.svg){kind=small}
TL;DR
- Top-down forecasting starts with a market or growth assumption and allocates it downward. Bottom-up builds from individual deals, accounts, or cohorts upward.
- Bottom-up is more accurate for the current quarter. Top-down is more useful for annual planning and board decks.
- The best forecast is a blend: build both, compute the gap, and investigate. A gap >15% means one of the inputs is wrong.
- Healthy forecast accuracy is within 5% of quarterly commit. Over 15% variance either way is a process problem, not a sales problem.
- Fairview maintains both views simultaneously by joining pipeline, cohorts, and historical close rates from the CRM, Stripe, and the ad platforms.
The best-run finance teams I have worked with do not pick one forecasting method. They build both, compare the gap, and treat the disagreement as the actual output. The question is not which method is more accurate. It is what the two methods, honestly reconciled, tell you about the quarter ahead.
Bottom-up vs top-down forecasting is usually framed as a debate. It is not a debate. They answer different questions. Bottom-up tells you what the pipeline will likely produce if current execution continues. Top-down tells you what the business has to produce for the plan to hold. If those numbers disagree, the gap is where the real operating conversation lives.
This post explains how each method works, where each one breaks, the blended approach, and the accuracy benchmarks operators should hold themselves to. It complements the Cluster 2 post on RevOps vs Sales Ops and the Cluster 1 guide to finding profit leaks.
What is top-down forecasting?
Definition
Top-down forecasting: a forecast built from a market or growth-rate assumption, then allocated across segments, regions, or products. It starts with a number the business has to hit and works backward to plan.
Top-down is how boards think. Start with TAM, apply a market share assumption, overlay a growth rate, compare against the plan, and hand the number to the GTM leaders. It is fast, it anchors ambition, and it does not require clean pipeline data to produce.
It is also where most founder forecasts live in the first two years of a company. “We’ll triple revenue this year” is a top-down statement. It might be right. It is just not grounded in anything observable yet.
What is bottom-up forecasting?
Definition
Bottom-up forecasting: a forecast built from individual deals, accounts, or customer cohorts up to the total. For sales it means taking each open opportunity, applying a weighted probability, and summing. For D2C it means taking cohorts and applying observed retention and expansion curves.
Bottom-up reflects what is actually happening. The pipeline has real deals in real stages with real dollar values. Cohorts have real retention curves. The output is slower to produce, requires clean data, and usually disagrees with the top-down number.
For the current quarter, the bottom-up number is the one to trust. For next year’s plan, it is useless — there is no pipeline yet for Q3 of next year, and the cohorts that will drive that revenue have not been acquired.
Key insight
Top-down sets the goal. Bottom-up shows the path. If the path does not arrive at the goal, one of them is wrong.
The four differences that matter
| Dimension | Top-Down | Bottom-Up |
|---|---|---|
| Starting point | Market size, growth rate | Individual deals, accounts, cohorts |
| Key inputs | TAM, share, growth % | Pipeline $, stage conversion, close rates |
| Accuracy window | 12+ months | Current + next quarter |
| Best use case | Annual plan, board deck, fundraise | Commit call, hiring triggers, cash plan |
| Typical failure mode | Assumption fits the story, not reality | Missing pipeline for far-out periods |
A finance team that only does top-down misses the bottom-up signal that the quarter is going to miss by 12% until the final week. A sales team that only does bottom-up hits the quarter and still runs the business into a wall because nobody is setting an ambition that forces the hiring plan.
Which is more accurate?
Bottom-up wins on the current quarter. Top-down wins on the annual plan. The real accuracy gain comes from using both and watching the gap.
Industry data backs this up. Gartner’s sales forecasting research has found that bottom-up, pipeline-weighted forecasts outperform top-down anchored forecasts by roughly 10–15 percentage points on current-quarter accuracy in B2B companies with mature CRM data. The gap inverts for forward-year projections, where top-down market-based models are more stable than a bottom-up build that depends on pipeline that does not yet exist.
For most growth-stage operators, “which is more accurate” is the wrong framing. The right question is: when the two disagree, what are we learning?
The blended approach operators actually use
A workable cadence:
- Set the annual top-down at planning. TAM, growth rate, product split, region split. This is the commit to the board.
- Build bottom-up for the current quarter from pipeline. Weighted by stage, filtered for close date, cleaned weekly. The VP Sales owns this number.
- Compare the two weekly. Gap = (top-down allocated to quarter − bottom-up). Track the gap trend, not just the number.
- Investigate any gap >15%. Either top-down assumptions need revising, pipeline is missing, or execution assumptions (close rate, cycle time) changed.
