A revenue forecast is wider than a sales forecast. Sales forecasting predicts new bookings only. Revenue forecasting predicts everything that will hit the P&L next quarter, from fresh deals to churn to professional services.
Revenue forecasting is the practice of predicting total company revenue over a future period, usually a quarter or a year, across every revenue stream the business runs. It blends new-business bookings, expansion from existing customers, renewal revenue, and services into one top-line number. Finance and RevOps own the model, Sales owns the new-business input, and CS owns the retention input. The output drives board commits, hiring plans, and cash flow.
The six things every finance leader, RevOps lead, and CEO should know before building, running, or buying a revenue forecasting process.
A revenue forecast rolls up every revenue stream the company runs: new business, expansion, renewals, and services. It is the number the CFO commits to the board and the number the CEO reports to investors. Sales forecasting is one input. Finance owns the model, RevOps runs the data pipeline, and Sales and CS feed in their forward numbers.
A healthy B2B SaaS revenue forecast is split into four streams. New business is logos signed this quarter. Expansion is existing customers buying more. Renewals are contracts coming up for renewal and the predicted retention rate. Services are implementation and professional-services bookings. Each stream has a different owner, a different confidence profile, and a different cadence.
Top-down starts from a market model or board target and divides it across streams. Bottom-up builds from pipeline, renewal lists, and expansion plays. The gap between the two is the single most useful signal in the whole exercise. A $2M gap means someone is wrong, and the forecast meeting is a working session on which number to trust and why.
Revenue forecasts refresh monthly inside the quarter. The CFO commits a quarterly number to the board at the start of each quarter and updates mid-quarter if variance exceeds a defined threshold. The annual plan is reset each year during planning season. Weekly motion stays with sales forecasting. Revenue forecasting breathes on a slower, finance-led rhythm.
Finance builds and maintains the forecast model. RevOps runs the data pipeline that feeds it. The CRO or Chief Revenue Officer owns the committed number back to the CEO and board. Sales leaders own the new-business input. CS leaders own the retention and expansion input. Everyone has a lane. The forecast meeting is where lanes converge.
Public-market SaaS companies target quarterly revenue variance under 3% versus guidance and full-year inside 5%. Private companies without the earnings-call pressure target a wider band but still track variance as a per-quarter metric. Rolling thirteen-month accuracy charts tell the board whether to trust the model and whether to trust the team running it.
A revenue forecast is only as good as the streams feeding it. Each stream has a different data source, a different confidence profile, and a different owner. Treating them as one number hides the real risk. The CFO who separates the streams on the model sees where the quarter is actually breaking.
New business is the hardest stream to call. It comes from the sales pipeline and depends on open deals closing in the forecast window. Sales submits the number via the weekly sales forecast cadence. Finance pulls that commit into the revenue model. Confidence is medium: pipeline math is well understood, but outside events move numbers week to week.
Expansion revenue comes from upsell, cross-sell, and seat growth on live accounts. CS or account managers track the expansion pipeline the same way Sales tracks new-business pipeline. Confidence is higher than new business because the customer is already live, the buying center is known, and the sales cycle is shorter.
Renewals are the most predictable stream for a healthy SaaS business. The list of contracts up in the quarter is known months in advance. The retention rate against that cohort is the forecast lever. CS owns the number. Finance applies the predicted retention rate (gross or net) to produce the renewal contribution to total revenue.
Services revenue is often lumpy and project-driven. It is predicted from signed statements of work, utilization against the services backlog, and expected close dates on open proposals. For many SaaS companies this is 5-15% of total revenue. For services-heavy businesses it can be a majority. Services leaders own the input.
Not every stream adds. Logo churn, seat contraction, and downsell all reduce the recurring base. The revenue forecast deducts expected churn from the renewal cohort and expected contraction from the live base. Finance tracks gross retention and net retention side by side so the CEO sees both the attrition rate and the net effect of expansion offsetting it.
The four streams minus churn and contraction roll up to total revenue for the period. That is the number the CFO commits to the board and the CEO reports to investors. Reporting the streams separately gives leadership the lever map: if the quarter is at risk, the forecast says which stream is pulling it down and which team owns the fix.
Every revenue forecast is some blend of the four methods below. Public-market companies lean on driver-based models. Early-stage startups lean on bottom-up pipeline math. The right mix depends on data maturity, business model, and how much the forecast needs to defend. Finance teams that pick one method and ignore the rest get surprised at quarter end.
Top-down starts with a total number from a market model, a board commit, or an annual plan, then divides it across streams, segments, and regions. Fast and good for setting the target. Weak at calling the actual quarter because it ignores pipeline reality. Most teams use top-down to set the plan and bottom-up to call the quarter against it.
Bottom-up builds the forecast from every open deal, every renewal in the quarter, every signed SOW, and every expansion play. Each stream owner submits a number. Finance sums the rollup. Slow to build the first time, but once the data pipeline runs, it is the most defensible forecast and the one that stands up to board questions.
Trend-based forecasting takes last period's revenue and applies a growth rate derived from recent history or a planned ramp. Lightweight and useful as a sanity check against the bottom-up number. Weakest when the business is changing fast (new product, new segment, pricing change). Strongest in steady-state businesses with predictable growth.
Driver-based forecasting ties revenue to the operating levers behind it: pipeline coverage ratio, win rate, ACV, sales capacity, retention rate, expansion rate. Finance builds the model so a change in a driver flows through to the top line. The gold standard for mature SaaS finance teams. Supports scenario planning and sensitivity analysis the other methods cannot.
