more-arw search

Q&A Forum

How much detail is necessary to deliver accurate forecasts?

My question is mainly towards headcount forecasting methods but am curious to hear people's opinions  on the matter in general.

I work in a company with 1000+ employees, and we currently manage the forecast at the employee/position level and I am wondering at what point do we ditch detailed forecast and start managing at higher level. Perhaps total units and salaries by department and region. 

I think we spend alot of time managing the details yet we rarely achieve accuracy on our headcount forecasts.


Topic Expert
Wayne Spivak
Title: President & CFO
LinkedIn Profile
(President & CFO, |

Have you examined the reasons for your variance?

It is possible that the algorithm and assumptions you are using as the basis of your forecasts are flawed, so whether you ditch detail or not, you are not correcting the fundamentals.

Mark Thurmond
Title: Managing Partner
Company: Thurmond Consulting Group, LLC
(Managing Partner, Thurmond Consulting Group, LLC) |

In general, I believe the amount of effort or modeling put into an area of a forecast should be directly related to the materiality of that area. So if personnel costs are material, then it makes sense to model them at a detailed level, such as headcount, to gain accuracy in the overall forecast.

If you are not getting the accuracy you want in the personnel area, then there could be issues in your modeling. Personnel should be able to modeled quite accurately as the related costs (taxes, benefits, etc) are very calculable. Variances that come in the size and timing of items such as raises and bonuses are fairly easy to track. Also, if your forecasting tool is a good one and is setup properly, the maintenance to the Personnel data that drives the forecast in this area can be managed with proper procedures.

Obviously at some point this theory can break down if the volume of information gets too large. With 1000 employees, if one person is maintaining the data, it should be manageable. But if you have a distributed model and you have 10 departments who will each manage their 100 employees, it is very manageable. There is also the factor of the source data to feed the forecast with accurate personnel data. If you set up an interface with your HR system, things get much easier. So the cutoff as to when to switch from Headcount to something less granular is going to depend on many factors that will be unique for each organization.

Topic Expert
Marc Faerber
Title: CFO
Company: Amarantus
(CFO, Amarantus) |

Wayne and Mark are correct. If your business is dynamic, high-tech etc., well that can be part of the nature of the process. How clear and dependable are the starting assumptions and why do you think they are dependable? If the process is solid and you still are missing your forecast identify the drivers of that and do some sensitivity analysis so you can be responsive to the impact of the fluctuations.


Get Free Membership

By signing up, you will receive emails from Proformative regarding Proformative programs, events, community news and activity. You can withdraw your consent at any time. Contact Us.

Business Exchange

Browse the Business Exchange to find information, resources and peer reviews to help you select the right solution for your business.

Learn more

Contribute to Community

If you’re interested in learning more about contributing to your Proformative community, we have many ways for you to get involved. Please email to learn more about becoming a speaker or contributing to the blogs/Q&A Forum.