Feature · Sales Forecasting Methods

Sales forecasting methods that produce a number finance will actually trust.

Commit, best case, pipeline coverage, weighted stage, time series, regression, and AI-assisted. Every method runs on the same live pipeline inside the shared record store, side by side, with the reconciliation report that shows exactly where the methods disagree and why. The sales leader walks into the finance meeting with the real number and the real receipts.

Start free trial See pricing
The seven forecasting methods every revenue org eventually tries

When each one works, when each one breaks, how to use them together.

Sales forecasting is not one method. It is a family of methods, each with specific strengths and specific failure modes. The sales leader who uses only one method eventually gets burned. The sales leader who uses all seven and reconciles the differences produces a number finance actually trusts. These are the seven methods every mature revenue org ends up running, with the specific conditions under which each one earns the seat.

Rep commit

The number the rep stands behind.

The rep looks at the pipeline, picks the deals they will close this quarter, and commits a number. The method captures seller conviction and local knowledge. It fails when reps sandbag (too low) or over-promise (too high) under pressure. Useful as a baseline that other methods calibrate against.

Best case

Everything that could close, if everything breaks right.

The rep lists every deal that could realistically close in the window, including the stretch deals. The method surfaces the full upside. The spread between commit and best case is the first signal of forecast conservatism or ambition. Finance wants both numbers.

Pipeline coverage

Total pipeline divided by target.

Total open pipeline in the quarter divided by the quarterly target. A healthy ratio for most B2B businesses sits between 3x and 5x depending on close rate. A ratio under 3x is a signal of a thin funnel. The method is coarse but useful for program-level health.

Weighted stage

Each stage multiplied by its historical win rate.

A deal in stage 3 at a 40 percent historical close rate contributes 40 percent of its amount to the forecast. A deal in stage 5 at an 80 percent close rate contributes 80 percent. The method captures stage-level probability and is the single most common forecasting method in B2B SaaS.

Time series

Last four quarters, extrapolated forward.

A time-series method extrapolates recent quarterly performance forward with trend adjustment and seasonality. The method works well for mature businesses with stable patterns. It fails on new segments, new motion, or disruptive market events. Useful as a sanity check against pipeline-driven methods.

Regression

Multi-factor model fit to closed-won history.

A regression model takes stage, age, deal size, segment, source, rep tenure, and other features and predicts probability of close. The method is more accurate than weighted stage when the model is well-fit and well-maintained. It is a specialist method and benefits from an ops team with analytical capacity.

Scenario

Best, likely, worst, with the plan behind each.

A scenario forecast ships three numbers (best, likely, worst) with the specific deals that distinguish each scenario. Finance sees not just the number but the plan behind each. The scenario method is especially useful for forecasts shared outside the sales team because it carries the uncertainty transparently.

AI-assisted

The assistant projection as the eighth method.

Strkr AI reads the pipeline nightly and produces an AI-projected number using deal shape, stage age, activity signal, past accuracy, stakeholder count, and segment baseline. The AI number sits alongside the human methods. The spread between the AI number and the rep commit is a diagnostic signal, not a replacement for the human judgment.

The combined view

All methods, side by side, reconciled.

The forecast surface in Strkr shows all seven methods side by side with the deltas highlighted. A quarter where every method converges within 10 percent is a quarter with a credible forecast. A quarter where methods diverge by 30 percent is a quarter with a specific calibration issue worth investigating before the finance review.

Rep commit and best case in Strkr

The seller-driven methods with the right scaffolding.

Rep commit and best case are the methods that rely on seller judgment. They work when the seller has real conviction, enough visibility to make a call, and a surface that captures the call accurately. They fail when the capture is a spreadsheet that nobody fills out until the Friday before the quarter close. Strkr builds the capture surface on the deal record with the specific controls that keep the method honest.

Commit field

A yes-no flag on every open deal.

Every open deal in the quarter carries a commit flag. The rep marks yes on the deals they commit to closing. The number aggregates to the rep forecast, the team forecast, and the org forecast live. No spreadsheet, no copy-paste, no reconciliation.

