Most founders start forecasting only after they get burned.
A few deals slip into next month. A hire starts before the pipeline is ready. Marketing keeps spending because the target still looks possible, while the actual close dates say otherwise. Then finance asks a simple question: what are we really going to sell this quarter?
If your answer is a mix of hope, rep opinion, and a spreadsheet last updated two Fridays ago, you don't have a forecast. You have a guess.
That's why "what is forecasting sales" matters more than it sounds. For a small B2B SaaS team, forecasting isn't a finance exercise. It's how you decide whether to hire, how hard to push outbound, how much runway you really have, and whether your current pipeline can support the plan.
The problem with guessing your revenue
The painful part isn't missing a number. It's making operating decisions off the wrong number.
A founder thinks two deals will close this month, so they approve a hire. One buyer goes quiet, another pushes legal review, and suddenly payroll was planned against revenue that never landed. An AE says the quarter still looks fine because the pipeline is “strong,” but half the opportunities don't have real next steps. Marketing keeps feeding top of funnel, but no one can tell whether that pipeline will convert in time.
This is what bad forecasting looks like in real life. Not a dramatic crash. Just a series of small decisions made with low visibility.
Target versus forecast
A lot of small teams confuse a sales target with a sales forecast.
A target is what you want to happen. It's the goal.
A forecast is what is likely to happen based on the deals, timing, conversion patterns, and market reality in front of you.
Those are not the same thing.
A healthy business can miss a target and still have a good forecast. A business with a bad forecast gets surprised late.
Why founders feel this first
In an early-stage company, the forecast touches everything:
- Hiring decisions: Can you add an SDR or AE now, or should you wait?
- Cash planning: Are expected collections likely to arrive when you need them?
- Campaign pacing: Should you keep spending to fill next quarter, or protect budget now?
- Founder time: Do you stay in sales calls, or shift into recruiting, partnerships, or fundraising?
Without a forecast, every one of those decisions leans on confidence instead of evidence.
The good news is you don't need an enterprise RevOps team to fix this. You need a simple method, clean definitions, and a weekly habit of reviewing what changed.
What sales forecasting really means for B2B teams
Sales forecasting is the practice of predicting how much a company will sell in a future period, usually by week, month, quarter, or year, using historical sales plus market and pipeline data, as explained by IBM's overview of sales forecasting.
For a B2B team, that definition matters because it moves forecasting out of the “sales gut feel” bucket. A forecast is a working model of future revenue. It should change as your pipeline changes. It should get sharper as your data improves.
It's not just a number at the end of the month
A useful forecast answers practical questions:
- What is likely to close
- When it is likely to close
- How much confidence you should have in that number
- What assumptions the number depends on
That last point matters. If your forecast depends on three late-stage deals with shaky next steps, the number may look fine while the risk is high. Good forecasting exposes that.
IBM also notes that forecasting informs budgeting, hiring, inventory, and resource allocation, not just sales planning. That's exactly how small SaaS teams should think about it. The forecast is a planning tool for the whole business, not a leaderboard metric for the sales team.
Why B2B forecasting is harder than it looks
B2B sales cycles usually stretch across multiple calls, stakeholders, and approval steps. Deals move. Budgets change. Buyers disappear for two weeks and then come back.
That's why a forecast has to be tied to a time horizon and updated often. Weekly, monthly, quarterly, and annual views all matter, but they answer different questions.
A practical way to think about it:
| Forecast horizon | Best used for |
|---|---|
| Weekly | Spotting slips, deal movement, and immediate execution risk |
| Monthly | Managing cash expectations and rep pacing |
| Quarterly | Hiring, spend decisions, and board-level planning |
| Annual | Capacity planning and overall growth model |
What a small team should include
Your first B2B forecast doesn't need advanced tooling. It needs a few basic inputs:
- Historical closed business
- Open pipeline
- Deal stage
- Expected close date
- Deal value
- A simple probability model
If you're still building your prospecting stack, it helps to understand the trade-offs between research platforms and data sources. A comparison like Orbbit versus ZoomInfo can help you think through how lead and account data quality affects the top of your funnel, which eventually affects forecast reliability too.
Practical rule: If your close dates and stage definitions are sloppy, the forecast will be sloppy, no matter how polished the spreadsheet looks.
Four common sales forecasting methods compared
There isn't one best method. The right approach depends on your data quality, how stable your market is, and how complex your sales motion is. Xactly's guide to sales forecasting models lists common approaches such as straight line, moving average, linear regression, time series, ARIMA, exponential smoothing, econometric models, and cohort analysis. The same guide notes that different methods fit different conditions, especially when markets are changing.
For small B2B SaaS teams, four methods come up most often.
