Revenue Operations Playbook

Executive summary

A revenue operations playbook is a set of definitions, data, and weekly routines that make the revenue number predictable before anyone tries to make it bigger. The core claim of this one is simple: fix forecast accuracy before you set growth targets. A target set on top of a forecast you cannot trust is a wish with a quota attached.

The reasoning is practical. Every growth lever in a mid-market company depends on a number that someone has to predict. Sales capacity depends on attainment, pipeline coverage depends on win rates, and retention depends on which customers renew. If the people making those predictions are consistently 8 to 15 percent off, the plan built on them is off by the same amount, and the miss lands on cash, hiring, and covenants. If you want to see how that plays out for a lender, read the covenant compliance headroom guide.

The playbook has four moves:

1. Measure the forecast. Compute bias and error for the last four to eight periods, by segment and by owner.
2. Fix the definitions. One definition of qualified pipeline, one of a booking, one of a renewal, one of a churned customer.
3. Build the plan bottom-up. Retained base from net revenue retention, new revenue from capacity and pipeline, price from realized rather than list increases.
4. Set the target last. Treat the gap between the bottom-up number and the ambition as a list of named decisions, not a number to be spread across the sales team.

The sections below give the diagnostic, the framework, the 90-day plan, templates, metrics, and an illustrative worked example. Three deeper methods have their own pages: sales capacity planning, pipeline coverage and forecast accuracy, and net revenue retention.

When to use this

Use this playbook when:

• The company has missed its revenue forecast in two or more of the last four quarters, in either direction.
• The board or sponsor has set a growth target and nobody can show the arithmetic that gets there.
• Sales, finance, and customer success each report a different revenue or pipeline number.
• A new sales leader, CFO, or operating partner has arrived and needs a baseline in 30 days.
• You are about to hire sales capacity and want to know whether the pipeline can feed it.
• You are preparing for a sale and a buyer will test the forecast process during diligence.

Do not start here if bookings and invoiced revenue cannot be reconciled to the general ledger. Fix that first. Every measurement below assumes that the closed-won number and the revenue number are two views of the same set of deals.

This playbook describes operating method. It does not give accounting advice on revenue recognition. Follow your own accounting policies and involve your auditors or advisers where the treatment of contracts, discounts, or multi-element arrangements is in question.

Diagnostic questions

Answer these before changing anything. Most can be answered from the CRM, the ERP, and last year's board packs.

1. For each of the last eight periods, what did you forecast at the start, what did you forecast at the midpoint, and what actually happened?
2. Is the error mostly in one direction (bias) or scattered (noise)?
3. Which segment, product line, or sales owner is responsible for most of the error?
4. Is there a written definition of qualified pipeline, and does every rep apply it the same way?
5. What is the stage-by-stage conversion rate for deals created in each of the last four quarters, measured from your own CRM history?
6. How much of the open pipeline is older than your typical sales cycle, and is anyone removing it?
7. What is the quota attainment distribution across reps, and what share of reps made quota last year?
8. How long does a new rep take to reach full productivity, measured from your own hires?
9. What are your gross and net revenue retention numbers, and who owns each?
10. How much of last year's revenue growth came from price, how much from volume, how much from new customers, and how much from expansion?
11. Is the price you quoted the price you collected? See the price realization rate drift guide for the check.
12. Does the revenue plan tie to the cash forecast, and does the cash forecast use the same customer-level assumptions?

Reading the answers: if questions 1 to 3 show a consistent over-forecast, the problem is optimism in stage definitions or deal dating. If question 5 shows that conversion varies a lot across quarters, the pipeline has no stable definition. If question 9 cannot be answered, you do not know whether growth is coming from new customers or being offset by leakage from old ones.

The framework: forecast, capacity, pipeline, retention

Revenue operations is easiest to run when you treat revenue as four linked layers. Each layer has one question, one owner, and one number.

