When forecasting revenue, most teams default to one method and stick with it. That creates a blind spot. Bottom-up forecasting — building from individual deals — is accurate when the deals are real but misses what isn’t yet in the pipeline. Top-down forecasting — starting from targets or market assumptions — is good at asking whether you have enough pipeline, but disconnected from the actual deals that will determine whether you close.

The most reliable forecasts use both. This guide explains how each method works, where each one breaks down, and how to build a layered forecast that combines them into something more accurate than either approach alone.

Two Ways to Predict Revenue: Starting from Different Directions

Think of bottom-up and top-down as two different questions about the same number.

Bottom-up asks: “Based on what’s in our pipeline right now, how much will we close this quarter?” It starts from deal-level data and adds upward.

Top-down asks: “Given our target and our historical close rates, do we have enough pipeline to get there?” It starts from a target and works backward to a requirement.

These are genuinely different questions, and the gap between the two answers is often the most important number in your forecast. If your bottom-up forecast says $800,000 and your top-down analysis says you need $1,200,000 in pipeline to hit quota, you have a coverage problem — not just a forecast number.

How Bottom-Up Forecasting Works

In a bottom-up forecast, each rep reviews their open deals and makes a commit: which ones will close this quarter, at what value, by what date? Those commits roll up to a rep total, which rolls up to a team total, which becomes the bottom-up forecast.

In practice:

1. Each rep reviews open deals in the CRM and flags which will close this quarter
2. Manager reviews the rep’s list, asks questions, and adjusts based on deal knowledge
3. Adjusted totals per rep are aggregated for the team
4. That aggregate becomes the official bottom-up commit for the period

Strengths of Bottom-Up Forecasting

The biggest advantage is that it’s grounded in real deals. Your reps know things about individual opportunities that don’t appear in CRM data: the champion’s internal status, the competitor’s pricing move, the procurement timeline that opened up last week. Bottom-up gives you access to that qualitative context.

It also creates accountability. When a rep commits a deal to the forecast, there’s a conversation on record about that deal’s status. Managers can track commit accuracy over time and use it to calibrate how much to trust each rep’s future commits.

Weaknesses of Bottom-Up Forecasting

Rep optimism is the core problem. Reps invest time and emotional energy in their deals. They naturally believe in the ones they’ve worked hardest. That means bottom-up forecasts are systematically tilted toward the optimistic end — not because reps are dishonest, but because the perspective from inside a deal looks different than the perspective from the data.

Bottom-up also misses what isn’t yet in the pipeline. If you need five more deals to close this quarter but none of them are in the pipeline today, bottom-up forecasting won’t flag that gap. It only sees what exists.

How Top-Down Forecasting Works

In a top-down forecast, you start from either your revenue target or your total addressable pipeline and apply historical rates to arrive at an expected outcome.

The most common top-down calculation:

• Open pipeline value × historical stage-weighted close rate = expected revenue
• Or: Revenue target ÷ historical win rate = required pipeline coverage

You’re not forecasting from individual deals. You’re applying statistical patterns to the aggregate.

Strengths of Top-Down Forecasting

Top-down forecasting forces the coverage question. Before you even look at which deals your reps are excited about, you can calculate whether you have enough pipeline to hit the number at historical conversion rates. If you have $3M in pipeline and a 25% win rate, your expected revenue is $750,000 — regardless of how optimistic your reps feel about their deals.

Top-down is also less contaminated by recency bias and deal-specific optimism. It doesn’t care that a rep is excited about a particular deal. It applies rates derived from dozens or hundreds of historical outcomes.

Weaknesses of Top-Down Forecasting

Top-down forecasting assumes historical patterns will hold. If you’re targeting a new market, trying a new pricing strategy, or working through a partner channel for the first time, your historical rates may not apply. The top-down forecast will be miscalibrated without you knowing it.

It’s also disconnected from deal reality late in the quarter. If you know a specific $400,000 deal will close in the next two weeks because the contract has been returned for signature, that certainty doesn’t show up in a top-down calculation using aggregate probabilities.

Forecasting Method	Data Source	Best For	Key Weakness	CRM Fields Used	Accuracy Indicator
Bottom-up (rep commit)	Rep assessment of individual deals	Late-quarter precision; qualitative context	Rep optimism; misses pipeline gaps	Close date, commit flag, deal value	Rep commit accuracy vs. actual close rate
Top-down (target-based)	Revenue target ÷ historical win rate	Coverage gap identification; early quarter planning	Doesn’t reflect current deal reality	Pipeline total, historical win rate, quota	Pipeline coverage ratio vs. target coverage needed
Weighted pipeline	Stage probability × open deal value	Mid-quarter projection; systematic approach	Stage probabilities may be outdated	Stage, stage probability, deal value	Weighted forecast vs. actual close rate over time
Combined (layered)	All of the above	Maximum accuracy; scenario planning	Most complex to maintain	All pipeline fields + manager adjustment	Range accuracy: actual result within stated range

Building a Layered Forecast That Uses Both Methods

The most accurate teams don’t choose between bottom-up and top-down. They use both — plus a third layer — to build a range rather than a single number.