Best-case field

A separate flag for the stretch deals.

Beyond commit, the rep marks best-case flags on the stretch deals. The best-case number aggregates separately so the sales leader sees commit, best case, and the delta. The delta becomes the coaching conversation: which stretch deals need the push to move into commit.

Submit lock

Weekly forecast locked at Friday noon.

The weekly forecast submits with a lock. After the lock time, changes write to a new version but the submitted number stays as the forecast of record. Finance has a defensible number that does not shift under its feet. The next week starts from a fresh submit.

Change log

Who moved what, when, why.

Every change to a commit flag writes to the deal timeline with the actor, the timestamp, and the reason (if the rep adds a note). The sales manager reviewing a quarter sees exactly which deals moved in and out of commit and the reasoning behind each move. Pattern-level accountability.

Pull-in flags

Deals from next quarter that could pull.

Separate from commit, the rep can flag deals from the next quarter as pull-in candidates. A deal with a Q3 close date flagged as potentially pullable to Q2 shows up on the Q2 forecast as upside. The pull-in flag is the honest way to track the moonshot capacity without inflating commit.

Push-out flags

Deals currently in the quarter that may slip.

The symmetric flag captures deals currently in the quarter that may push out. A deal with a current close date in Q2 flagged as likely slipping to Q3 shows up as risk on the Q2 forecast. Finance sees the specific slip candidates rather than discovering them one deal at a time at month-end.

Manager override

The leader can disagree with the rep.

The sales manager reviewing the rep forecast can set a manager-adjusted commit on specific deals. The override logs with the actor and the reason. The rep sees the adjustment and discusses in the 1:1. The forecast of record is the manager-adjusted number for the leader review path.

Accuracy tracking

How accurate was last quarter commit vs actual.

Each quarter, the accuracy report compares the submitted commit against the actual closed-won. Reps who consistently sandbag or over-commit stand out. The pattern feeds coaching. Over time, the forecast calibration improves because the measurement is systematic rather than vibe-based.

Call-coach integration

The 1:1 opens on the forecast changes.

The weekly rep manager 1:1 opens on the forecast changes from the past week: new commits, removed commits, slipped close dates, moved stages. The 1:1 structure is forecast-first rather than activity-first because the forecast is the leading indicator of the next quarter outcome.

Weighted stage and pipeline coverage

The quantitative methods, done honestly.

Weighted stage and pipeline coverage are the quantitative baseline methods. Done well, they produce a credible forecast that reads the pipeline as it actually sits. Done poorly, they produce a vanity number that nobody trusts. Strkr runs both methods on the live pipeline inside the record store with the specific controls that keep the output honest.

Historical close rates

Not vendor defaults, your real numbers.

The weighted-stage method pulls close rates from the actual closed deals in the tenant history, not from vendor defaults or industry averages. A stage 3 deal at your actual 42 percent close rate contributes 42 percent, not the 40 percent default. The number is grounded in your real motion.

Segment-aware rates

Enterprise and SMB have different curves.

Close rates vary by segment. Enterprise deals convert differently than SMB deals. The weighted-stage calculator respects segment, so a stage 3 enterprise deal uses the enterprise close rate while a stage 3 SMB deal uses the SMB rate. The segmentation materially improves forecast accuracy.

Age-adjusted rates

Deals aging in stage lose probability.

A deal that has sat in stage 4 for 90 days has a different probability than one that landed in stage 4 last week. The age-adjustment multiplies the base close rate by a decay factor. Stale deals surface in the forecast at their reduced probability rather than their stated probability.

Pipeline coverage ratio

Total pipeline divided by target, by segment.

The pipeline coverage ratio shows total open pipeline divided by the quarterly target, broken out by segment, by rep, and by product line. The coverage view surfaces thin funnels at the segment or rep level before the quarter runs out of time to build pipeline.

Coverage trend

How coverage changes week over week.

The coverage trend chart shows pipeline coverage ratio over the past 12 weeks. Rising coverage is a healthy signal. Flat or dropping coverage is a problem that compounds. The trend view surfaces the direction before the quarterly report shows the outcome.