The layers run in a deliberate order. The forecast comes first because it is the instrument you use to judge the other three. If you cannot tell whether a coverage ratio of 4x is safe or reckless, you cannot judge pipeline. Capacity comes second because it tells you the ceiling for new bookings given today's team. Pipeline tells you whether that ceiling is reachable. Retention sets the floor, because a base that leaks makes every new dollar cost more.

A fifth input runs through all four: price. A list price increase that does not reach the invoice is not a lever. Track realization separately, using the method in the pricing waterfall guide, and feed the realized number into the plan.

Definitions that must be written down

Most forecast error is definitional. Write these once, in a single page, and get sales, finance, and customer success to sign it:

• Qualified pipeline: the minimum conditions for a deal to count (for example a confirmed budget owner, a defined problem, and a next meeting on the calendar). Your conditions, not a template.
• Stage exit criteria: what evidence moves a deal from one stage to the next. Evidence is something the buyer did, not something the rep believes.
• Close date rule: a deal with no buyer activity in a stated number of days has its close date reset or is closed out.
• Booking: the event that makes a deal count as booked, and whether it requires a signed contract, a purchase order, or both.
• Renewal and churn: when a customer is counted as churned (contract end, last invoice, written notice) and how a downgrade is treated.
• Expansion versus new: whether a new product sold to an existing customer is expansion or new business, and who is credited.

If your agreements, commission plans, or lender documents define any of these terms, the contractual definition governs for those purposes. Operate a management definition alongside it and reconcile the two, and involve counsel or your auditors where the difference matters.

Step-by-step: a 90-day revenue operations plan

Weeks 1 to 2: Baseline the forecast

1. Collect the forecast submitted at the start and midpoint of each of the last eight periods, plus actuals. Owner: FP&A. Output: a forecast-versus-actual table by period.
2. Compute bias (signed error) and error (absolute error) at total, segment, and owner level, using the formulas below. Owner: FP&A. Output: a one-page baseline that states plainly whether the company over-forecasts, under-forecasts, or is noisy.
3. Reconcile CRM closed-won to ERP invoiced revenue for the last four quarters. Owner: Controller with sales operations. Output: a list of deals where the two disagree and why.
4. Write the one-page definitions sheet described above. Owner: CRO and CFO jointly. Output: signed definitions.

Weeks 3 to 4: Clean the pipeline and measure conversion

5. Age out stale deals. Any open deal past its close date with no recent buyer activity gets a new date supported by evidence or is closed. Owner: each sales manager. Output: a cleaned pipeline file.
6. Measure stage conversion from your own history, by cohort. Take deals that entered each stage in a past quarter and count how many reached closed-won. Owner: sales operations. Output: a conversion table by stage, segment, and deal-size band.
7. Compute required coverage for the current quarter from those conversions, and compare it to actual coverage. Owner: sales operations. Output: a coverage gap table.
8. Start a weekly forecast call with a fixed agenda (see templates). Owner: CRO. Output: a committed call, an upside list, and a risk list, each with deal names.

Weeks 5 to 8: Build capacity and retention views

9. Build the capacity model: ramped reps, quota, historical attainment, ramp curve, expected attrition. Owner: sales operations. Output: bookings capacity for the next four quarters.
10. Compute gross and net revenue retention by segment and by cohort. Owner: customer success lead with FP&A. Output: a retention table and a list of the ten largest decreases and increases.
11. Tie price to the plan. Pull realized price by customer or product from invoices and compare to list and quote. Owner: pricing or commercial finance. Output: realized price change for the last year.
12. Review account coverage for the largest customers. A named owner, an agreed next step, and a contact beyond the primary one. See the key account management cadence guide.

Weeks 9 to 12: Build the bottom-up plan and set the target

13. Assemble the bottom-up revenue bridge: opening base, retention and expansion, price, new customers from capacity and pipeline. Owner: FP&A. Output: a plan with each line tied to a driver.
14. Compare the bridge with the board or sponsor ambition. Owner: CEO and CFO. Output: a gap, expressed in dollars and in the assumptions that would have to change.
15. Decide which levers close the gap and by how much, each with an owner and a date: more capacity, more pipeline creation, better retention, higher realized price, or a change in the ambition. Owner: CEO. Output: a decision log.
16. Link the plan to the cash forecast. Revenue timing, billing terms, and collections drive cash. See the 13-week cash flow forecast template. Owner: CFO.

