The Economics of a Funnel: a pipeline is tied-up capital

A pipeline is tied-up capital: the costs for leads, qualification and selling work are spent, and revenue arrives only at the close. The time in between measures how long the money is tied up. That is why manufacturing measures every step and every cycle time. Our thesis: between lead and close, this measurement is mostly missing. Four separate quantities measure this time: the age of the stock, the time per stage, movement and the length of the whole run.

Why is a half-finished good tied-up capital?

In a plant, a half-finished part is tied-up capital. For as long as the part stands as stock, it has only cost money: running costs and a share of machine time. Only the run through to the finished product turns the tied-up capital into revenue.

That is why manufacturing measures every step and every cycle time. The reason is a sober one. The money is spent. The only open question is how long the money stays tied up and whether anything comes back.

This thought carries over to sales. For every open opportunity the acquisition costs have been spent. So has the selling work that has already gone into the opportunity. Revenue arrives only at the close, not before and not in proportion.

For the finance leader who checks the sales forecast, that is the core. Between money spent and revenue lies a stretch of time. This stretch is the subject of this whitepaper.

For every open opportunity the acquisition costs have been spent, and revenue arrives only at the close. The time in between measures how long the money is tied up.

What do three market figures say about the way from lead to revenue?

Three published figures give a reason to look at this stretch. All three are market values from surveys. None of the three figures is a measurement by Funnel Economics, and none measures how much capital is tied up.

Median spend on sales is 15% of annual recurring revenue (ARR), and median spend on marketing is 8%. The survey covered private B2B SaaS companies. The two medians are separate values and cannot be added up to a total.

Source: SaaS Capital, “2026 Spending Benchmarks for Private B2B SaaS Companies”, 2026, https://www.saas-capital.com/blog-posts/spending-benchmarks-for-private-b2b-saas-companies/

Sales reps report spending around 70% of their time on tasks other than selling. The figure comes from a survey of sales professionals in several countries. Salesforce is a vendor of software for sales teams.

Source: Salesforce, “State of Sales”, sixth edition, 2024, https://www.salesforce.com/news/stories/sales-ai-statistics-2024/

Four in five of the sales and finance leaders surveyed missed a quarterly forecast in the prior year. 52% of sales leaders were off their forecast by 10% or more. Sales and finance leaders in North America were surveyed. The figures come from a vendor of forecasting software surveying the problem it sells products for.

Source: Xactly, press release on the “2024 Sales Forecasting Benchmark Report”, 2024, https://www.xactlycorp.com/node/11261

The three figures show spend, working time and forecast errors. The three figures do not prove that the way from lead to revenue goes unmeasured. That statement is our thesis, not an established fact.

This whitepaper draws no comparison with manufacturing in figures. A funnel and a plant have no common unit. The manufacturing picture carries the reason for measuring, not a league table.

What does the chain from lead to close look like as a picture?

The manufacturing picture can be drawn as a chain. In the plant, raw material arrives, is refined and inspected, goes onto the line and leaves the plant as a finished product.

The chain is a drawing, not a finding. The picture does not supply two things. The picture supplies no rate from one lead stage to the next. Nor does the picture supply a fixed order of the stages: which stage sits where is something each company sets in its own configuration.

Which three valuations exist for the same stock?

Funnel Economics calculates with the stage win rate. The stage win rate is the share of won deals among the decided deals that stood in that stage. The stage win rate is not a transition into the next stage.

Three valuations of the same stock follow from the stage win rate.

bEFV (Baseline Expected Funnel Value) is the structural expected value of the funnel: each opportunity's amount multiplied by its stage win rate, without regard to current sales hygiene. bEFV, the structural expected value, describes the potential of the stock.

rEFV (Risk-Adjusted Expected Funnel Value) is bEFV weighted by SLA Health. The weighting uses measured factors per band. rEFV, the risk-adjusted revenue expectation, values the same stock. rEFV is an expectation, not a floor.

EVL (Execution Value Leakage) is the difference between bEFV and rEFV: the distance between two expected values of the same open pipeline. EVL is an amount of money, but not lost revenue. More on this is in the entry EVL, the distance between bEFV and rEFV.

One sentence holds for all three valuations: an expected value is not a payment stream. An expected value says what the stock is worth as a forecast. An expected value does not say what has been earned.

An expected value is not a payment stream. The money is spent; whether anything comes of it has not been decided yet.

Which four questions does the finance team ask about the time capital is tied up?

The time between lead and close is not one quantity. Four questions can be asked about this time, and each question has its own quantity. The four quantities are not interchangeable.

In a plant, nobody answers the question of how long a part sits at a station with the age of the part. The same holds in the funnel.

How old is the stock?

SLA Health classifies each open opportunity as HEALTHY, DELAYED or STALLED. SLA Health compares the opportunity's total age in the funnel with the two SLA values the customer sets per stage and opportunity type. The entry SLA Health, the classification by total age, describes the three bands.

