Divide the number of deals that reached the chosen result by the number of deals you started with, then multiply by 100% — that is how to calculate a conversion rate. For sales, the result is a purchase or payment; for an individual stage, it is moving to the next stage. The deals in the numerator must come from the same group as those in the denominator. This formula lets you calculate conversion for the whole journey from enquiry to payment and for each transition along it.
A business owner usually needs the percentage to make a decision: who should receive more enquiries, where to help the team, whether to change the terms of sale. This is where correct arithmetic can easily lead to the wrong decision. The conversion rate of a salesperson who receives repeat orders and that of a salesperson who introduces new customers to the company answer different questions.
I consider it dangerous to allocate enquiries using an overall conversion ranking before it shows the mix of work. Otherwise, the head of sales will give the person at the top of the ranking the buyers who are easier to sell to. That person will show an even higher percentage, and the sales team will take the result of the allocation as proof of skill. You can make sense of this without complex statistics. Let us start with exactly what we are dividing.
Conversion rate formula
Conversion rate = number of deals with the required result ÷ number of deals at the start × 100%. Both numbers relate to the same chosen group, and the result is counted as of the specified date.
The starting point can be enquiries, meetings held or proposals sent, depending on the question. If you need to know what share of proposals ended in payment, the denominator contains deals with a proposal. If you need the share of payments from all enquiries, the denominator will be larger. The label must show both points: “enquiry → payment” or “proposal → payment”.
Before calculating, write down three conditions:
- What counts as one unit: a separate purchase request, a deal or a buyer.
- Which event counts as the result: a signed contract, the first payment or full payment.
- Which enquiries to include and the date on which you check their outcome.
These conditions must not change between rows in the report. One buyer can place several orders: for buyer conversion, they count once; for conversion of individual requests, you count the requests. An advance payment and the balance on one deal do not make two successful deals.
With zero deals at the start, there is no conversion rate: you cannot divide by zero. If there were deals at the start but none reached the result, the rate is 0%. The difference matters: in the first case, there was no work in the chosen group; in the second, the work received has produced no result.
Conversion above 100% in this calculation means that the numerator is no longer part of the denominator. Look for a deal counted more than once or results from enquiries outside the starting group. This is a reason to correct the calculation before discussing a bonus.
How do you calculate conversion rate as a percentage?
Divide the number of results achieved by the number of opportunities to achieve them, then multiply the resulting fraction by 100%. A fraction of 0.15 corresponds to a conversion rate of 15%.
Take your own exported data and enter two numbers: “Payment received for ___ deals out of ___ incoming deals; ___ ÷ ___ × 100% = ___%”. If 24 deals out of 160 enquiries have been paid, the result is 24 ÷ 160 × 100% = 15%. This means that, as a proportion, fifteen enquiries out of every hundred in the chosen group ended in payment.
In a spreadsheet, simply divide the cell containing the results by the cell containing the starting number and apply percentage formatting to the result. There is no need to multiply by 100 as well when displaying it this way: otherwise, instead of 15%, the screen will show 1500%.
Growth in conversion can also be calculated in two ways. A rise from 15% to 18% is an increase of 3 percentage points. The relative increase is 20%: divide the difference of 3 by the previous 15 and multiply by 100%. Writing “increased by 3%” without clarification confuses two different changes here.
Convert percentage points into deals for your own volume: number of enquiries × increase in percentage points ÷ 100. For 200 enquiries, an increase of 3 points corresponds to six additional payments. This is arithmetic with the starting number unchanged, not a promise of results: you still need to find an action that will bring those six purchases.
How do you calculate sales conversion rate?
For the overall sales conversion rate, divide the number of deals in the chosen group that ended in a purchase by the group’s original size. For conversion between stages, divide the number of deals that reached the next stage by the number that entered the previous stage.
Gather the numbers for one group of enquiries. If 200 enquiries led to 120 meetings, followed by 60 proposals sent and 30 deals paid, the calculation looks like this:
| What we are calculating | Reached the result / entered | Conversion rate |
|---|---|---|
| Enquiry → meeting | 120 / 200 | 60% |
| Meeting → proposal | 60 / 120 | 50% |
| Proposal → payment | 30 / 60 | 50% |
| Enquiry → payment | 30 / 200 | 15% |
The table describes a journey in which every paid deal went through a meeting and a proposal. Use your own stage names and counts. Count how many deals have been at each stage by the calculation date: the deal records currently remaining in a column of your CRM system are not suitable for this formula.
For a sequence of stages, overall conversion equals the product of the fractions for each transition: 0.60 × 0.50 × 0.50 = 0.15, or 15%. You cannot add the three percentages or take their average. If the total across the stages differs from directly dividing payments by enquiries, first look for different groups, skipped stages and duplicate counting.