The formulas

Forecast bias and error

Bias = (Forecast − Actual) ÷ Actual
Absolute error = |Forecast − Actual| ÷ Actual
Weighted absolute error across periods = Sum of |Forecast − Actual| ÷ Sum of Actual

Bias tells you direction. A company that over-forecasts by 8 percent on average has a bias of +8 percent. Absolute error tells you size and ignores direction, so errors that cancel each other out do not hide. Report both. Bias can be corrected with definitions and haircuts. Noise needs better inputs.

Required pipeline coverage

Effective yield = Expected closed-won revenue from open pipeline ÷ Open pipeline
Required coverage = 1 ÷ Effective yield
Coverage gap = Actual coverage − Required coverage

Compute effective yield from your own stage-by-stage history, not a rule of thumb. Details are in the pipeline coverage guide.

Capacity

Productive capacity per ramped rep = Quota × Historical attainment
Ramped-equivalent headcount = Σ (rep months at each ramp level × ramp factor) ÷ 12
Bookings capacity = Ramped-equivalent headcount × Productive capacity per ramped rep

Details, including ramp and attrition, are in the sales capacity planning guide.

Retention

Gross revenue retention = (Starting revenue − Contraction − Churn) ÷ Starting revenue
Net revenue retention = (Starting revenue + Expansion − Contraction − Churn) ÷ Starting revenue

Both use a fixed cohort: customers that existed at the start of the period. New customers acquired during the period are excluded. Details are in the net revenue retention guide.

The revenue bridge

Plan revenue = Opening base × Net revenue retention + In-year revenue from new customers

This simple structure forces every plan to answer two questions: what do we assume about the base we have, and what do we assume about the customers we do not have yet. Price and expansion are inside net revenue retention. Capacity and pipeline set the second term.

Worked example (illustrative)

The numbers below are made up to show the method. They are not benchmarks and describe no real company.

A distributor has $60.0M of revenue from the customers it served last year. The board has set a target of 15 percent growth, or $69.0M. Nobody has shown the arithmetic.

Step 1: Baseline the forecast

The finance team collects the last four quarterly forecasts made at the start of each quarter:

These figures are for the new-business bookings forecast. Total bias is +7.8 percent: the team over-forecasts. Weighted absolute error is 12.9 percent, larger than the bias, which means errors in both directions are hiding inside the average. The Q4 miss of 25 percent is the largest, and Q3 was an under-forecast. The pattern suggests late-quarter optimism in deal dating plus a conservative Q3. The team spends its first two weeks on the definitions sheet and a close-date rule.

Step 2: Build the bottom-up bridge

• Opening base: $60.0M.
• Net revenue retention, from the cohort table: 97 percent. This includes expansion and realized price. Retained base: 60.0 × 0.97 = $58.2M.
• In-year revenue from new customers, from the capacity and pipeline model with the new definitions: $4.5M.
• Bottom-up plan: 58.2 + 4.5 = $62.7M, which is 4.5 percent growth.

Step 3: Compare to the ambition

The target is $69.0M. The gap is 69.0 − 62.7 = $6.3M.

What would have to be true to close it with one lever alone:

Neither is credible alone. The CEO and CFO use the table to decide on a mix: hire sales capacity early enough to ramp in-year (the sales capacity guide shows why month of hire matters), fix the largest churn segment, and recover realized price. Each gets an owner, a date, and a dollar estimate. Whatever is left becomes an explicit stretch above the plan, labelled as such, not a quota spread across the team.

Step 4: Apply the forecast haircut

Because the new-business forecast has been 7.8 percent too high on average, the CFO does not use the sales-submitted figure for the cash forecast. The CFO uses the new-customer revenue from the capacity and pipeline model and tracks the gap to the sales number weekly. If bias falls inside a band the executive team agrees, the haircut comes off.