Layer 1: Rep Commit (Bottom-Up)

The rep commit is the most optimistic layer. It’s every deal your reps say will close this quarter. Without adjustment, this number is typically 15-30% higher than what actually closes. That doesn’t make it useless — it defines the ceiling.

To get rep commits, run a weekly forecast call where each rep walks through their committed deals. In your CRM, add a “forecast category” field so deals can be tagged as: Committed, Best Case, Pipeline, or Omitted.

Layer 2: Weighted Pipeline (Data-Driven)

Apply your historical stage close probabilities to all open deals. This is your most systematic layer — it doesn’t take into account what reps say about specific deals, only the statistical likelihood based on where deals are in the process.

The weighted pipeline number tends to be more conservative than rep commits. If it’s significantly lower than the rep commit number, that gap deserves scrutiny: either your reps are being optimistic, or there are specific deals in late stages that will pull the actual close rate above the historical average.

Layer 3: Management Adjustment (Top-Down Context)

The manager reviews both layers and applies judgment: Are there known factors that make this quarter’s pipeline more or less likely to convert than historically? Are there specific large deals that will dominate the outcome in either direction? Is there an unusual seasonal pattern this year?

This isn’t about overriding the data with gut feeling. It’s about applying context the numbers can’t capture — a competitive move, an internal organizational change, an unusually large single-deal concentration.

The final forecast is presented as a range:

• Conservative: weighted pipeline × a downward adjustment factor
• Expected: weighted pipeline with manager adjustment applied
• Best case: full rep commit, assuming optimistic deals close

Reporting a range rather than a single number is more honest and more useful. Leadership can make decisions based on what’s likely while understanding the upside and the risk.

When Each Method Is More Reliable

The balance between methods shifts across the quarter.

Early in the quarter, top-down is more reliable. Most deals that will close are not yet in late stages, and rep commits at the start of a quarter are inherently speculative. At this point, the coverage ratio question — do we have enough pipeline? — is more actionable than any individual deal prediction.

Late in the quarter, bottom-up is more reliable. The deals that will actually close this quarter are visible and progressing. Reps and managers know specifically which commitments are firm. The weighted pipeline calculation may actually lag behind reality because deals that closed quickly from early stages aren’t well captured by historical averages.

For stable, repeating businesses, both methods tend to converge: your reps commit what the data suggests, and historical rates are predictive because your business is consistent. When both methods produce similar numbers, you have a high-confidence forecast.

For high-growth businesses, top-down tends to reality-check rep optimism. When every quarter is a new record and the team is confident, the top-down question — “do we have the pipeline structure to support that growth?” — is a valuable counterweight.

---

Frequently Asked Questions

Which forecasting method do most CRMs support natively?

Most CRMs natively support weighted pipeline forecasting — applying a probability percentage to each deal based on its stage and summing to a forecast number. Some CRMs also support rep commit fields where reps can categorize deals. True top-down forecasting — working backward from a target using historical rates — typically requires a spreadsheet or a BI tool alongside your CRM, since it’s a calculation about the pipeline as a whole rather than about individual deals.

How do we get reps to commit accurately instead of optimistically?

Track commit accuracy over time and make it visible. When a rep consistently commits ten deals and closes six, that’s a 60% commit accuracy rate. Share that with the rep not as criticism but as calibration: “Your actual close rate on committed deals is 60%, so when you have eight committed deals, we should plan our forecast on five or six closures.” Reps who see their own commit accuracy data become better at forecasting. They’re not lying — they’re learning.

How do we use forecast data without creating pressure that distorts it?

The risk is real: if forecast accuracy is tied to compensation or treated as a test of confidence, reps will sandbag. Keep forecast data in a separate conversation from performance reviews. Frame the goal as “accurate forecast so we can plan resources” rather than “prove your pipeline is healthy.” When reps see that an honest forecast leads to useful support — more marketing spend in a low coverage territory, manager engagement on a stuck deal — they’ll trust the process.

What’s a reasonable forecast accuracy target for a growing sales team?

For a team with fewer than 20 reps and less than two years of historical data, forecasting within ±20% of actuals is a reasonable baseline. As you build more historical data, improve data quality, and develop manager calibration skills, you can push toward ±10-15%. Very few teams sustain better than ±10% accuracy consistently, especially in businesses with lumpy large-deal revenue. If someone claims they forecast within ±5% every quarter, ask whether their forecast is genuinely predictive or whether it’s being adjusted after the fact.