Comparison of sales forecasting methods
| Method | Best For | Data Required | Complexity |
|---|---|---|---|
| Historical | Teams with stable sales patterns and enough past closed-won data | Historical sales by month or quarter | Low |
| Opportunity Stage | Teams with a defined pipeline and clear stage progression | Deal value, stage, close date, stage probabilities | Low to medium |
| Pipeline-Based | Teams that want a bottom-up view of current quarter revenue | Full open pipeline, rep input, timing assumptions | Medium |
| Predictive or AI | Teams with more data, more variability, or more complex motions | Historical outcomes, pipeline data, timing, multiple variables | High |
Historical forecasting
This is the simplest starting point. You look at past sales and project forward.
It works best when your market is fairly stable and your sales motion hasn't changed much. If last quarter looked similar to this quarter, historical trends can give you a useful baseline.
Where it fails is obvious. If you changed pricing, hired your first AE, moved upmarket, or launched outbound recently, the past may not be a reliable guide.
Opportunity stage forecasting
This method assigns a probability to each deal based on stage. A deal in proposal might count more than a deal in discovery.
Small teams like this method because it's easy to understand and easy to maintain. It gives structure without requiring a data science setup.
The trade-off is that stage labels can hide bad assumptions. If reps move deals forward too early, the forecast inflates fast.
Pipeline-based forecasting
This is more operational. You review the actual pipeline, deal by deal, and build a forecast from what is expected to close in a period.
It often combines stage, deal quality, close timing, and rep judgment. Done well, it gives founders the most useful view of the quarter because it reflects what's really in motion now.
Done badly, it becomes a meeting where everyone defends optimism.
Predictive or AI forecasting
This method uses broader inputs and more statistical modeling. In more complex or changing markets, models that include multiple variables can outperform simpler approaches, as noted in the Xactly source above.
For a small team, this usually isn't where you should start. It can be powerful, but only if the inputs are clean and the process is trusted. Otherwise, it produces complex noise.
Start with the method your team can actually run every week. A simple forecast that gets updated beats a smart-looking model nobody believes.
A step-by-step guide to creating your first forecast
If you're building your first forecast, keep it simple. The goal isn't perfection. The goal is to create a version you can trust enough to use, then improve it over time.
Step 1 choose one method
For most founder-led and small AE teams, start with opportunity stage forecasting.
It's simple. You take each open deal, assign a win probability based on stage, and multiply the deal value by that probability. Zendesk describes this as a weighted forecast in its sales forecasting techniques guide.
If you want a deeper primer on time-series thinking for future use, this guide to business predictions is useful context once you've outgrown the first-pass model.
Step 2 gather the minimum useful data
Pull these fields from your CRM or spreadsheet:
- Account name
- Deal value
- Current stage
- Expected close date
- Owner
- Next step
You don't need a huge dashboard. You need current, believable deal records.
If you're reviewing sales engagement and workflow tools while setting this up, a page like Orbbit versus Salesloft can help you think through where outreach systems fit compared with forecasting inputs and pipeline hygiene.
Step 3 standardize your stages
At this stage, many first forecasts fail.
Zendesk notes that forecast error often comes from inconsistent stages and weak process discipline. If one rep uses “proposal” to mean “sent pricing” and another uses it to mean “buyer confirmed budget and timeline,” your probabilities are meaningless.
Create stage definitions that match observable reality. For example:
- Discovery: First real qualification call completed
- Qualified: Problem, buyer, and basic fit confirmed
- Proposal: Pricing or proposal shared
- Negotiation: Commercial terms under active review
- Commit: Clear path and expected close timing confirmed
Step 4 calculate a weighted forecast
Use this formula:
Forecasted deal value = deal value × win probability
Example:
| Deal | Value | Stage | Probability | Weighted value |
|---|---|---|---|---|
| Deal A | High-value annual contract | Proposal | Medium-high | Value × probability |
| Deal B | Mid-market pilot | Discovery | Lower | Value × probability |
| Deal C | Expansion opportunity | Negotiation | Higher | Value × probability |
You can do this in a spreadsheet in less than an hour.
The exact percentages should come from your own historical win rates by stage once you have them. Until then, use conservative estimates and adjust them over time. Age-based forecasting can also help later, because close timing should reflect observed sales-cycle duration, not rep intuition alone, which Zendesk also highlights in the source above.
After your initial calculation, review the total in plain English:
- Which deals are carrying too much of the quarter?
- Which close dates feel weak?
- Which opportunities have no meaningful next step?
Step 5 review every week
A forecast is only useful if it moves with reality.
Watch this walkthrough if you want a visual explanation before building your sheet:
In a weekly review, ask:
- What moved forward
- What slipped
- What changed in value or scope
- Which deals should leave the forecast entirely
- What new pipeline entered the period
The first version of your forecast should feel slightly conservative and very easy to explain.
If you can't explain why a deal is in the number, it shouldn't be in the number.
How to improve your forecast accuracy
Most forecast problems aren't math problems. They're input problems.
