Dynamic number pool calculator for call tracking

How to use it

In GA4, find the hour with the most sessions from the sources you will track with dynamic numbers (for example, Google Ads only) and their average session duration. Enter both figures, the safety factor and the minimum number of numbers the vendor requires. If the vendor has confirmed that a number stays reserved for longer than the session, use that reservation time instead of the duration. Add the cost per number to see the monthly and annual bill for the pool. Everything is calculated in your browser.

What it calculates and the assumptions

The tool applies the method in the guide to sizing a dynamic number pool, which builds on Nimbata's estimator guide:

Numbers = peak-hour sessions × duration in minutes ÷ 60 × safety factor, rounded up and never below the vendor minimum.

• Average concurrent sessions = peak-hour sessions × duration ÷ 60. This is how many numbers are reserved at the same time on average.
• Occupancy = average concurrent sessions ÷ numbers in the pool.
• Monthly cost = (numbers − included numbers) × cost per number. The annual figure is the monthly one times 12.
• Sessions that find the pool full: if sessions arrive at random across the hour, the number of concurrent sessions follows a Poisson distribution with that mean, whatever the shape of their duration. The tool calculates the probability that a new session arrives when every number is reserved, and how many numbers it would take to bring that down to 1%. This is an added indicator, not part of Nimbata's method.

The 1.3 factor is Nimbata's proposal, not a standard. The guide does not set a rounding rule or a minimum either; WhatConverts, in its dynamic pool set-up, suggests a minimum of four numbers.

Worked example

Fictitious data, matching the default values: 150 sessions in the peak hour, a 4-minute average duration, a 1.3 factor, a minimum of 4 numbers and 3 dollars per number per month, the additional local number price published in WhatConverts' pricing for Spain for its single-account plans.

• Average concurrent sessions: 150 × 4 ÷ 60 = 10.
• Numbers: 10 × 1.3 = 13.
• Average occupancy: 10 ÷ 13 = 76.9%.
• Cost: 13 × $3 = $39 a month and $468 over 12 months, excluding taxes.
• New sessions that find the pool full: 20.8%. Bringing that down to 1% would take 19 numbers.

The sensitivity table shows how the result changes with half, 1.5 times and double the sessions and minutes:

The pool grows in proportion to both variables: doubling the reservation time costs as much as doubling the traffic.

The risk of reusing a number

Each dynamic number identifies one session while it is reserved. If the pool runs out, the system has to show a number another session still holds or fall back to a default number. In the first case, a call can be attributed to the wrong visit: someone else's campaign, keyword or page. In the second, the call has no source. WhatConverts sums it up in its capacity alerts guide: when every number in a pool is in use, new calls may be missed or misattributed.

That is why average occupancy should not approach 100%. At 77% occupancy the pool already fills up at some moments of the peak hour, and a campaign that concentrates visits into a few minutes makes it worse. The tool warns you when occupancy exceeds 80%.

How to read the result and its limits

The result is a starting estimate for ordering numbers, not a guarantee of attribution. It does not model peaks within the hour, returning visitors, calls made after the page is closed or numbers saved in a customer's contacts. Nor does it know each vendor's reservation rules: ask what extends the assignment and which number is shown when none is free. Validate the size with a capacity alert and by reviewing actual occupancy in the first campaign with peaks.

The cost covers only the pool numbers. Fees, minutes and transcription are budgeted separately with the guide to call tracking prices in Spain, and it is worth getting this figure in writing when you evaluate call tracking software in a demo.

The values you enter are calculated in your browser and are not sent to CallsIQ. The page measures visits with Google Analytics as described in the privacy policy.