Jonty Pearce · Nov 3, 2025
Jonty Pearce: Check your forecasting tools for the Erlang error
Markdown · JSON · Original source
’s Jonty Pearce argues that a forecasting tool’s reputation is no guarantee that it returns the right answer: the same algorithm can produce different results in different software. He highlights a possible Erlang C problem when a calculation uses factorials that grow beyond what Excel can handle. For a test with 2,000 calls per half hour, a 180-second average handling and an 80/20 service level, Pearce says the base result should be 208 agents, with no shrinkage or occupancy limit. He advises checking tools against this example rather than trusting their outputs. The practical concern is that an incorrect staffing calculation could go unnoticed; the article does not establish how many tools are affected or independently document the two online calculators Pearce says gave wrong answers. What algorithm do the tools you use apply, and do they return Pearce’s stated result for this test?
Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims.
Press play to watch. If playback is unavailable, open the source below.
Original source
· Article
Source details
Original source date: Nov 3, 2025 (2025-11-03T07:30:00+00:00).
Date basis: Date stated by the publisher.
Added to Escape Meridian: 2026-11-05T20:41:25.465Z. First observed: 2026-11-05T15:42:07.080Z. Retrieved: 2026-11-05T20:40:23.246Z.
Referenced by
Recorded citations and public file attachments. A reference alone does not establish agreement or influence.
No public references recorded here yet.
Limits
The article does not name the two online calculators Pearce says gave wrong answers or show their results. The article gives a test case and a warning, but does not establish how widespread the Erlang error is across forecasting tools. The expected result is presented by Pearce; the evidence does not include an independent verification of the calculation.
Open question
Which forecasting tools return Pearce’s stated result for the test, and what algorithm do they use?
Source-stated
Jonty Pearce warns that the same algorithm can yield different results in different software, so users should check their forecasting tools.
the same algorithm used in different software can actually produce different results.
Source-stated
Pearce says an Excel-based calculation for more than 171 agents needs careful checking because factorials can exceed Excel’s limit.
if you’ve got more than 171 agents in your contact centre and you’re using Excel, you need to be very careful of the formula that you’re using.
Source-stated
Pearce gives 208 agents as the expected base result for the specified Erlang calculator test, with no shrinkage or occupancy limit.
The answer you should come out with is 208 agents, assuming that you don’t have maximum occupancy.
Source-stated
The article reports Pearce’s claim that two online calculators gave wrong answers, but does not identify them or provide their results.
I went online and I found two online calculators that are actually giving the wrong answers.
Escape Meridian interpretation
An incorrect staffing calculation could go unnoticed, according to the article’s warning about tools returning wrong results.
there’s a real risk you’re getting incorrect results – without even knowing it.
Source and contributor context
Jonty Pearce · Article author
These are credits stated by the source; they do not establish a profile match or authorship of every passage.
Source media
Media declared by the original publisher; availability checked separately from the editorial reading.
Terms and references
Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.
Call Centre Helper
Call Centre Helper is a trade publication publishing guidance on contact center operations, performance metrics and customer service.
TIME
TIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services.
Editorial provenance
Version 1. Publication ID: 51cdbb93-33e9-41e6-8cea-af1a0dda3b5f.
Prepared by gpt-6-luna; grounding review by gpt-6-luna.
Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-05T20:41:25.465Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: useful_perspective.
Observation: 6c90475657b4c3af681be8b331697c431377a605b8d4bf72889b5ebe5a40d97c. Source text hash: e910d14c81de74509b11f95a16d6eec30d6f0f8bbafe0b334d77ec32d2a2ba0f.
Revision and source-check history
Version 1 · 2026-11-05T20:41:25.465Z
Jonty Pearce: Check your forecasting tools for the Erlang error
’s Jonty Pearce argues that a forecasting tool’s reputation is no guarantee that it returns the right answer: the same algorithm can produce different results in different software. He highlights a possible Erlang C problem when a calculation uses factorials that grow beyond what Excel can handle. For a test with 2,000 calls per half hour, a 180-second average handling and an 80/20 service level, Pearce says the base result should be 208 agents, with no shrinkage or occupancy limit. He advises checking tools against this example rather than trusting their outputs. The practical concern is that an incorrect staffing calculation could go unnoticed; the article does not establish how many tools are affected or independently document the two online calculators Pearce says gave wrong answers. What algorithm do the tools you use apply, and do they return Pearce’s stated result for this test?
Original source
· Article
Source details
Original source date: Nov 3, 2025 (2025-11-03T07:30:00+00:00).
Date basis: Date stated by the publisher.
Added to Escape Meridian: 2026-11-05T20:41:25.465Z. First observed: 2026-11-05T15:42:07.080Z. Retrieved: 2026-11-05T20:40:23.246Z.
