Forecasting for SMBs: A Practical Cadence That Works at $1-20M Without Enterprise Tooling

A weekly forecasting cadence built for SMBs at $1M to $20M in revenue. Stage exit criteria. The rep submission template. The manager rollup template. The four forecast call types. The single metric that tells you whether your forecast is honest. And when to graduate from spreadsheets to Clari or BoostUp.

1. Why SMB forecasts drift (and what good looks like)

The typical SMB forecast is wrong by 25 to 40% per quarter. Bridge Group, CSO Insights, and Clari's own benchmarking all land in roughly the same range. That much variance creates real business cost: cash plans miss, hiring plans overshoot or undershoot, board credibility erodes, and over time the founder stops trusting the team's numbers and reverts to running the forecast personally.

Good SMB forecasts land within 10% of actuals consistently. Not because the team is psychic, but because the team has installed three disciplines that most SMBs skip:

1. Stage exit criteria that mean something. Most CRMs let the rep advance a stage when they feel like it. Honest forecasts require evidence-based stage exit.
2. Deal categorization at the call level, not just the CRM stage level. The CRM stage is a backwards-looking artifact. The category (commit, best case, pipeline) is a forward-looking judgment.
3. A weekly cadence the manager actually runs. Not a monthly forecast review. Weekly. Most forecast drift happens between weeks 1 and 4 of a month, and only a weekly cadence catches it.

This guide gives you the templates, the cadence, and the call types to install all three.

2. Stage exit criteria, the foundation of an honest forecast

Every CRM has stages. Most teams have not defined what it takes to leave each stage. Without exit criteria, reps drift deals forward based on optimism rather than evidence, and the forecast inherits the optimism.

A workable 5-stage process for SMB B2B selling, with exit criteria you can actually verify:

The exit criteria are stricter than most SMB teams are used to. That is the point. Strict criteria produce honest forecasts. Loose criteria produce hopeful forecasts.

3. Categorize every deal: commit, best case, pipeline, omitted

The CRM stage tells you where the deal is. The category tells you whether the rep thinks it will close this period. These are different judgments and both belong in the forecast.

• Commit. The rep is willing to put their name on this deal closing this period. If asked one-on-one, they say "yes." Commit deals roll up to the manager forecast and are reported to the founder or VP.
• Best case. Could close this period if things break right. The rep would not be shocked but would not bet on it. Best case is what defines the upside number, not the committed number.
• Pipeline. Real opportunity but unlikely to close this period. Important for next-period forecast and for current-period coverage analysis.
• Omitted. The deal is open in the CRM but the rep is honest that it is dead or stalled. Surfaces stale pipeline before it pollutes the forecast.

The honesty test: every deal in the manager forecast should be one of these four categories. Anything in "pipeline" that has not progressed in 30 days should be flagged for omitted unless there is a specific reason to keep it open.

4. The rep submission and manager rollup templates

Every Friday by 5pm, every rep submits their forecast to the manager in this format:

The manager runs all rep submissions through one consolidated rollup:

Both templates fit on a single page. Forecasts that need 12 tabs in a spreadsheet are forecasts that nobody reads. Brevity is what makes them survive the weekly cadence.

5. The weekly and monthly cadence calendar

6. The four forecast call types

Not every forecast conversation has the same purpose. Four distinct call types, each with a different agenda:

Type 1: weekly pipeline review (manager + rep). 30 minutes. Deal-by-deal. The manager is in coach mode, not interrogation mode. Goal is to surface what is real and what is not, and to identify the specific coaching the rep needs on each deal.

Type 2: weekly forecast rollup (manager solo). 30 minutes. The manager's own analysis after consolidating rep submissions. Goal is to land on a committed number the manager will personally defend to the founder.

Type 3: monthly committed forecast call (manager + founder/VP). 60 minutes. The manager presents the committed number. The founder asks the hard questions: what is the biggest risk to the commit, what is the biggest upside, what would change if we moved resources around. Goal is alignment on the number and the levers.

Type 4: mid-month pulse (manager + founder/VP). 30 minutes. Honest check on whether the commit still holds. Most slippage shows up here. The point is not to recommit; it is to know early so cash and hiring plans can adjust.

7. The single metric that tells you the forecast is honest

Track commit-to-close ratio rolling over the last 6 months. Calculate it as dollars actually closed in a month divided by dollars committed at the start of the month, expressed as a percentage.

• 105% to 115%: Honest. Manager is slightly sandbagging, which is fine.
• 95% to 105%: Excellent. Forecast is calibrated.
• 85% to 95%: Acceptable for SMB. There is room to tighten exit criteria.
• Below 85%: Forecast is hopeful, not honest. The fix is almost always stricter Stage 3 exit criteria (the champion + economic buyer + use case + proposal requirements above).
• Above 120%: Manager is over-sandbagging. The team will be denied resources because the forecast looks worse than reality. Coach the manager toward submitting a higher commit number.

One metric, six-month rolling, reviewed quarterly. This is the gauge.

8. When to graduate from spreadsheet to Clari or BoostUp

The spreadsheet-based cadence above works cleanly for SMBs up to roughly $10M in revenue with 6 to 10 reps. Past that, three things start to break:

• Manual aggregation eats the manager's Friday. 6 to 10 reps takes 90 minutes. 15 reps takes 3 hours. 25 reps takes a full day.
• Trend analysis becomes important. Knowing this week's commit is not enough. The manager needs to see how the commit number moves week over week, which deals slip from commit to best case, and which reps are calibrated vs hopeful.
• The founder or VP wants drill-down. Without a tool, every drill-down request triggers a manual spreadsheet exercise. With a tool, the founder pulls the answer themselves.

The right time to evaluate Clari, BoostUp, or InsightSquared is when you cross 15 reps or 12 months of consistent weekly cadence in spreadsheets, whichever comes first. The right time to skip them is when you are under 6 reps; the tools are powerful but they are also slow to configure and the manual cadence will outpace them at small scale.

If your AI dimension on the maturity scorecard is at "Multi-tool stack" or "Agentic workflows," a forecasting AI is the third tool you should add, after conversational intelligence (call recording and coaching) and AI role-play (rep practice).