Sales Operations Management: The Layer Between the Plan and the Number

Two reports of the same quarter put a seller on opposite sides of quota. One counted bookings by signature date, the other by the date revenue was recognised, and a deal signed in the last week of the period fell either side of the line depending on which rule was applied. Nobody made a mistake, and the seller's commission depends on which report wins.

Settling that argument, permanently and in writing, is sales operations management. The function is often described as reporting and tooling, which is the visible part. The load-bearing part is that it owns the arithmetic that turns a plan into a number somebody is paid against, and the definitions underneath it.

The sales operations role is usually described by its job description; what follows is what it owns once the tooling questions are set aside.

Sales operations org structure: centralised, embedded, or one person with a queue

Sales operations responsibilities, in the order they get picked up

The sales operations skills that decide the outcome are query literacy, a tolerance for definition work, and the willingness to publish a number that contradicts a favourite report.

The duties list on a sales operations job posting collapses into those four, plus whatever reporting the last quarter demanded.

What the function owns

Four areas come up consistently, and they are more specific than the alignment language the category usually attracts.

The plan mechanics. Quota setting and the model behind it, territory boundaries and the file that holds them, account assignment, and the credit rules that decide who gets paid when two sellers touched a deal. These are the rules that make an attainment figure mean the same thing twice, and they are the ones most often carried in somebody's head until the first disputed commission. Quota attainment is only interpretable once the plan behind it is written down, which is why the plan is the function's artefact rather than the number.

The process. The forecast meeting schedule and what a seller has to bring to it, the stage criteria and their enforcement, deal desk and approval routing for non-standard terms, and the operating rhythm of the quarter. The distinction that matters is between owning the forecast, which belongs to the sales leader, and owning the forecast process, which belongs here. Confusing the two produces a function that is blamed for a miss it had no authority to prevent.

The systems the sales organisation runs on. CRM configuration, the fields that the definitions depend on, the sales tool stack and its administration, and the routing that decides who works what. Most of the visible work sits here and most of the value does not, which is a difficult thing for the function to explain in its first year.

The reporting everybody argues from. Not dashboards for their own sake, but a small number of figures whose definitions are stable enough that a movement in them means something. A reporting layer built on drifting definitions produces confident answers to questions nobody can reconstruct, and rebuilding trust in a number is considerably more expensive than defining it correctly the first time.

Revenue operationsAcross the whole revenue motion

• Shared definitions across marketing, sales and success
• Systems and integrations between them
• Data quality standards and maintenance
• Attribution and the join between functions

Sales operationsInside the sales organisation

• Quota, territory and credit rules
• Forecast process and stage enforcement
• Deal desk, approvals and non-standard terms
• CRM configuration for the selling motion
• The reporting sales leadership argues from

The sales managerOwns the outcome, not the machinery

• Coverage decisions inside the assigned patch
• The forecast itself and its honesty
• Coaching and deal strategy
• Hiring, ramp and performance decisions

At smaller companies these are one person and the boundaries are academic. They stop being academic at the point where two of the three exist, because the work that falls between two named functions is the work that reliably goes undone. Where a separate development team exists, its manager owns a further set of decisions again, and what the SDR manager role owns is the cleanest statement of that boundary.

The sales operations manager role

The title covers a wide band, and the level matters more than the label.

At the individual contributor end, a sales operations manager runs the CRM, maintains the territory and quota files, builds the reporting, and administers the tool stack. At the leadership end the same title owns the design of the plan itself: how quotas are set, how territories are carved, what the comp plan rewards, and which arguments the process is built to prevent. The first version executes decisions made elsewhere. The second version makes them.

Hiring the first without meaning to hire the second is the common failure, and it shows up as a function that produces excellent reports and changes nothing, because every recommendation it makes requires an authority it was not given. The question that separates the two in an interview is what the candidate would change about a quota model they inherit, and whether they can explain the trade the change makes rather than only its benefit.

Reporting lines vary and none of them is obviously right. Under the sales leader, the function is responsive and risks becoming a service desk for whatever the leader wants shown. Under finance, it is independent and risks being read as an audit function the sellers route around. The arrangement that works is less about the line and more about whether the function can decline a request, which is the same authority test as the level question above.