Stage conversion funnel

Rates between each stage, by rep.

The stage conversion funnel shows the conversion rate between each stage, aggregated by rep and by segment. Reps with low conversion at a specific stage flag for coaching. Segments with low conversion at a specific stage flag for playbook refinement. The funnel is the diagnostic view for the forecast.

Discount drag

Weighted amount factors in realized discount.

Deals carry a stated amount and an expected discount. The weighted-stage calculation factors in the expected discount so the forecast is net of discount rather than list price. Finance sees the real number rather than the nominal number that reality shaves down by month-end.

Multi-year weighting

Multi-year deals weight by year-one ARR.

For multi-year contracts, the forecast weights by year-one ARR rather than TCV. The number aligns with the way most SaaS finance teams actually recognize revenue. The TCV view ships as a separate view for teams that forecast on TCV instead.

Carry-forward

Deals pushed out explicitly, not silently.

Deals that slip past the current quarter get a push-out event on the record. The forecast on the current quarter removes the deal. The forecast on the next quarter adds the deal with a carry-forward flag. The flow is explicit so finance sees how much of the next quarter forecast is new pipeline versus carry.

Time series, regression, and AI-assisted methods

The quantitative methods that need a data posture to run.

Time series, regression, and AI-assisted forecasting are the methods that go beyond the live pipeline. They look at history, at correlations, and at pattern matching. Done well, they produce accuracy improvements over weighted stage. Done poorly, they produce a number that nobody can defend. These are the specific implementations Strkr ships with the controls that keep the output honest.

Time series baseline

Last four quarters, trend, seasonality.

The time-series method ingests the last eight quarters of closed-won data, extracts trend and seasonality, and projects the next two quarters. The method works best for mature businesses with stable patterns. The output ships as a line with a confidence interval, not a point estimate, so finance sees the real uncertainty.

Regression model

Multi-factor fit to historical closed-won.

The regression model takes stage, age, deal size, segment, source, rep tenure, stakeholder count, discount depth, and other features and predicts probability of close. The model recalibrates monthly against new closed-won data. The output per deal is a probability. The sum across the pipeline is the regression forecast.

AI projection

Strkr AI reads the pipeline nightly.

The AI-assisted method runs the assistant on the full pipeline nightly. For each open deal, the assistant computes a projected close probability using the deal shape, the activity signal, the stakeholder movement, the past accuracy of the handling rep, and the segment baseline. The assistant produces a per-deal number and a reasoning sentence.

Explainability

Every AI number ships with its receipts.

For every AI-projected deal, the assistant surfaces the top three driving signals in plain English: "Deal has not advanced in 29 days. Primary champion left the account 14 days ago. Discount depth 32 percent is above segment median." The receipts let the sales leader argue with the reasoning rather than the number.

Accuracy comparison

Which method predicted which quarter best.

Each quarter, the accuracy comparison shows which forecasting method produced the closest number to the actual closed-won. Over time, the pattern reveals which method works best for the business. The leader picks the primary method and uses the others as sanity checks.

Method weighting

Combine methods with configured weights.

For teams that want a composite forecast, the method weighting feature combines the methods with configured weights. 40 percent weighted stage, 30 percent AI projection, 20 percent rep commit, 10 percent time series. The composite ships as the forecast of record with the method breakdown as the audit trail.

Model drift detection

Alert when the model stops being accurate.

The forecast engine monitors model drift. When the regression or AI projection accuracy drops below the historical baseline, the alert fires. The ops team investigates whether the business motion changed, whether a segment dynamic shifted, or whether a model recalibration is needed. Drift surfaces fast rather than at year-end review.

Scenario planning

Model what-if changes.

The scenario tool lets the sales leader model what-if changes: "What if we close the top 5 enterprise deals in Q2 instead of Q3." "What if close rates on stage 4 drop by 10 percent." The scenarios ship alongside the base forecast so the leader walks into the finance review with the stress tests ready.

Confidence interval

A range, not a false-precision point.

