Ask a sales team how many open deals they have and you will usually get two numbers. One is in the CRM. The other is in somebody's head, or in a spreadsheet on their laptop, and it is the one they actually believe.
The gap between those two numbers is the whole reason sales operations exists. Somebody has to keep the system true enough that the team uses it, the owner can read it, and the forecast is worth the meeting it takes up. In a large company that somebody has a job title. In most companies it is nobody, and the CRM slowly turns into a list of good intentions from last spring.
This post covers what sales ops is, what a sales operations specialist does day to day, how the role differs from a sales operations analyst, a sales operations manager, RevOps and sales enablement, the KPIs a sales ops person watches, and which parts of the work an AI can do. We make Sanaf, an AI employee that takes on this job among others, so we have an interest in the last part. The product turns up near the end. The rest would be true without us.
The short answer
Sales operations (sales ops) is the work that keeps a sales team's systems, data and process usable: cleaning and maintaining the CRM, keeping deal stages and fields consistent, routing new leads to the right person, building pipeline and forecast reports, and looking after the sales tools. A sales operations specialist is the person, or increasingly the AI, who does that maintenance week in and week out so the sellers can sell and the numbers can be trusted.
What is sales ops, in plain terms?
The easiest way to understand sales ops is to look at what goes wrong without it.
A rep closes a deal and forgets to move it to won. Another rep logs the same company twice, once as "Acme" and once as "Acme Plumbing LLC". A third rep has twenty deals sitting in "proposal sent" with close dates that passed in July. Nobody notices any single one of these. Six months later the pipeline report says you have twice as much business coming as you actually do, the owner stops trusting it, and the weekly sales meeting turns into people reading numbers off their own notes.
Sales ops is the job of preventing that drift. It is unglamorous. Most of it is checking, fixing, chasing and reporting, and none of it closes a deal directly. What it does is make every other part of selling work: the forecast, the commission sums, the handoff to the account manager, the owner's view of the next quarter.
There is also a design side. Someone has to decide what the deal stages are, what each one means, which fields are required, how leads get assigned, and which tools the team uses. In bigger companies that design work is a large part of the role. In smaller teams it gets decided once, badly, and then nobody revisits it.
What does a sales operations specialist do?
A sales operations specialist is the hands-on version of the role. If you read a sales operations job description for this title, the duties usually fall into five groups.
1. CRM hygiene
CRM hygiene means keeping the records accurate and complete. In practice that is finding and merging duplicate contacts and companies, filling in missing fields, fixing bad email addresses and phone numbers, closing out deals that are clearly dead, and making sure every open deal has a next step and a close date that has not already passed.
None of it is hard. All of it is tedious, and it is the first thing to slip when the team gets busy. That is why pipeline hygiene is a weekly habit or it is nothing. A CRM cleaned once a year is dirty again by the second month.
2. Keeping the process consistent
Three reps will define "qualified" three different ways unless somebody writes it down and keeps checking. The specialist keeps the stages, required fields and picklists consistent, spots deals that skipped a stage, and notices when one rep's pipeline looks nothing like everyone else's. That last one is usually a data problem before it is a performance problem.
3. Lead routing
Lead routing is the set of rules that decides who gets a new lead: by territory, by industry, by deal size, by round robin, or by whoever owns the existing account. Without rules, leads go to whoever saw the notification first, and the slow-to-respond ones sit unassigned for a day. The specialist sets the rules up in the CRM, watches for leads that fell through, and fixes the edge cases, like the lead from a company that is already a customer.
4. Reporting
Every sales team has a weekly report someone builds by hand: the pipeline by stage, deals closing this month, activity by rep, wins and losses. The specialist builds it, or better, builds it so it builds itself, and checks that the numbers behind it are real before anyone reads it.
5. Looking after the tools
The CRM, the email sequencing tool, the dialer, the e-signature tool, the spreadsheet everyone still uses. Somebody has to manage the seats, keep the integrations running, and answer "why didn't this sync?" That is usually sales ops too.