What a first hire should do first

The temptation is to build the dashboard everybody asked for. It is visible, it is quick, and it makes people happy for about a month.

The higher-value opening move is an inventory of the definitions currently in use, gathered from the artefacts rather than from what people say. Take the last two quarters of reported attainment and reconstruct it from the raw records. Where the reconstruction differs from the published figure, the gap is a definition nobody wrote down, and the list of those gaps is the actual work plan. It also arrives with evidence attached, which is what buys a new function the standing to change anything.

Publishing that reconstruction is the second move, and it is a political act as much as an analytical one. A leader who has been quoting an attainment figure weekly will not enjoy learning it was computed three ways, so the reconstruction lands better as a dated statement of which rule the function will apply from now on than as a correction of what was said before. Choose the rule that is cheapest to compute repeatedly rather than the one that is theoretically best, because a rule that requires manual reconciliation each quarter will quietly stop being applied by the third one.

1. Step 1Before the period: set and publish the plan

   Quotas, territories, credit rules and the stage criteria, dated, with the previous version still readable.

2. Step 2During: enforce the definitions quietly

   Stage entry, deal desk approvals and routing, so that the reporting layer is measuring what it claims to.

3. Step 3Weekly: run the forecast process

   Own the schedule and the required inputs. The number belongs to the sales leader, the process does not.

4. Step 4After: reconcile and record what changed

   Attainment against payroll, and every definition change dated on the chart it moves.

The failure mode worth naming in advance

Every sales operations function is at risk of becoming a report factory, and the mechanism is ordinary rather than dramatic.

Requests arrive individually and each one is reasonable. A leader wants a cut by segment, finance wants a reconciliation, a manager wants a view of their team. None can be refused without seeming obstructive, and each takes a couple of hours. Within two quarters the function's entire capacity is spent answering questions, none of the underlying definitions has been fixed, and the reports themselves are increasingly built on fields nobody maintains.

The defence is structural rather than personal. Publish a small standard set on a fixed schedule, make anything outside it a request with a stated cost, and reserve a fixed share of capacity for definition and data work that nobody asked for. Data hygiene is the part of that reserve which is least visible and most load-bearing, because a segment query against inconsistent fields reaches a population nobody can reconstruct afterwards.

The second defence is a retirement rule for reports. Anything built on request gets an owner and a review date, and a report nobody has opened in two quarters is switched off rather than maintained. Without that, the function's standing workload only ever grows, because every report ever requested is still being produced for somebody who has forgotten they asked.

Where outbound intersects with the function

Outbound is the part of the motion that most rewards this discipline, because every input to it is chosen rather than inherited.

The segment is a decision, the message is a decision, and both are measurable against a clean population if the definitions hold. When they do not, a campaign result teaches nothing whether it worked or not, since the population it reached cannot be described afterwards. The two definitions that carry the most weight are what counts as a qualified conversation and where an opportunity enters the pipeline, covered in what qualified has to mean when money depends on it and pipeline stages that earn their place. Cost figures inherit the same problem, which is the substance of why cost per lead misleads even when computed carefully.

Our own commercial standard is the same instrument this page recommends. Meetings are qualified against criteria agreed in writing before launch, and budget, timing and authority are never billing conditions, so the definition exists before anybody has a result to read off it. Campaigns carry one message with no bumps and no thread replies, which has an operational consequence that gets underestimated: one message per campaign means one clean attribution per contact, so the question of which message produced which reply never requires untangling a thread.

The short version

Sales operations management owns the plan mechanics, the process, the sales systems and the reporting layer. The distinctive part is the arithmetic that makes an individual seller's number computable and defensible, which is why quota, territory and credit rules are the function's core artefact rather than its paperwork.

Keep the boundaries explicit. Revenue operations spans the whole motion, sales operations sits inside the sales organisation, and the sales manager owns the outcome rather than the machinery. Hire for the level you actually need, since a function given no authority produces recommendations and no changes. Start with an inventory of the definitions in use, reconstructed from records rather than collected from opinions, then fix exactly one of them completely. Reserve capacity permanently for the work nobody requests, or the function becomes a report factory within two quarters.

Where the constraint is the number of qualified conversations rather than the machinery around them, no amount of operating discipline creates supply: see what a first campaign produces.