Do some buyers pay without a meeting? Separate that route out. Otherwise, dividing the number of all payments by the number of proposals from another branch will produce a convincing percentage with no clear meaning. The article “Sales funnel: stages, an example and where money gets stuck” explains how to separate real stages of a purchase from labels in CRM.
Two rows with the same 50% do not call for the same decision. In the calculation above, 60 enquiries did not progress from meeting to proposal, and 30 did not progress from proposal to payment. At the same time, not sending a proposal after a meeting can be the right outcome: it turned out that the company does not do the required work. Making a salesperson send a proposal to everyone is a quick way to raise an intermediate percentage without additional purchases.
So a changed conversion rate needs an analysis of the reasons alongside it: who the proposal is unsuitable for, who did not have one prepared for them and who is still deciding. For the business owner, a clear reason from the head of sales and a decision on how to address it are enough. A low figure in a row is not, on its own, a reason to change the price or send the whole sales team for training.
Enquiry conversion
Enquiry-to-sale conversion is the share of unique purchase enquiries that ended in the chosen result, such as payment. Website visitor-to-enquiry conversion is a different measure: its denominator is visitors, and its numerator is those visitors who made an enquiry.
Even within the sales department, two calculations are useful. The first shows purchases as a share of all enquiries. The second shows purchases as a share of suitable enquiries only: those for which the company actually sells the required product on terms the buyer can meet. Both are needed, but they must have different labels.
If your data contains 200 enquiries, of which 100 are suitable and 30 led to payment, the overall conversion rate is 15%, while conversion among suitable enquiries is 30%. The number of purchases is the same. The difference comes from changing the denominator. The report must show both counts; otherwise, excluding inconvenient enquiries will look like an improvement in the work.
Agree on the rule for a “suitable enquiry” before counting. A request for a product the company does not sell and an enquiry where the salesperson could not reach the buyer by phone mean different things. In the second case, you do not know what the buyer needed. You cannot remove that contact from the starting group just because they did not answer.
A repeat message about the same enquiry does not create another sales opportunity. Duplicates are merged, and test enquiries and spam are excluded using a consistent rule. The report retains the number of excluded records: then an increase in the percentage after cleaning the data is explained by the change in how records are counted and is not credited to salespeople.
To compare periods, enquiries also need the same observation period. Recent enquiries have not had time to follow the same path as older ones. The article “Increasing conversion rates matters to us. Why do new enquiries distort the report?” examines this separate mistake and its effect on advertising decisions.
Why averaging salespeople’s percentages distorts the team’s result
Calculate the sales team’s conversion rate from total results and the total number of incoming deals. If each employee has the same number of incoming deals, a simple average of their individual percentages gives the same result; when workloads differ, use the totals.
Enter your own counts in the following table. If one salesperson has 10 payments from 20 enquiries, and the second has 18 from 180, the result is as follows:
| Salesperson | Enquiries | Paid deals | Conversion rate |
|---|---|---|---|
| First | 20 | 10 | 50% |
| Second | 180 | 18 | 10% |
| Sales team | 200 | 28 | 14% |
The average of the two percentages is 30%: (50 + 10) ÷ 2. The team’s conversion rate is 14%: (10 + 18) ÷ (20 + 180) × 100%. The first calculation gave equal weight to twenty and one hundred and eighty enquiries. For the team, weight depends on the number of enquiries, so the second group has a greater effect on the total.
The same arithmetic works for five salespeople: add up their paid deals, add up their enquiries separately, then divide the first total by the second. If one deal was passed between employees, it must appear only once in the total number of incoming deals. For individual results, decide in advance whose work you are assessing: the person handling the enquiry at the start or the person taking it through to payment.
But even correctly calculated rates of 50% and 10% do not yet identify the best salesperson. Separate individual figures by new and repeat buyers, then, within comparable groups, by source and product. Compare the same transitions over the same observation period. If a difference remains at one point, the head of sales now knows whose work and which action to examine.
Keep the group size next to the percentage. With ten enquiries, one additional purchase changes conversion by 10 percentage points; with a hundred, by one. A decision to send all enquiries to the person with the best percentage from a small group makes the company dependent on a few outcomes.
When you have figures but no basis for changing how you work
If every comparison means having to establish the source, the salesperson and what the stage means all over again, the business owner gets percentages before they have a basis for a decision. A report is needed that retains the starting number, the result and what makes up the group for every measure; preparing it remains the head of sales’ job.
RENTROP has AI ROP for this task — sales team analytics: figures, the sales funnel, salespeople and money by source. The service reads data; a person decides how enquiries are allocated.
Instead of a line saying “sales team conversion — 14%”, the business owner needs a statement: “28 payments from 200 enquiries; new and repeat buyers counted separately; individual figures compared within matching groups”. If the leader’s advantage has disappeared after separating the groups, the previous ranking is no longer suitable for choosing who gets all the enquiries. Using it, the company risked rewarding its own allocation rule and penalising the person it had itself assigned the harder work to.