Modern revenue models overlay machine-learning signals from CRM history, pipeline velocity, renewal risk scoring, and engagement data. The AI does not replace the finance model. It flags where the rolled-up number is likely wrong, which specific deals or renewals are at risk, and where the model drifts from historical patterns. Finance uses the signals in the forecast meeting.
A production revenue forecast runs top-down for the target, bottom-up for the quarter commit, and driver-based for scenario work. The gaps between the three are the signal. When they agree, confidence is high and the CFO commits with conviction. When they diverge, the forecast meeting has a specific list of streams and drivers to pressure test before the number locks.
Revenue forecasting runs on a slower cadence than sales forecasting. Finance refreshes the full model monthly, with a quarterly commit at the start of each quarter and a mid-quarter check-in. The shape of the month is predictable: stream owners submit, Finance aggregates, the CRO and CFO reconcile, and the committed number lands on the executive dashboard.
Sales submits the new-business forecast from the weekly sales cadence. CS submits renewal and expansion forward numbers. Services submits the services backlog and expected bookings. Each submission includes assumptions, confidence band, and change versus last month. Finance pulls each submission into the model on a fixed date.
Finance rolls up every stream, deducts expected churn and contraction, and compares the total against the top-down target. The variance between top-down and bottom-up is the opening agenda for the forecast meeting. Finance also runs the driver-based model and reports where drivers (coverage, win rate, retention) are tracking against plan.
The CRO, CFO, VP Sales, VP CS, VP Services, and RevOps lead meet to walk the forecast. The agenda is: variance to top-down, variance to last month, at-risk deals in new business, at-risk renewals, expansion softness, and services backlog risk. The meeting ends with an updated commit number everyone signs off on.
The CFO commits the number into the executive dashboard, the board update, and (for public companies) the guidance range. The CRO owns the number back to the CEO. Any mid-quarter revision requires a formal re-forecast and a documented reason. The commit is a signed artifact, not a running estimate.
If actuals or pipeline health move more than a defined threshold (typically 3-5%) from the committed number mid-quarter, Finance runs an unscheduled re-forecast and the CFO issues a revised commit to the board. The threshold is set in advance so the trigger is objective, not political. Most quarters never need the trigger. The ones that do are the ones that save credibility.
When the quarter closes, Finance reconciles actual revenue against every intra-quarter forecast version. The accuracy of each forecast gets scored. Model drift (where the driver-based number diverged from reality) gets analyzed. Stream owners see their accuracy. The scorecard is reviewed in the quarterly business review and feeds the next quarter's weights.
Strkr runs the CRM, the renewal pipeline, the expansion motion, and the services backlog on one platform, with Strkr AI risk flags on the deals and renewals most likely to miss. Finance pulls one clean rollup instead of chasing four spreadsheets.
Sales forecasting predicts new-business bookings only, usually weekly, owned by the Sales org. Revenue forecasting predicts total company revenue across new business, expansion, renewals, and services, usually monthly, owned by Finance with the CRO on the hook. Sales forecasting is one input into the revenue forecast. The revenue forecast is the number the CFO commits to the board.
Finance owns the model and builds the roll-up. RevOps runs the data pipeline that feeds it. The CRO owns the committed number back to the CEO and board. Each stream has an owner: Sales leaders own new business, CS leaders own renewals and expansion, Services leaders own services bookings. Everyone has a lane. Finance is the integrator, not the predictor.
Top-down starts with a total target (from the board, a market model, or the annual plan) and divides it across streams and segments. Bottom-up builds the forecast from every open deal, renewal, expansion play, and SOW. Mature finance teams run both and the gap between them is the single most useful signal in the exercise. The gap tells leadership where the model is wrong.
New business is logos signed in the period, owned by Sales. Expansion is existing customers buying more, owned by CS or AMs. Renewals are contracts up for renewal, owned by CS and the renewals team. Services is implementation and professional services bookings, owned by the Services team. Each stream rolls up to total revenue after subtracting expected churn and contraction.
Public-market SaaS companies target quarterly revenue variance under 3% versus guidance and full-year inside 5%. Private companies without the earnings-call pressure target a wider band but still track variance as a per-quarter metric. Rolling thirteen-month accuracy charts tell the board whether to trust the model and whether to trust the team running it. Below those bands, credibility erodes fast.
Monthly inside the quarter is standard. The CFO commits a quarterly number at the start of the quarter, refreshes monthly, and issues a mid-quarter revision if variance crosses a defined threshold (typically 3-5%). Weekly cadence belongs to sales forecasting. Revenue forecasting breathes on a slower, finance-led rhythm so the committed number has weight when it lands.
Driver-based forecasting ties revenue to the operating levers behind it: pipeline coverage ratio, win rate, average contract value, sales capacity, gross retention, net retention, expansion rate. Finance builds the model so a change in any driver flows through to the top-line forecast. It supports scenario planning and sensitivity analysis that pure top-down or bottom-up models cannot. The gold standard for mature SaaS finance teams.
Smaller teams run revenue forecasts in spreadsheets. Mid-market and enterprise teams use the CRM's forecast module alongside a financial planning and analysis platform. The must-have inputs are: CRM pipeline data, renewal and expansion data from CS, services backlog data, actuals from the general ledger, and a scenario-capable model. Spreadsheets break the moment the streams and segments multiply past a few dozen rows.
No sales call, no migration consultant, no four-month implementation. Enter your card, get 14 days of the full Pro tier, cancel any time before day 14 with zero charge. Spin up a workspace, import your CSV, and have something useful before lunch.