- Update the annual from actuals quarterly. The top-down number for Q3 should get more grounded every quarter as actual Q1 and Q2 data lands.
Quote-ready
A forecast that never disagrees with the plan is not a forecast. It is a wish.
Forecast accuracy benchmarks
| Motion | Best-in-class | Healthy | Review below |
|---|---|---|---|
| Mid-market B2B SaaS | Within 5% | Within 10% | > 15% |
| Enterprise B2B | Within 8% | Within 15% | > 20% |
| D2C / eCommerce | Within 4% | Within 8% | > 12% |
| PLG self-serve | Within 6% | Within 12% | > 18% |
A quarter missed by 18% is not a sales problem. It is a forecasting process problem. Either the pipeline was not clean, the conversion rates being applied were outdated, or the top-down assumption never actually matched the market. Any of those three is fixable — but not by hiring more reps.
The three forecasting mistakes operators make
- Using stale close rates. The bottom-up is built on weighted probabilities. If the Stage 4 close rate was 62% a year ago and is 41% today, the weighted pipeline number is inflated until the rate is refreshed. Refresh stage-by-stage close rates every quarter, not every year.
- Collapsing segments. A blended 45% close rate hides that enterprise closes at 28% and SMB closes at 61%. Forecast at the segment level whenever the mix is material, which is almost always.
- Confusing ambition with forecast. The top-down number is the plan. The bottom-up is the forecast. Telling the board the forecast when you mean the plan is how trust erodes. Tell them both, and the gap.
How Fairview runs both views simultaneously
Fairview connects to HubSpot, Salesforce, Pipedrive, Stripe, QuickBooks, Xero, Shopify, Google Ads, Meta Ads, and HubSpot Marketing Hub via native OAuth. Once the sources are linked, the operating view computes bottom-up pipeline forecasts by segment and stage, compares to the top-down annual plan, and tracks the running gap week over week.
When the gap crosses a configured threshold, Fairview writes a named next-best action: "Mid-market bottom-up slipped 18% below plan this week. Driver: Stage 4 close rate dropped from 54% to 38% on deals >$75K. Coverage still adequate but conversion is the leak. Review discovery quality."
See pricing and tiers for the plan that fits your stack.
Both views
Bottom-up and top-down on one screen
Weekly
Gap tracking cadence
10
Native integrations live today
Key takeaways
- Top-down forecasting anchors ambition. Bottom-up reflects reality.
- Bottom-up wins current-quarter accuracy. Top-down wins annual planning.
- The best forecast is a blend: build both, track the gap, investigate >15%.
- Forecast within 5% of commit is best-in-class. Over 15% means the process needs work.
- Three common mistakes: stale close rates, collapsed segments, confusing ambition with forecast.
See the gap between your plan and your pipeline.
Connect HubSpot or Salesforce, Stripe, and your ad accounts. Fairview returns the bottom-up forecast and the gap to plan on day one. 14-day trial, no card required.
Frequently asked questions
Bottom-up forecasting builds the revenue number from individual deals, accounts, or cohorts up to the total. Top-down forecasting starts with a market size or growth-rate assumption and allocates it across segments and products. Bottom-up reflects what the current pipeline will likely produce. Top-down reflects what the plan requires the business to produce.
For the current quarter, bottom-up is more accurate because it reflects the actual pipeline in stage with real probabilities. For annual planning beyond two quarters, top-down tends to be more stable because most of next year’s pipeline does not yet exist. Best accuracy comes from running both and investigating the gap.
Annual planning, board decks, fundraise models, and hiring plans. Top-down is the right tool any time ambition has to anchor the business before real pipeline exists for the target period. It is also useful as a sanity check against a purely bottom-up build, which can under-call ambition when pipeline is still loading.
Current-quarter and next-quarter commit calls, weekly sales forecast reviews, and any place the execution team is held accountable to a specific number. Bottom-up is also the right input to cash planning and hiring triggers, because it reflects what the business will likely close, not what it hopes to.
Build both. Allocate the annual top-down down to the current quarter and compare against the bottom-up pipeline-weighted number. Track the gap as a percentage of the top-down each week. A gap under 10% is normal. A gap above 15% signals that either pipeline is missing, conversion assumptions are stale, or the top-down number is unrealistic.
Mid-market B2B SaaS teams should land within 5% of their quarterly commit to be considered best-in-class, and within 10% to be healthy. Enterprise motions are more volatile, so 8% is best-in-class and 15% is healthy. D2C tends to be tighter because transactional motions have more data density. Variance greater than 15% in either direction usually indicates a process issue rather than a sales issue.