Sales operations specialist vs. analyst vs. manager
Job boards use these three titles loosely, and they overlap a lot. Roughly:
| Sales operations specialist | Sales operations analyst | Sales operations manager | |
|---|---|---|---|
| Main focus | Doing the maintenance | Reading the data | Owning the function |
| Typical work | CRM cleanup, lead routing, report building, tool admin | Forecast models, territory and quota analysis, win-rate and cycle-length studies | Process design, tool choices, comp plans, working with sales leadership |
| Question they answer | Is the data right? | What does the data say? | How should the team run? |
| Who has one | Teams with a few reps and up | Teams big enough to have a forecast to model | Teams big enough to have a sales ops team |
In a company with one sales ops person, that person does all three. In a company with none, the owner or the sales lead does a thin slice of each on a Sunday night.
Sales ops vs. RevOps vs. sales enablement
Two other titles come up whenever people search "what is sales ops", and they are easy to mix up.
What is RevOps?
Revenue operations, or RevOps, is sales ops stretched across the whole customer lifecycle. It joins sales operations, marketing operations and customer success operations under one function so that the data, tools and handoffs line up from first visit to renewal. The question in revops vs sales ops is mostly one of scope: sales ops looks after the sales team's systems, and RevOps looks after the systems for everyone who touches revenue. Many companies that say RevOps mean "sales ops, plus the marketing automation tool".
Sales operations vs. sales enablement
Sales enablement is about the sellers rather than the systems: training, onboarding new reps, call coaching, the pitch deck, battle cards against competitors, and the content reps send to buyers. Sales ops makes sure the CRM and the process work. Enablement makes sure the people know how to use them and what to say. In small teams both land on the same sales lead, which is part of why both get done badly.
Sales operations KPIs worth watching
Sales operations KPIs fall into two groups: numbers about the pipeline, and numbers about whether the pipeline data can be believed. The second group gets skipped, and it matters more, because every number in the first group is only as good as the data under it.
Data you can believe:
- Share of open deals with a next step and a future close date
- Share of open deals untouched in the last 30 days
- Duplicate contact and company rate
- Time from a new lead arriving to its being assigned, and to its first touch
- Share of required fields actually filled in
The pipeline itself:
- Pipeline value by stage, and how it changed since last week
- Win rate, by stage and by rep
- Average sales cycle length
- Forecast accuracy: what the team said would close against what did
If you track only one hygiene number, make it the first one. A deal with no next step and a close date in the past is somebody's hope, and a pipeline full of them is a forecast nobody should act on.
What happens when nobody owns sales ops
Most companies under a certain size never hire a sales operations specialist, even though they need the work done. The work is too small to justify a full salary and too constant to fit around the other jobs on someone's plate. So it gets done in bursts: a big cleanup before a board meeting or a fundraise, then six months of drift.
You can usually tell from three signs. The weekly sales report takes somebody a morning to build. Reps keep their real pipeline in a notebook or a private spreadsheet, because the CRM is a chore that gives nothing back. And nobody uses the forecast, because the last three were wrong in the same direction.
Each one has a cost, even if nobody adds it up. The morning spent on the report is a morning not spent selling. The private spreadsheet means the owner cannot see the business. The ignored forecast means hiring and spending decisions get made on gut feel, which is fine until the gut is wrong in a slow month.
Which parts of sales operations can AI do?
This is where the job splits cleanly in two.
The maintenance half is a very good fit for an AI. Finding duplicates, spotting deals with no next step, flagging close dates in the past, filling gaps from the emails and meeting notes already in your systems, assigning new leads by the rules you wrote, and building the Monday report are all rule-shaped and repetitive. They also need doing every week, which is exactly what people stop doing when they get busy.
The judgment half is not. Whether a deal is really at the proposal stage, what it is worth, who should own an account, and whether two records are the same customer are claims about what is happening with a real buyer. The rep or the owner knows things the CRM does not. So a sensible AI sales operations setup draws the line here:
- It cleans, fills and flags, and it can do that on its own once you are happy with how it does it.
- It proposes merges, with the evidence, and a person confirms anything that is not an obvious copy. A wrong merge throws away history you cannot get back.
- It chases the record, not the rep. A short message in the team's channel asking for the missing close date on a named deal, while the deal is fresh, lands differently from a manager asking at the end of the quarter.
- It never changes a deal's stage, value or owner without the rep or the owner agreeing.
- It keeps a record of every change it made, so anyone can see what moved and why.
The other thing an AI changes is cost. A full sales operations hire is a salary. An AI doing the maintenance half is priced on the work it does, which makes it worth having for a team of three reps, where a person in the role never would be.
