Sales Bottleneck Analysis: Find Where Your Revenue Gets Stuck in 30 Minutes

Why does your sales team work hard and still sell too little?

Everyone is busy. The phone rings, quotes go out, the calendar is full. Yet revenue stays below what the effort would suggest. In this situation many owners reach for a list of measures: a new website, more advertising, sales training, a CRM. Each measure sounds sensible. Together they cost a year and a lot of money, and often little changes.

The reason is simple. Sales is a chain of stages, and a chain carries as much as its weakest link. If your quotes take two weeks, a better first conversation gains you almost nothing. Prospects just wait in the same queue in a slightly better mood. Bottleneck analysis reverses the order. It first asks which stage limits the result and leaves everything else alone for now. This piece asks: at which stage is your sales process stuck, and what do you do once you have found it? To answer it, we first clarify what a bottleneck is, then introduce the Three-Reading Test with a worked example, and show how to read the values, make full use of the bottleneck, and avoid typical mistakes.

If you are improving everywhere at once, you have not yet found out where things are stuck.

What is a bottleneck in sales?

The term comes from the Theory of Constraints, which goes back to the physicist and management author Eliyahu Goldratt. Its core: in every workflow there is, at any given time, one point that determines the throughput of the whole. In sales, throughput means how many orders come out at the end of the month. An hour lost at the bottleneck is lost for the entire workflow. An hour gained anywhere else does not produce a single additional order.

The method is well studied in production. Lewandowska-Ciszek and colleagues (2025) describe a single case study of a sterilization unit for medical devices. The team observed the workflow, measured times, and analyzed performance indicators. The unloading station turned out to be the bottleneck. Only that one station was improved: the robot that handles the trays was redesigned, its gripper holds more, and the buffer positions are used better. After that, cycle times fell and throughput rose. The authors stress that hidden capacity can be unlocked this way without major investments. They point out themselves that a single case cannot simply be generalized, and they call for studies across several sites and longer periods.

A bottleneck does not have to be a machine or a person. Knop (2025) studied a machinery company that produces bearing cages. The most important constraint there proved to be missing work instructions at every production stage. As the underlying cause, the study names insufficient investment in training and management development. Applied to sales, this means your bottleneck may be an overloaded person, but just as easily a missing rule, such as who follows up after a quote.

Looking at bottlenecks complements the work on handoffs between stages. How to close those handoffs with fixed rules is covered in the article on how to optimize your sales process. Bottleneck analysis tells you beforehand at which handoff to start.

The Three-Reading Test: bottleneck analysis in 30 minutes

The Three-Reading Test is the mental model of this article. It takes the same three readings at every stage of your sales operation: how many cases arrive there (volume), what share reaches the next stage (conversion), and how many days a case sits there on average (waiting time). A stage that stands out on two or three readings is your bottleneck.

You need no software for this. Your inbox, your calendar, your list of quotes, and the invoices from the last three months are enough. Set a timer. The time limit protects you from turning a quick diagnosis into a project.

1. Minutes 0 to 5: define the stages

   Write your stages in one row, for example inquiry, first conversation, quote, order. Four to five stages are enough. Use the names people in the company already use.

2. Minutes 5 to 15: count volumes

   For the last three months, count how many cases reached each stage. Count all of them, including the rejected and the dormant ones.

3. Minutes 15 to 20: calculate conversion

   Divide the volume of each stage by the volume of the stage before. This shows you at which step the largest share is lost.

4. Minutes 20 to 27: estimate waiting times

   Take ten cases per stage and note how many days passed between stages. The average will do. Precision to the day is not necessary.

5. Minutes 27 to 30: mark the bottleneck

   Mark the stage with the lowest conversion and the one with the longest waiting time. If they coincide, you have your bottleneck. If not, the pile-up decides: where do most open cases sit today?

A missing baseline is the normal case. Surdziel and Gdowska (2026) accompanied a medium-sized Polish IT consulting firm while it reorganized its presales work. The study draws on twelve interviews and two workshops with sales, technical staff, finance, and the delivery organization. No prior performance data existed. The baseline values were estimated from several sources: a win rate of roughly 25 percent, an average response time of 14 days, and an average margin of 25 percent. Only with these numbers could targets be set, namely a win rate of at least 35 percent and a response time of at most 10 days. These are targets. Whether they were reached cannot be seen from the study’s summary.