The health band is a statement about age, not about movement. An opportunity can move on a stage every week and still stand in the STALLED band. An opportunity without any movement can stand in the HEALTHY band.

How long does a deal sit in one stage?

BPV (Build-to-Perform Variance) measures, per stage, by what percentage the actual time opportunities spend in the stage deviates from the stage’s SLA. BPV, the deviation of time in a stage from the SLA, thus answers a different question from the health band.

The health band asks whether a deal is old for its stage. BPV asks whether a stage is slow. The two quantities are not the same statement, and neither confirms the other.

Does the deal move at all?

Dormant Opportunities are open opportunities that show not a single movement during an entire period. The quantity belongs to the view of how the pipeline is being built, not to the look back at a completed period. Where the question is asked for a look back, a question comes back instead of a number. Where the chosen model carries no data for the period, the quantity counts as not measurable; a zero is not reported.

When an opportunity is created during the period, that counts as a movement. Dormant Opportunities, open deals without movement, are not the STALLED band. A deal can stand in the STALLED band and have moved, and a deal can stand in the HEALTHY band without movement.

How long is the whole run?

Sales cycle length is the average number of days from the creation of a deal to the win. The measurement uses the deals won, with one value per deal won. The number of deals the average rests on belongs beside the length.

Sales cycle length serves as the denominator for daily values. A daily value over the full sales cycle is a normalisation, not a payment stream. The entry on Sales Velocity, the daily value of a completed period, explains this reading.

Age of the stock, time per stage, movement and length of the whole run are four questions with four quantities. None of the four quantities replaces another.

What has the stock cost?

The acquisition costs fall before the stock is worked. The opportunities sales is working on today mostly came in as leads in earlier periods. The costs for them fell in those earlier periods.

The costs arise at several points: at the lead, at qualification and at the opportunity. Each point has its own unit cost.

Unit costs come from the company's own figures; the method does not supply any of its own. Where unit costs are missing, there is no cost figure.

This whitepaper therefore names no cost ranges per lead or per opportunity and no cost per sales rep. Whoever wants to know what their own stock has cost calculates with their own figures.

What does acceleration say, and what does it not say?

Acceleration is a statement about time, not about probability. Shortening a run shortens the run. The acquisition costs then stay tied up for a shorter period.

Whether a deal wins more often once it is made healthy is unchecked. An old deal can be old because it is weak. The health band is a statement about age, and it enters rEFV through measured factors per band. What happens when a deal is deliberately moved into a better band, the band does not say.

This whitepaper therefore claims neither more revenue from a shorter sales cycle nor that it brings no additional revenue. A company can check the question only with comparison cases from its own deals.

Where does the manufacturing picture end?

The manufacturing picture is an aid to understanding. The measurement begins where the picture ends. Six limits belong to the picture.

The six limits come from the method documentation of Funnel Economics.

The picture explains why the quantities exist. The picture supplies no quantity. Every measured figure comes from the company's data or from a named external source.

What follows for the CFO and the CEO?

For the finance team, a sequence of questions follows. First comes the question of which valuation of the stock is meant: potential, expectation or the distance between them. Then comes the question about the time: age, time per stage, movement or length of the run.

For the chief executive looking at the revenue forecast, a rule for the conversation follows. Whoever names a figure about the pipeline says which of the four questions the figure answers. A figure without its question cannot be checked.

The yardstick from within the company is a subject of its own. The article on Internal Benchmark Arbitrage, the gap to the top quartile, covers it.

Funnel Economics Core calculates bEFV and a simplified rEFV by health status in your Salesforce org. rEFV with measured factors per band and BPV are calculated by the optional add-on Funnel Economics Strategist.

Funnel Economics is entering its beta phase. During the beta, you start by running it against a Salesforce sandbox.

Discover is meant for trying Funnel Economics before any installation. It is built on a demo org.

Method and data basis

This whitepaper describes quantities of the method: definition, basic formula, interpretation and delimitation. The statements about the method come from the method documentation of Funnel Economics. The numbers of the example are invented. This whitepaper contains no measured values of any organisation.

The quantities of the method

The definitions and basic formulas come from the method documentation of Funnel Economics.

How the quantities are read

rEFV is an expectation: a calculated value, not a result. EVL is a distance between two expected values, not a loss. This whitepaper names no target value and no threshold for any of the quantities.

EVL is read as a series: the share of EVL in bEFV against its own history, not as a single value. A single value without its series says little.

The external figures

The three figures in the second chapter come from published surveys by SaaS Capital (2026), Salesforce (2024) and Xactly (2024). The three figures are market values. Salesforce is a vendor of software for sales teams, and Xactly is a vendor of forecasting software.

What the example shows and what it does not

The example in the chapter on the four questions uses invented numbers. The example shows that four questions about the same stock give four different answers. No reference value for age, time, movement or length can be derived from the example.

A pipeline is tied-up capital, and the time to the close can be measured. The manufacturing picture explains the reason; the figures come from the company's own data.