Every forecasting method ships with a confidence interval rather than a single point. The time-series method surfaces its statistical uncertainty. The regression ships model variance. The AI projection carries aggregate uncertainty from per-deal confidence. Finance sees a range and the leader stops pretending to deliver a specific dollar amount with false precision.

Running the forecast meeting

The weekly cadence that makes the methods compound.

Forecasting methods are only useful if they feed a real meeting cadence that produces decisions. Strkr builds the forecast surface for the specific meeting cadence most B2B sales orgs run: rep manager 1:1 on Monday, team roll-up on Tuesday, exec forecast call on Wednesday, finance review at month end. These are the specific surface designs that make each meeting productive rather than dreaded.

Monday rep 1:1

The forecast changes from the past week.

The Monday rep manager 1:1 opens on the forecast surface filtered to the rep. The surface highlights deals that moved commit, deals that slipped close dates, deals that lost stakeholder engagement, and deals that gained stakeholder engagement. The 1:1 is a working session on the forecast rather than a status update.

Tuesday team roll-up

Manager views the team forecast with reconciliation.

The team roll-up view shows the aggregated rep commits, the manager overrides, the weighted-stage number, and the AI projection. The manager sees the reconciliation and prepares for the Wednesday exec call with a defensible number and the specific deals driving any gap between methods.

Wednesday exec call

The org forecast, with the receipts.

The exec forecast call runs off the org forecast view with every team roll-up, every segment, every product line. The exec sees the number and the five biggest risk deals. The call is 30 minutes rather than 90 because the data preparation happened in the surface rather than in a spreadsheet reconciliation.

Month-end finance review

The number of record with the full audit.

The month-end finance review opens on the submitted forecast with the full audit log: every change, every actor, every reason. Finance sees the complete history and the sales leader defends the number with the real receipts. The review becomes a review rather than a reconstruction.

Risk watchlist

The specific deals that could blow up the quarter.

The risk watchlist is a sorted view of the five to ten deals whose outcome most swings the quarter result. Each deal carries current commit status, the AI risk score, the stakeholder situation, and the next-step owner. The sales leader works the watchlist weekly rather than the full pipeline.

Upside watchlist

The pull-in candidates with the best path.

Symmetric to the risk watchlist, the upside watchlist shows the pull-in candidates with the clearest path to pulling into the current quarter. Each deal carries the stakeholder work needed, the discount leverage, and the timeline risk. The upside becomes a managed program rather than a hope.

Weekly submit

The ritual that keeps the methodology honest.

The weekly submit is the ritual at the center of the forecast cadence. Reps submit their commit by Friday noon. The submission locks. The following week starts from the previous week submit. The ritual produces the time series of forecast accuracy that fuels the next quarter calibration.

Quarterly accuracy review

What did the methods predict, what actually happened.

Each quarter, the accuracy review compares every submitted forecast against the actual closed-won. The pattern surfaces whether reps sandbag or over-commit, whether the weighted-stage method ran accurate or drifted, whether the AI projection beat the human methods. The review feeds the next quarter calibration conversation.

Yearly methodology review

Which methods graduate, which retire.

Once a year, the sales ops team reviews which forecasting methods actually earned their keep in the past four quarters. Methods with chronic inaccuracy retire. New methods graduate to the primary rotation. The methodology evolves with the business rather than ossifying into a vendor default.

A forecast with every method, every receipt, every change tracked.

Rep commit, best case, pipeline coverage, weighted stage, time series, regression, AI projection, and the composite view all ship on every paid tier. Strkr AI runs with no credit meter. The weekly submit, the manager override, the risk watchlist, and the audit log all live on the shared record. Open your real pipeline and run the forecast your finance team actually wants.

Start free See pricing
Common questions

What buyers ask about this feature.

What are the main sales forecasting methods and when does each one work?