Arithmetic check: 60.0 × 0.97 = 58.2. 58.2 + 4.5 = 62.7. 62.7 ÷ 60.0 = 1.045. 69.0 − 62.7 = 6.3. required new-customer revenue is 4.5 + 6.3 = 10.8, and 10.8 ÷ 4.5 = 2.4. (69.0 − 4.5) ÷ 60.0 = 1.075.

Templates

Forecast accuracy log (copy into a spreadsheet)

Cause tags to use: stage inflation, close date slip, deal size change, lost to no decision, lost to competitor, timing of customer purchase order, definition difference.

Weekly forecast call agenda (45 minutes)

1. Forecast versus last week: committed, upside, and risk, each in dollars (5 min).
2. Deals that moved stage or close date since last week, with the buyer evidence for each (15 min).
3. Deals in committed that have had no buyer activity in the stated number of days (10 min).
4. Coverage versus required coverage for the current and next period (5 min).
5. Decisions needed from marketing, finance, or the executive team (10 min).

Definitions sheet

Revenue bridge

Gap decision log

Metrics and KPIs

Setting targets without invented benchmarks. Use three anchors. First, your own trailing best: the lowest weighted absolute forecast error you have achieved over four consecutive periods. Second, contractual floors: the revenue level your covenant or financing documents require, which sets a minimum for the plan, not a target. Third, named public peers: if you want an outside reference for retention or growth, use the filings of listed companies you can name, and read their definitions before you compare. Do not apply a number from a blog post to a different definition of retention or pipeline.

Set a bias target before an error target. Getting the average right is faster than getting each period right, and it removes the systematic part of the miss.

Common mistakes

1. Setting the target first. A top-down number spread across territories turns forecasting into negotiation. Build bottom-up, then decide on the gap in the open.
2. Measuring only the average. Total bias can be near zero while every period is wrong. Report bias and weighted absolute error together.
3. Using CRM stage probabilities as truth. Default probabilities rarely match your conversion history. Measure them from cohorts.
4. Counting stale pipeline. Coverage of 5x built from deals older than the sales cycle is not coverage.
5. Hiring late and expecting in-year revenue. A rep hired in the last half of the year contributes little or nothing to that year. Model ramp.
6. Treating retention as a customer success metric only. It is the base of the revenue plan and it belongs in the board pack.
7. Counting price increases at list. Announced, quoted, and collected are three different numbers.
8. Letting each function define its own numbers. Three versions of closed-won guarantee three versions of the forecast.
9. Changing definitions mid-year without restating history. Trend comparisons break. Version the definitions and restate prior periods once.
10. Skipping the cash link. A revenue plan that does not tie to billing and collections will miss on cash even when it hits on revenue.

AI prompt assistance

AI is useful for cleaning data, tagging causes, and drafting summaries. It does not know your deals. Supply the data, keep it inside your own approved environment, and verify results against the system of record.

1. Compute forecast bias and error.

I will paste a table with period, segment, owner, forecast, and actual. Compute bias and absolute error for each row, then totals and weighted absolute error by segment and by owner. Show the formula and one worked row. List the three owner and segment combinations responsible for most of the absolute error.

Check: recompute two rows and the grand total by hand, and confirm that actuals tie to the ERP.

2. Tag missed-forecast causes.

Here are 30 deals that slipped or were lost versus forecast, with the rep's notes. Tag each with one cause from this list: stage inflation, close date slip, deal size change, lost to no decision, lost to competitor, customer purchase order timing, definition difference. Return a count by cause and flag any note you could not classify with confidence.

Check: have the sales manager review ten tags. Do not change a stage rule on the basis of the tags alone.

3. Draft the definitions sheet.

Draft a one-page definitions sheet for qualified pipeline, stage exit criteria, booking, churn, and expansion for a B2B distributor. For each term, propose a definition, the evidence a manager can verify in the CRM, and the question a finance reviewer would ask. Mark every point where the definition might conflict with a customer contract or commission plan.

