Sales Estimation · Top-Down vs. Bottom-Up
Sales Estimation (Top-down & Bottom-up)
What is sales estimation?
- Top-down for opportunity, bottom-up for evidence
- Cross-check the two to test realism
- Present scenarios and sensitivity together
Estimating sales in two directions changes the language of the meeting. A target declaration like 'we have to hit KRW 10 billion (≈ US$7,481,700) next year' turns into an execution discussion: 'where do we generate 34 more inquiries a month?' Top-down sizes the place we can claim in the market, while bottom-up shows how much activity it takes to fill it. So the real output of an estimate is not a number but a list of verifiable assumptions, and what boards and investment reviewers probe is not the size of the number but the source of each assumption.
In practice, if the two methods land within 2x of each other, the gap is considered explainable. If they diverge by 3x or more, the customer definition or the pricing basis differs between them, so align the definitions before adjusting any numbers. One thing often overlooked here is the supply capacity constraint. Even if the bottom-up result is KRW 12 billion (≈ US$8,978,000) a year, if instructor availability and PM headcount can only handle KRW 8 billion (≈ US$5,985,300) worth of work, the real ceiling is KRW 8 billion (≈ US$5,985,300). Sales estimates and sales targets must also be kept distinct. An estimate is the range reachable with current capabilities, while a target is that number plus management’s commitment, so blending the two undermines the entire plan.
Top-down falls apart too if done carelessly. TAM, SAM, and SOM cut the market down in three steps. Even if public data can take you as far as SAM, from SOM onward you must narrow it to the range your sales team can actually reach. A new business has no track record, so build the bottom-up estimate on the first six months of results from similar offerings or on conversion data from three to five pilots. For an existing business, it is faster to use last year’s results as the baseline and adjust only for churn and price increases.
The most common misconception is that a bottom-up estimate is automatically credible. If you drop in values like a 25% conversion rate or 60% repurchase rate without evidence, that is just a top-down estimate in different packaging. Next to each cell, mark whether the value comes from the past 12 months of actuals, industry conventional wisdom, or hope. Scenarios work the same way: if you build optimistic and pessimistic cases by multiplying the base case by ±20%, it looks like three lines but rests on a single assumption. A scenario only means something when you change variables such as win rate and unit price directly rather than the result, and that lets you decide in advance whether to cut hiring or marketing spend first in a downside case.
John KimInstructor John Kim, CEO of Nanum Consulting and author of “B2B Marketing Design”
Kwangdong MinExpert in business models, revenue models, and market research| Category | Top-down | Bottom-up |
|---|---|---|
| Starting point | Total market size | Customers we can reach |
| Strengths | Shows the size of the opportunity | Validating assumptions and linking to execution |
| Weaknesses | Market share assumptions are arbitrary | Can underestimate the market opportunity |
| Used for | Explains the opportunity · Market attractiveness | Goal setting · Resource planning |
It grew out of demand forecasting practice and, combined with the TAM-SAM-SOM framework, became a standard step in startup investment reviews.
Consider a hypothetical case: Company A, a corporate training firm with 40 employees and KRW 6.2 billion (≈ US$4,638,600) in annual revenue. It put KRW 10 billion (≈ US$7,481,700) in sales into next year’s business plan and sent it to the board, with a single line as its basis: '0.5% of the KRW 2 trillion (≈ US$1,496,334,000) Korean corporate training market.' When an outside director asked, who exactly is going to buy that 0.5%?, the meeting came to a halt.
- The head of strategic planning pulled the CRM records for the past 12 months and set the actuals as the baseline: an average of 34 inquiries a month, a 22% win rate, and an average deal size of KRW 48 million (≈ US$35,900).
- When the head of sales recalculated by splitting new and existing business, renewals from existing clients came to KRW 3.8 billion (≈ US$2,843,000) and new business to KRW 4.2 (≈ US$1) billion, totaling KRW 8 billion (≈ US$5,985,300). For the first time, the KRW 2 billion (≈ US$1,496,300) gap from the original KRW 10 billion (≈ US$7,481,700) plan became visible.
- The CEO told the team to “cross-check against instructor availability.” With 12 full-time instructors, annual deliverable revenue came out at KRW 8.6 (≈ US$1) billion, which showed that KRW 10 billion (≈ US$7,481,700) is out of reach without hiring.
- Working back top-down, they narrowed SAM to KRW 420 billion (≈ US$314,230,100) in public-sector and large-enterprise HRD procurement budgets. The required share came to 1.9%, which looked defensible next to last year’s 1.5%.
- The pessimistic scenario was built by lowering the win rate to 18% instead of applying a ±20% multiplier. A response plan for sales of KRW 6.8 (≈ US$1) billion, which delays new instructor hiring by three months, was attached beneath the table.
- The final plan was adjusted to KRW 8.8 billion (≈ US$6,583,900) and approved by the board without changes. The marketing team received a target of raising monthly inquiries from 34 to 46, and in the first half of the following year it reached 103% of plan.