Workday's summary of enterprise forecasting methods notes that the most reliable models start with historical sales as a baseline, then adjust for trend and seasonality using time-series methods such as moving averages, exponential smoothing, or ARIMA-style decomposition. It also stresses that forecast quality depends heavily on clean data, consistent stage definitions, and accurate close dates rather than subjective optimism, as outlined in Workday's forecasting methods guide.
Clean inputs beat clever formulas
Founders often ask how to make the forecast smarter. Usually the better question is how to make the underlying pipeline cleaner.
Focus on three things first:
- Stage discipline: Deals should enter a stage only when a real condition is met.
- Close date hygiene: Reps should change dates when reality changes, not after the period ends.
- Next-step quality: Every material deal should have a specific upcoming action, owner, and timing.
If those three are weak, the forecast won't improve much no matter what software you buy.
Add context, not just more numbers
Historical patterns help, but context sharpens judgment.
For example, a deal may look healthy in CRM while the buying team has gone quiet, priorities shifted, or the champion changed roles. On the other hand, an account might still be early-stage while showing strong signs of urgency because they're hiring into the problem, entering a new market, or changing systems.
That's why many teams combine internal pipeline review with external account context. This broader view is also reflected in practical resources like Jumpstart Partners on sales forecasting, which is useful if you want another plain-English take on how projected sales connect to operating decisions.
Build a simple accuracy loop
Use a lightweight review cycle:
| Review habit | What to check |
|---|---|
| Weekly | Stage changes, slips, new opportunities, next steps |
| Monthly | Closed-won versus forecasted deals, timing misses, rep bias |
| Quarterly | Stage probabilities, sales-cycle assumptions, pipeline quality |
If you're auditing prospecting systems and outbound data sources as part of that process, a comparison like Orbbit versus Apollo can help you evaluate where your lead flow and account data may be helping or hurting pipeline quality upstream.
Better forecasting starts before the deal is created. Weak-fit pipeline makes the whole model less reliable.
That matters for small teams. If the top of funnel is full of poor-fit accounts, forecasting gets harder because the pipeline looks fuller than it really is.
Common forecasting pitfalls and key metrics to track
A weak forecast usually breaks in familiar ways.
Sometimes reps leave close dates untouched because moving them feels like admitting the deal slipped. Sometimes founders keep dead opportunities in the quarter because the target still needs coverage. Sometimes everyone knows the CRM is messy, but nobody wants to stop and clean it because the quarter is busy.
Those are process problems, not edge cases.
Common mistakes
- Confusing quota with forecast: Quota is the goal. Forecast is the likely outcome.
- Trusting stage names too much: A deal in “proposal” can still be very weak.
- Letting stale deals stay active: If there's no buyer action, the deal may not belong in the call.
- Using rep optimism as evidence: Confidence isn't the same as proof.
- Ignoring timing risk: A deal can be real and still not close this period.
Metrics worth tracking
You don't need a giant KPI pack. A few measures go a long way.
| Metric | Why it matters |
|---|---|
| Forecast accuracy | Shows how close your projected revenue was to actual closed revenue |
| Pipeline coverage | Tells you whether open pipeline is enough to support the period goal |
| Sales cycle length | Helps you judge whether close dates are realistic |
| Stage conversion | Shows where deals actually move or stall |
| Slip rate | Highlights how often deals move out of the forecast period |
What to do with those metrics
Use them to challenge assumptions.
If forecast accuracy is poor, inspect stage definitions and close dates first. If pipeline coverage looks healthy but closed revenue still misses, your issue may be deal quality, stage inflation, or timing discipline. If sales cycle length keeps stretching, stop accepting end-of-month close dates at face value.
If the same deals keep showing up in the forecast and never closing, the problem isn't visibility. It's qualification.
From guesswork to a predictable growth engine
Sales forecasting doesn't need to start as a complex system. It needs to start as a repeatable habit.
The practical version is simple. Pick a method. Use real pipeline data. Define stages clearly. Review the number every week. Learn where your assumptions break. Then tighten the process.
That's how a founder moves from “I think we'll be okay” to “here's what is likely to happen, here's where the risk sits, and here's what we need to do next.”
If you've been asking what is forecasting sales, the shortest answer is this: it's the discipline of turning pipeline reality into a planning tool. Not a wish list. Not a target slide. A working view of future revenue that helps you run the business with fewer surprises.
Start with the current quarter.
Pull your open deals. Remove the obvious fiction. Weight the rest. Review the result with honesty. You'll learn more from a basic forecast that gets used than from a complex model that sits untouched.
Orbbit helps you find better-fit leads, research them faster, and turn that research into personalized outreach. If you want a stronger pipeline behind your forecast, Orbbit gives B2B founders and sales teams a practical way to spot relevant accounts, understand why they matter now, and act on that insight with personalized outreach.