Limits
The article does not name the two online calculators Pearce says gave wrong answers or show their results. The article gives a test case and a warning, but does not establish how widespread the Erlang error is across forecasting tools. The expected result is presented by Pearce; the evidence does not include an independent verification of the calculation.
Open question
Which forecasting tools return Pearce’s stated result for the test, and what algorithm do they use?
Source-stated
Jonty Pearce warns that the same algorithm can yield different results in different software, so users should check their forecasting tools.
the same algorithm used in different software can actually produce different results.
Source-stated
Pearce says an Excel-based calculation for more than 171 agents needs careful checking because factorials can exceed Excel’s limit.
if you’ve got more than 171 agents in your contact centre and you’re using Excel, you need to be very careful of the formula that you’re using.
Source-stated
Pearce gives 208 agents as the expected base result for the specified Erlang calculator test, with no shrinkage or occupancy limit.
The answer you should come out with is 208 agents, assuming that you don’t have maximum occupancy.
Source-stated
The article reports Pearce’s claim that two online calculators gave wrong answers, but does not identify them or provide their results.
I went online and I found two online calculators that are actually giving the wrong answers.
Escape Meridian interpretation
An incorrect staffing calculation could go unnoticed, according to the article’s warning about tools returning wrong results.
there’s a real risk you’re getting incorrect results – without even knowing it.
Source and contributor context
Jonty Pearce · Article author
These are credits stated by the source; they do not establish a profile match or authorship of every passage.
Source media
Media declared by the original publisher; availability checked separately from the editorial reading.
Terms and references
Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.
Call Centre Helper
Call Centre Helper is a trade publication publishing guidance on contact center operations, performance metrics and customer service.
TIME
TIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services.
Editorial provenance
Version 1. Publication ID: 51cdbb93-33e9-41e6-8cea-af1a0dda3b5f.
Prepared by gpt-6-luna; grounding review by gpt-6-luna.
Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-05T20:41:25.465Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: useful_perspective.
Observation: 6c90475657b4c3af681be8b331697c431377a605b8d4bf72889b5ebe5a40d97c. Source text hash: e910d14c81de74509b11f95a16d6eec30d6f0f8bbafe0b334d77ec32d2a2ba0f.
uncertain · 2026-11-05T20:41:25.465Z · anthropic/claude-opus-5.5
Publication: 51cdbb93-33e9-41e6-8cea-af1a0dda3b5f
attribution — pass: The draft correctly attributes the argument, the 208-agent test result and the two-wrong-calculators claim to Jonty Pearce of Call Centre Helper. It keeps them as his assertions rather than presenting them as established fact.
Call Centre Helper’s Jonty Pearce explains why you need to double-check your forecasting tools
summary — uncertain: The summary is faithful, but a general reader is left with unexplained central terms. It does not say Erlang C is a formula that contact centres use to work out staffing needs, or what a factorial is. It also omits that N, the number fed into the factorial, is the number of agents, so the 171-agent limit is unclear in the summary. The terms 80/20 service level (80% of calls answered in 20 seconds), shrinkage and occupancy are undefined. Smallest repair: add brief plain definitions and state that the risk applies above roughly 170 agents in Excel. The word possible also softens the source's firmer warning slightly.
Many contact centre forecasting tools use the Erlang C formula to work out staffing needs.
date — pass: originalPublished is 2025-11-03T07:30:00+00:00. This matches the publisher page's stated publication time.
Published On: 3rd Nov 2025 07:30
support — pass: Each claim matches a verbatim source passage. The limitations correctly note that the two calculators are unnamed and that the 208 figure is not independently verified. The unnoticed-error point is stated in the source itself. The question follows from the source's own advice to test tools and check their algorithm, and no unsupported benefits or costs are implied.
If your tool gives you anything other than 208 agents at the base level, something’s gone wrong, and you should check what algorithm it’s using behind the scenes.
media — pass: The selected media is the verified direct link to the original publisher article, which carries the full transcript and test. The embedded video is an alternative, but the article link is direct and relevant.
Challenge or correct this reading
Send a correction or an opposing source with this reading’s URL. Disagreement and uncertainty can remain unresolved.
Call Centre Helper is a trade publication publishing guidance on contact center operations, performance metrics and customer service.
Explore Call Centre HelperTIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services.
Explore TIMECall Centre Helper is a trade publication publishing guidance on contact center operations, performance metrics and customer service.
Explore Call Centre HelperCall Centre Helper is a trade publication publishing guidance on contact center operations, performance metrics and customer service.
Explore Call Centre HelperTIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services.
Explore TIMECall Centre Helper is a trade publication publishing guidance on contact center operations, performance metrics and customer service.
Explore Call Centre Helper