Introduction
There is a management style that consists almost entirely of asking about the number. What is the forecast, where are we against quota, what is coming in this week. It feels like leadership because it is focused on the thing that matters, and it produces a reliable pattern: mild reassurance for eleven weeks, then a surprise in week twelve.
The problem is not the attention on the number. It is that the number cannot be managed. It is a lagging outcome, visible only after the work that produced it is finished, and pressure applied to it late in the quarter arrives after every decision that could have changed it. This article is about what to manage instead, what that looks like in practice, and the test that keeps it from turning into the bureaucracy everyone fears.
Why the Number Cannot Be Managed
Take a shortfall in week eight of a thirteen-week quarter. The manager sees it and applies pressure. What can the team actually do?
Deals that will close this quarter are already in the pipeline; deals created now will mostly close after it ends. The rep can accelerate a few late-stage opportunities, usually by discounting, and can chase everything else, usually with diminishing returns. The interventions available at week eight are weak, and the two most common ones — discounting and chasing — both damage next quarter.
The decisions that determined week eight were made in weeks one to five: how much pipeline got created, which deals were qualified in that should have been out, whether next steps were set. Those decisions were manageable when they happened. By week eight you are managing their consequences.
This is the whole argument. Managing by the number means intervening at the point where interventions are weakest.
What Managing by Process Actually Means
It is a specific set of things, not a philosophy. Four components carry most of the weight.
Stage exit criteria a rep can verify
Every stage needs a condition that is checkable rather than felt. “Customer is interested” is not one. “Next meeting booked with the budget owner” is. The test is whether two people would independently place the same deal in the same stage — if not, your stages are labels rather than criteria, and every number derived from them is soft.
This is the highest-leverage item on the list, because forecasting, coaching and pipeline review all inherit their reliability from it.
A next-step rule
One rule, applied without exception: every open deal past qualification has a scheduled next step with a date. Deals without one are not being progressed by anybody, and they are the largest single category of pipeline that is forecast as alive while being effectively dead.
The reason to enforce exactly this rule and not twelve others is that it is cheap for the rep, unambiguous, and directly predictive.
A cadence built on exceptions, not status
The weekly meeting should be about the deals that broke a rule — no next step, past close date, stage age beyond the norm, a single contact in a large opportunity — not a walk through everything in alphabetical order.
Status meetings scale linearly with pipeline size and teach reps to prepare a narrative. Exception meetings stay the same length as the team grows and teach reps to keep the rules, because the rules are what put them on the list.
Coaching one skill at a time
Feedback across five dimensions changes nothing. One skill, worked for several weeks, until it shows up in recordings. Then the next. The one-on-one is the vehicle for this, which is why it should not be a second pipeline review.
The Objection: This Becomes Bureaucracy
The objection is legitimate and worth taking seriously, because process management does degenerate. It happens when steps accumulate: each one is added for a good reason, nobody removes any, and after two years the team spends more time reporting on selling than selling.
The test that prevents this has two parts, and a step must pass both:
- It changes a decision. Name the decision. If a field, a rule or a meeting exists and no decision would change based on it, it is documentation, not process.
- Someone can say what happens when it is skipped. If the honest answer is “nothing,” the step is already optional in practice, and keeping it nominally mandatory only teaches the team that rules are theatre.
Run that test against your existing process once a quarter and delete what fails. A process that only ever grows is on its way to becoming the thing its critics say it is.
A second guard: keep the number of enforced rules small. Three that everyone follows beat ten that are followed selectively, because selective enforcement destroys the data — you can no longer tell whether a missing next step means a stalled deal or a rule nobody bothers with.
Worked Example: Two Managers, One Shortfall
An illustrative example. Two teams, same size, same quota of $600,000 for the quarter. At week eight both are at $310,000 closed with a weighted pipeline that suggests a shortfall.
Manager A manages the number. She calls a mid-quarter push. Daily forecast updates, pressure on every late-stage deal, approval for discounting to bring deals forward. The quarter lands at $540,000 — better than the trajectory suggested, and two deals from next quarter were pulled in at reduced value to get there. Next quarter opens with a thinner pipeline and a customer base that has learned to wait for quarter-end. The pattern repeats, slightly worse each time.
Manager B manages the process. Same data, different questions. He looks at inputs and finds three things: qualified pipeline created in weeks one to six was about 40 percent below normal; 55 percent of open deals have no scheduled next step; and four deals in late stages have a single contact each.
His interventions follow from that. Two days of concentrated pipeline generation, aimed at the shortest-cycle segment. The next-step rule enforced on every open deal by end of week nine, which surfaces eleven deals nobody was actually working — six get closed lost, five get a real next action. Multi-threading pushed on the four single-threaded deals, one of which turns out to have lost its sponsor a month earlier.
His quarter lands at $505,000 — worse than Manager A’s, and here is the honest part: managing by process does not rescue a quarter that was lost in week five. What it does is fix the causes. His next quarter opens with normal pipeline coverage, a forecast built on deals that have next steps, and no customers trained to wait for a discount.
By the end of the following quarter the two teams are not comparable, and nothing in Manager A’s approach was ever going to change that, because she was managing an output at the only point in the cycle where it could not be influenced.
Making the Inputs Visible
Managing by process depends on being able to see the inputs, which means the CRM has to reflect reality. That is a real dependency: “share of open deals with a next step” is only a management signal if the field is maintained, and a manager who has to assemble the exception list by hand every week will stop doing it by week four.
This is where automation earns its keep rather than adding to the bureaucracy. SalesBond audits how the pipeline and CRM are structured — stage logic, required and missing fields, data quality, next-step rules — and turns the findings into a prioritised Fix Plan, with nothing changed until an authorised user approves it. Deal Health then watches active deals continuously and detects stalled deals, missing next steps and overdue activities, feeding an at-risk queue, so the exception list arrives already built.
FAQ
What does managing by process mean in sales?
Managing the inputs that produce revenue — pipeline creation, qualification, next-step discipline, stage integrity — rather than the revenue figure itself, which can only be observed after the fact.
Does this mean I should stop looking at the forecast?
No. Look at it, and use it as a trigger to examine inputs rather than as something to apply pressure to. The forecast is a symptom; the inputs are where the treatment is.
How many process rules should a team have?
Few enough that all of them are actually enforced. Three well-kept rules produce more usable data than ten that are followed when convenient.
How do I stop process management becoming bureaucracy?
Apply the two-part test quarterly: each step must change a decision, and someone must be able to say what happens when it is skipped. Delete anything that fails either part.
Does this work for a small team?
It matters more on a small team, where one badly qualified deal is a larger share of the quarter and there is no volume to average out mistakes.
Conclusion
The number is the score, and the score cannot be coached. What can be managed is the small set of inputs that produce it: stages with criteria a rep can verify, a next-step rule enforced without exception, a weekly cadence built on exceptions rather than status, and coaching on one skill at a time.
Keep it honest by testing every process step against two questions — does it change a decision, and does anything happen when it is skipped — and deleting what fails. This week, run the exception list rather than the status list in your pipeline meeting: same half hour, and it will surface deals the walkthrough has been quietly skipping for months.