Check: compare with your contracts and compensation plan. Involve counsel or your auditors where a definition touches contractual rights or revenue recognition.

4. Summarize the weekly call.

Using this week's forecast log, write a half-page note for the CEO: change in committed and upside versus last week, five deals most likely to slip with the buyer evidence for each, coverage versus required coverage, and decisions needed.

Check: confirm every dollar figure against the CRM export before sending.

Implementation guide

Governance: the CFO owns the forecast measurement, the CRO owns capacity and pipeline, the head of customer success owns retention, and the CEO owns the gap decisions. Report bias, error, coverage gap, and net revenue retention to the board monthly, next to cash.

What to expect: definitions and a weekly call reduce noise in the first quarter. Bias usually takes two to three quarters to read, because you need several periods to separate bias from chance. Do not announce an accuracy target until you have the baseline. Capacity changes show up in bookings only after the ramp period you measured, and retention changes show up with the contract cycle.

A 30, 60, 90 view for a smaller team: in 30 days, definitions and the forecast baseline; in 60 days, cleaned pipeline and a conversion table; in 90 days, the bottom-up bridge and a first gap decision log.

Downloadable assets

The forecast accuracy log, weekly call agenda, definitions sheet, revenue bridge, and gap decision log above are designed to be copied into a spreadsheet.

Related resources

• Sales capacity planning: how many productive reps the bookings plan needs.
• Pipeline coverage and forecast accuracy: the coverage and bias calculations in detail.
• Net revenue retention: calculation and levers: the retention numbers that set the base.
• Customer profitability analysis: which revenue is worth growing.
• B2B customer segmentation for pricing and sales: how to segment forecasts and retention.
• Key account management cadence: the routine behind large-account retention.
• CPQ and quote discipline: keeping quoted price close to collected price.
• Price realization rate drift: measuring the gap between quoted and collected price.
• Rolling forecast vs annual budget: how a rolling forecast supports the revenue plan.

Hub: Revenue growth guides.

Frequently asked questions

What is revenue operations?

Revenue operations is the practice of running sales, marketing, pricing, and customer success on shared definitions, shared data, and one weekly forecast routine. Its job is to make revenue predictable and to show where it leaks. In a mid-market company it is usually a small team in finance or sales operations, not a new department.

Why fix forecast accuracy before setting growth targets?

Because every growth lever depends on a prediction, and a plan built on biased predictions inherits the bias. If you over-forecast by 8 percent, a 15 percent growth plan is really a 7 percent plan wearing a bigger number. Measure bias first, correct it, then set the target from what capacity, pipeline, and retention can support.

How do you measure forecast accuracy?

Measure forecast bias, (forecast minus actual) divided by actual, and weighted absolute error, the sum of absolute misses divided by the sum of actuals. Bias shows direction, absolute error shows size. Compute both at the start and midpoint of each period, by segment and by owner, for at least four periods before drawing conclusions.

Who should own revenue operations?

One named person should own the definitions and the forecast measurement, usually the CFO or a head of revenue operations reporting to the CEO. Sales owns capacity and pipeline, customer success owns retention, and pricing owns realization. Ownership of measurement should sit outside the team being measured.

How long does it take to improve forecast accuracy?

Expect a visible reduction in noise within one quarter after definitions and a weekly call are in place, and two to three quarters to judge bias, since you need several periods to separate a pattern from chance. Capacity and retention changes take longer because they follow hiring ramps and contract cycles.

How this works in a System of Performance

A revenue plan is only as good as the link between what was predicted, what was decided, and what happened next. In most companies those records sit in the CRM, the ERP, spreadsheets, and email threads, so each review starts from memory. A Performis routine can take on a recurring job like this, with the scope and fit agreed with you: it works from the records your team supplies, prepares the work for your named reviewer, and keeps each approval on record. If you want to see how far your own forecast sits from your actuals, bring it to us as a launch partner.