Most B2B SaaS companies don't set up RevOps — they accumulate it. A CRM gets picked early, a few spreadsheets multiply, someone gets the title of "Sales Ops" and spends their time fixing data, and eventually leadership realises that nobody actually trusts the forecast.
This guide is for companies that want to build it properly from the start — or fix what's already broken.
Step 1: Define your ICP before you touch any tooling
The most common RevOps mistake is configuring systems before you know who you're selling to. Your CRM stages, your lead scoring, your qualification criteria, your territory model — all of it depends on a clear Ideal Customer Profile.
Your ICP should answer:
- Firmographic: Company size, industry, geography, funding stage, tech stack
- Behavioural: What triggers make them ready to buy? What does the buying committee look like?
- Value: What problems do they share? What does success look like for them?
Practical tip: Start with your 10 best customers. What do they have in common? That's your ICP — not a committee exercise, but a pattern recognition exercise on real data.
Step 2: Design your pipeline stages before configuring your CRM
Your pipeline stages should reflect buyer behaviour, not seller activity. "Proposal Sent" is a seller activity. "Evaluating Alternatives" is a buyer stage. The difference matters because one tells you what your rep did; the other tells you where the deal actually is.
A clean B2B SaaS pipeline typically has 5–7 stages:
- Qualified Lead — meets ICP criteria, shows intent
- Discovery — first conversation, pain confirmed
- Evaluation — actively assessing your solution
- Proposal — commercial conversation started
- Negotiation — terms being finalised
- Closed Won / Closed Lost — with a mandatory loss reason
Define entry and exit criteria for each stage. A stage without criteria is just a label — and labels don't forecast.
Step 3: Set up your CRM with the end in mind
Your CRM configuration should make it easy for reps to do the right thing and hard to do the wrong thing. That means:
- Required fields at stage transitions (not optional, not a reminder — required)
- Clean object structure: Accounts, Contacts, Deals — not everything crammed onto one object
- Consistent naming conventions that don't require a decoder ring
- Automation for the repetitive: lead routing, task creation, stage transitions triggered by activity
The CRM should reflect reality. If your reps treat it as a reporting tool they fill in on Fridays, your data is fiction and your forecast is guesswork.
Step 4: Build reporting that changes behaviour
Most RevOps teams build dashboards. Far fewer build reporting that actually changes what people do.
The reports that matter most at early and growth stage:
Pipeline health
How much pipeline do you have by stage, by rep, by segment? Is it enough to hit the quarter? Where are deals stalling?
Conversion rates by stage
What percentage of leads reach Discovery? Discovery to Evaluation? This tells you where you're losing — and whether the problem is top of funnel volume or mid-funnel execution.
Deal velocity
How long does a deal spend in each stage on average? Deals that sit in Evaluation for 60 days when your average is 14 are either going to die or need an intervention.
Win/loss analysis
Why do you win? Why do you lose? If you don't have clean loss reason data, you're flying blind on both product and sales.
Step 5: Define the Marketing-to-Sales handoff
The handoff between Marketing and Sales is where most revenue leaks. Leads get passed without context. Sales don't follow up. Marketing blames Sales. Sales blames lead quality.
A clean handoff requires:
- An agreed definition of a Sales-Qualified Lead (SQL) — not a feeling, a set of criteria
- SLA on response time (the data on speed-to-lead impact is overwhelming — every hour matters)
- Context transferred with the lead: what they engaged with, what they said, what problem they described
- A feedback loop: Sales tells Marketing which leads converted and why, not just complained about quality
Step 6: Don't forget Customer Success
RevOps ends at closed won in most early-stage companies. That's a mistake — especially in SaaS where expansion and retention drive as much revenue as new business.
CS needs to be in your RevOps infrastructure from the start:
- Handoff from Sales to CS with deal context, commitments made, and stakeholder map
- Health scoring that flags at-risk accounts before they churn
- Expansion triggers built into your CRM so opportunities don't get missed
- NRR tracked alongside new ARR — because growth without retention is a leaky bucket
The RevOps mantra: Build systems that make the right behaviour the easiest behaviour. If your reps have to work harder to do things correctly than incorrectly, you've built the wrong system.
The mistake most companies make
They buy tools before they have process. They configure CRMs before they know their stages. They hire a RevOps person and task them with "fixing the CRM" when the real problem is that nobody agrees on what good looks like.
RevOps is a discipline before it's a tech stack. Get the thinking right first. The tools will follow.
