens at 8am tomorrow is pulling from a dbt model that I now know is broken. The morning ACV problem was the symptom; the cause is a column rename in the upstream opportunities_enriched model that my downstream revenue_metrics model wasn't updated for.

I open the dbt project. I find the offending column. Old name was acv_first_year, new name is acv_year_one. Whoever made the upstream change didn't propagate. I fix the downstream SQL across three references in two models. I run dbt test locally. Two assertions fail because of an unrelated null issue I don't have time to chase tonight, so I scope-fix the original problem, push the PR, ping our analytics engineer to review tomorrow morning, and validate the model rebuilt cleanly in production.

11:14pm. I open Looker, refresh the dashboard. ACV is back to the right number. I close my laptop. Forecast call is at 8am.

What the JD Doesn't Say

The JD describes the title. It says you'll own forecast accuracy and drive cross-functional alignment. It doesn't say what those phrases actually mean.

You're a translator. CRO speaks pipeline. CFO speaks margin. CMO speaks MQLs and CAC. AEs speak commission. Your job is to make all four numbers reconcile to the same underlying truth, every day, in time for the next meeting. Most days nobody notices. The day they do, it's because something broke and you're the only one who can see why.

You're also a debugger of systems nobody else thinks about. Salesforce field renames, Zapier flows, dbt column drift, Outreach campaign mappings, Looker explore caches: these are your weather. AEs see weather as bugs. You see it as Tuesday.

And you're a politician without portfolio. You broker disagreements between executives who outrank you, using data that they didn't ask for and don't entirely trust, on deadlines that aren't yours.

What Good Looks Like at 6 Months

If you're new in the seat, these are the leading indicators that you're going to make it:

  1. You can name the three biggest data quality leaks in the org. Out loud, without notes, in the order of revenue impact.
  2. You've killed at least one recurring report nobody read. Bonus points if it was a CRO favorite that turned out to be vanity.
  3. The forecast call runs 15 minutes shorter than when you started. Because the questions get answered before the call, not during it.
  4. At least one AE has stopped Slacking you for things they can self-serve. They learned because you built a saved report and stopped answering DMs that the report would answer.
  5. Finance trusts your pipeline coverage number. Trust here means they stop asking "is this real" and start asking "what does it imply."

If you're hitting three of those five at six months, you're on track. If you're hitting zero, the role isn't broken; your scope is. Talk to your manager about what you've actually been hired to fix versus what you've been firefighting.

The Stack You'll Touch Every Day

The tools change every two years. The categories don't. Here's what's typically on a RevOps Manager's belt at a $5M-$200M ARR B2B SaaS:

  • CRM: Salesforce or HubSpot. System of record for accounts, contacts, opportunities. Salesforce wins on customization and stage logic; HubSpot wins on speed-to-setup and Marketing integration. Either way, it's where 60% of your data quality fights happen.
  • Forecasting layer: Clari or BoostUp. Sits on top of the CRM, weights deals with AI signals, surfaces risk. You'll spend more time validating their numbers than trusting them. Treat them as a hypothesis engine, not a source of truth.
  • Sales engagement: Outreach or Salesloft. Sequences, dialer, task logging. The first place that breaks when Marketing renames a campaign. Keep the integration health dashboard pinned.
  • Cross-team execution: Rework. For B2B SaaS teams who need the deals/tasks/projects layer that ties CRM activity to actual delivery (implementation, onboarding, success). Strong fit when Sales hands off to a CS or Services team and the work needs to live somewhere outside the CRM. Useful when your team is mid-size (20-500) and you need cross-functional ops without stitching three tools together.
  • Transformations: dbt. Owns the modeling layer between raw warehouse and BI. If you don't know dbt, learn it. If you do, you're already 11pm-on-call.
  • BI: Looker or Tableau. What executives actually open. Looker for explore-driven self-serve, Tableau for executive dashboards. Either one is downstream of dbt; if dbt is broken, BI is broken.
  • Slack: where most of the actual job happens. The DMs, the threads, the async fights. If your Slack hygiene is bad, your day is bad.

You'll also touch Excel and Google Sheets more than you'd like to admit. Sometimes the fastest way to reconcile three numbers is to dump them into a sheet, eyeball the deltas, and move on. Don't apologize for it.

Closing

If this Tuesday (the diagnosis, the brokering, the 11pm SQL, the holding email to the CFO, the comp-impact argument with the SDRs) sounds energizing, you're built for the role. The job isn't easy, but it's coherent. There's a reason for every part of the day, and the parts compound.

If it sounds like a punishment, the JD lied to you. There's no shame in stepping out before the next forecast call. RevOps doesn't need martyrs; it needs translators who can sleep.

For the JD that this article unmasks, see the companion Revenue Operations Manager Job Description. The gap between the JD and this Tuesday is where most RevOps hires fail in their first 90 days. Read both. Decide.

Learn More