How to start, if you have never had sales ops
- Write down your stages and what each one means, in a sentence each. If the team argues about it, good: that argument was always happening in the data.
- Pick the required fields for an open deal. Next step and close date at the very least.
- Run one cleanup pass and look at what it finds: duplicates, dead deals, missing fields. Fix the obvious ones and send the rest to their owners.
- Write your lead routing rules down, even if they are "round robin, except existing customers go to their account owner".
- Then make it weekly. The first cleanup is the easy part. Keeping it clean every Monday after that is the part a busy team never manages on its own.
Where Sanaf fits
Sanaf is an AI employee. It works in your team's Slack or Microsoft Teams and inside the apps you connect. Sales operations is one of the jobs it can take on, and the AI sales operations page lays out exactly what that job covers. It works with HubSpot, Salesforce and Pipedrive, along with Google Sheets and Gmail.
Day to day, it does the maintenance half from the section above. It cleans, merges and enriches CRM records. It chases the missing next steps, close dates and notes in the team's channel. It keeps stages, fields and sequences consistent, routes new leads by the rules you set, spots deals that have stalled or slipped, and builds the weekly pipeline and activity report so nobody spends a morning on it. You can ask it in a sentence, such as "which deals closing this month have no next step?", and it answers in the same thread.
The limits are the ones above. It tidies and reports. It does not change a deal's value, stage or owner without the rep or the owner agreeing, and merging two records that might be two real customers is always a person's call. Anything it cannot undo asks first, and every job leaves a work log the team can read. If you want it running more than one job, the AI chief of staff post explains how Sanaf hands work to agents that report to it.
On price, you start with $120 of free credit and no card, and that free credit lasts 30 days. After that you buy credits from $20 when you need them, and credit you buy never expires. You pay for the work, not per seat, so adding the whole sales team to the channel costs nothing. The pricing page has the details, and hiring an AI employee shows what the first day looks like.
A good first test is the cleanup pass from step 3. Connect the CRM, ask for the list of open deals with no next step, and see how long that list is.
Questions people ask about sales operations
What is sales ops?
Sales ops, short for sales operations, is the work that keeps a sales team's systems, data and process usable. It covers CRM hygiene, consistent deal stages and fields, lead routing, pipeline and forecast reporting, and managing the sales tools.
What does a sales operations specialist do?
A sales operations specialist maintains the sales team's CRM and process. They clean and merge records, make sure deals have next steps and close dates, set up and fix lead routing, build the weekly pipeline and activity reports, and look after the sales tools and integrations.
What is the difference between a sales operations analyst and a sales operations manager?
A sales operations analyst works with the data: forecast models, territory and quota analysis, win rates and sales cycle length. A sales operations manager owns the function: process design, tool choices, compensation plans and working with sales leadership. In small teams one person often does both, along with the specialist's maintenance work.
What is RevOps, and how is it different from sales ops?
RevOps, or revenue operations, joins sales operations, marketing operations and customer success operations into one function so the data, tools and handoffs line up across the whole customer lifecycle. Sales ops covers the sales team's systems; RevOps covers everything that touches revenue.
What is the difference between sales operations and sales enablement?
Sales operations looks after the systems and data: the CRM, the process, routing and reporting. Sales enablement looks after the sellers: training, onboarding, coaching, pitch materials and competitor battle cards.
What is CRM hygiene?
CRM hygiene is the habit of keeping CRM data accurate and complete: merging duplicates, filling missing fields, fixing bad contact details, closing dead deals, and making sure every open deal has a next step and a future close date. It only works as a regular routine, because a CRM starts to drift again within weeks of a cleanup.
What are the most important sales operations KPIs?
The pipeline KPIs most teams watch are pipeline value by stage, win rate, sales cycle length and forecast accuracy. The data-quality KPIs that sit under them are the share of open deals with a next step and future close date, the share untouched in 30 days, the duplicate rate, and the time from a new lead arriving to its first touch.
Can AI do sales operations?
AI can do the maintenance half of sales operations well: finding duplicates, flagging deals with no next step or a past close date, filling gaps from existing emails and notes, routing leads by your rules, and building weekly reports. Decisions about a deal's stage, value or owner, and merges that might combine two real customers, should stay with a person. Sanaf works this way by default.