What does the result look like? A worked example

The following numbers are freely chosen. They show what the Three-Reading Test can look like for a business with four stages.

In the example, the lowest conversion and the longest waiting time coincide: after the quote, 21 days pass on average, and only 30 percent of quotes become orders. So the bottleneck sits between quote and decision. Conversion at the start is good at 80 percent. Anyone buying more inquiries for this business just adds to the queue behind the quote. Which numbers you should keep an eye on permanently is covered in the article on sales metrics for small and mid-sized businesses.

How do I read the values correctly?

The three readings tell different stories. Low conversion with a short waiting time usually means that the wrong cases enter the stage. A long waiting time with good conversion means the stage works but lacks capacity. If both coincide, a clear owner is usually missing.

Take care not to confuse the bottleneck with the loudest problem. An angry customer, a lost major deal, or a complaint from the team sticks in memory. The real bottleneck is often inconspicuous because simply nothing happens there. Similar patterns at the level of the whole company are described in the article on seven brakes on growth.

What do I do with the bottleneck? Exploit before you expand

The Theory of Constraints works with five focusing steps. The case study by Lewandowska-Ciszek and colleagues follows them as well. For sales, they translate as follows:

1. Find the bottleneck. That is what the Three-Reading Test does.
2. Exploit the bottleneck. Get out of the stage whatever is possible with existing means. Whoever writes quotes does nothing else during that time.
3. Subordinate everything else. The other stages work at the pace of the bottleneck. There is no point in holding more first conversations than quotes can be written.
4. Expand the bottleneck. Only now do investments come into question: a second person allowed to calculate prices, a template, a tool.
5. Start over. Once the bottleneck is resolved, it moves elsewhere. Then you repeat the test.

The second route costs almost nothing and shows within a few weeks whether the diagnosis is right. In the Polish case studied by Surdziel and Gdowska, it was likewise simple means that stabilized the workflow: a dedicated presales unit, clearer roles, early involvement of technical experts, and forms for barriers as well as for won and lost bids. The authors explicitly speak of lightweight interventions without advanced technology. They also describe why such rules become necessary: in fast-growing firms, workflows emerge on the side, depend on individual people, and over time move away from what was once agreed. The study calls this process drift. As results it reports early observations such as less drift, better collaboration across departments, and more visibility for management, but no measured improvements yet. How to set up such an improvement beyond sales is shown in the article on process optimization step by step.

Which mistakes are most common in bottleneck analysis?

• Counting only orders won. Then every stage looks fine. The message lies in the cases that were never brought to an end.
• Too many stages. With nine stages and six cases each, every rate is chance. Stick to four or five.
• Looking for the bottleneck in a person. Almost always a rule or time is missing. If you look for someone to blame, you will get polished numbers next time.
• Tackling several points at once. Then you will not know afterwards what worked.
• Measuring once and never again. The bottleneck moves. One test per quarter is enough to follow it.

If you keep track of the stages continuously anyway, the one-off test becomes a routine. How such a list is built is described in the article on pipeline management for small and mid-sized companies. And if the test shows that your bottleneck lies behind the quote, you will find the matching levers in increasing revenue without more leads.

What you can do this week

Block one slot of 30 minutes and a second one of 15 minutes. You do not need more to get started.

• Write the stages in one row, five at most.
• Count volumes per stage for the last three months, including rejected and dormant cases.
• Note conversion and waiting time per handoff, and mark estimates with a question mark.
• Mark the one stage where conversion, waiting time, or pile-up stands out.
• Define a single rule for this stage: who, what, by when.
• Put the repeat of the test in your calendar four weeks from now.

In the second slot you discuss the result with the people who work at that stage. Ask them what holds them up. Usually they have known the answer for a long time. Until now, nobody had asked in a way that put a number next to it.

We started by asking at which stage your sales process is stuck and what to do then. The Three-Reading Test finds it with three values per stage: volume, conversion, and waiting time. The bottleneck is where cases pile up, conversion drops, and waiting time is longest. First it is used to the full with existing resources; only after that is money for more staff or software worth spending.