The seven methods every mature revenue org eventually uses are rep commit (the number the rep stands behind, captures seller conviction and local knowledge), best case (everything that could close including stretch deals), pipeline coverage (total pipeline divided by target, healthy range 3x to 5x), weighted stage (each stage multiplied by its historical win rate, the most common B2B method), time series (last four to eight quarters extrapolated forward with trend and seasonality), regression (multi-factor model fit to closed-won history), and scenario (best, likely, worst with the plan behind each). Each has specific failure modes. Rep commit fails under sandbagging or over-promising pressure. Weighted stage fails when close rates vary by segment or deal age. Time series fails on new segments or disruptive events. The mature forecast runs multiple methods side by side with reconciliation.

How does Strkr AI fit into the forecasting methodology?

Strkr AI runs as an eighth method alongside the seven classical methods. The assistant reads the full pipeline nightly and produces an AI-projected number for each open deal using deal shape, stage age, activity signal, stakeholder count and movement, past accuracy of the handling rep, discount depth, and segment baseline. Every AI number ships with the top three driving signals in plain English so the sales leader argues with the reasoning rather than the black-box number. The AI number sits alongside the human methods in the forecast surface. The spread between the AI projection and the rep commit is a diagnostic signal: a rep with a commit well below the AI projection may be sandbagging, a rep well above may be over-promising. Strkr AI runs on every paid tier with no credit meter.

What makes rep commit and best case trustworthy versus noise?

Rep commit becomes trustworthy when the capture surface is on the deal record with the specific controls: a yes-no commit flag on every open deal that aggregates live, a separate best-case flag for stretch deals, a weekly submit lock at Friday noon so finance has a defensible number of record, a change log that captures every move with actor and reason, pull-in and push-out flags for cross-quarter candidates, a manager override that logs when the leader disagrees with the rep, quarterly accuracy tracking that measures each rep against actual outcomes, and a 1:1 cadence that opens on the forecast changes from the past week. Without these controls, rep commit becomes a Friday-night spreadsheet exercise. With them, rep commit becomes the leading indicator of quarter outcome.

How does weighted stage forecasting work in Strkr?

Weighted stage multiplies each open deal amount by the historical close rate for its current stage, then sums across the pipeline. Strkr runs weighted stage with four specific improvements over the vendor default: historical close rates come from the actual tenant data rather than industry averages, close rates break out by segment (enterprise versus SMB have different curves), age-adjustment multiplies the base rate by a decay factor for stale deals, and the weighted amount factors in expected discount so the number is net rather than list price. The forecast surface also ships pipeline coverage ratio (total pipeline divided by target), stage conversion funnel (rates between each stage by rep and segment), and multi-year weighting (year-one ARR for SaaS, TCV as a separate view for teams that forecast on contract value).

What forecast cadence does the methodology support?

The weekly forecast cadence most B2B sales orgs run has four meeting shapes that Strkr builds specific surfaces for. The Monday rep manager 1:1 opens on the forecast surface filtered to the rep with the deal changes from the past week highlighted. The Tuesday team roll-up shows aggregated commits, manager overrides, weighted-stage, and AI projection side by side. The Wednesday exec call runs off the org forecast with the five biggest risk deals surfaced. The month-end finance review opens on the submitted forecast with the full audit log. The weekly submit at Friday noon locks the number so finance has a defensible value of record. Risk and upside watchlists sit between the ritual meetings for ongoing management. The quarterly accuracy review measures each method against actual closed-won.

Does the forecasting methodology require a specific tier?

No. The forecast surface with all seven classical methods plus the AI projection ships on every paid tier at no extra charge. Strkr AI runs with no credit meter and powers the AI projection, the deal risk scoring, the manager-ready pipeline briefings, and the reasoning sentences on every projected number. The weekly submit, the lock, the change log, the manager override, the risk watchlist, the upside watchlist, the pull-in and push-out flags, the segment-aware close rates, and the age-adjusted probabilities are all native. HIPAA-compliant deployments with a signed Business Associate Agreement are available for qualified customers, and SOC 2 Type II is in progress with an external audit firm. The seat price on the pricing page is the full price for the forecasting capability.

Related features

Pair this with...

Try it free. Bring your team next week.

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.

Start free trial See pricing