BFS-logotype-rvb TORSA TABLE D148CM - TECK NERO/CERA
BFS-logotype-rvb TORSA TABLE D148CM - TECK NERO/CERA

BFS-logotype-rvb TORSA TABLE D148CM - TECK NERO/CERA

Forecasts? Never trust them! How to improve the quality of your forecasts and provide numbers you can actually trust.

05.10.2021

Forecasts? Never trust them! How to improve the quality of your forecasts and provide numbers you can actually trust.

“There are two kinds of forecasters: those who don’t know, and those who don’t know they don’t know.” ― John Kenneth Galbraith

Deco
Deco
Deco
Deco
Deco

A forecast is by definition inaccurate, incomplete or simply wrong. Still, a good forecast is key to set priorities and organize the work of your team while managing expectations

across the organization. If your business forecast is still a copy of

last year's results with a few changes here and there, then this article

is for you! Read on as we develop three ideas that will help small and medium business leaders improve the quality of their forecasts and better communicate their progress.

But first, what is a good forecast and why is it important? A good forecast correctly captures trends, seasonality and future events

that could impact your business metrics within a specific time frame in

order to better prepare today in the face of a possible future/outcome.

Good forecasts are indeed extremely helpful to organize your team's

work, review your strategy and better set your priorities although being

often inaccurate - and, to a certain extent, that's ok. Here is an

example:

Corporate treasurers are traditionally good forecasters

for a simple reason, if they run out of cash the business collapses.

Thus, they are constantly checking the business plan, account

receivables and payables, inventory and cash levels in order to improve

their cash forecast. As a result, one month forecasts are generally very

accurate while six month forecasts are much less so. However, longer

forecasts should signify a trend while giving treasurers time to work

upon its outcome: "after performing a thorough analysis of the treasury

schedule, the treasury manager realized the company may not be able to

pay its debtors on time in six months from now". In this case the

workable outcome could be: six months to obtain a credit facility in

order to pay debtors on time - which is a task that couldn't easily be

performed in one month.

As companies are ever more dependent on

accurate forecasts to organize their activities and optimize their

supply chains, managers should be able to provide and communicate on

reliable forecasts. Here are three things you can do to improve the

quality of your forecasts:

Choose a foreseeable time frame that would still give you enough time to act upon its possible outcome. Indeed,

when you look at a 30 days forecast you know exactly what to do but its

often too late to influence the outcome, while a 180 days forecast

gives you plenty of time to do something about it but you don't know

exactly what. Perhaps (and this depends a lot on your industry) a 90

days forecast is just what you need to figure out what to do while still

having enough time to materialize the outcome of your forecast.

Re-forecast. As

you move along the weeks, a series of assumptions are now being

confirmed or dismissed, update your forecasting model and improve the

quality of your previsions. An easy way to do that is by simply

replacing forecasted periods by their actuals when they become available

so that you improve the accuracy of your quarterly forecast (more on

this in the table below).

By creating a good forecast you have a first mover advantage:

use it! Indeed, when you act upon the outcome of your forecast you are

doing so before others do. Which means that you can assess your strategy

earlier (usually after 30 days) and adapt if needed (you still have 60

days left).

Now that we've learned to create reliable forecasts, let's see how to communicate them:

The

table below is a structured way to communicate on your forecasts but

many variations exist (it really depends on your industry). The table

below is specifically aimed at a quarterly sales forecast

and has been divided in 4 columns (initial situation (end of December),

end of the first month (January), end of the second month (February)

and end of the quarter (end of March)). Keep in mind that the outcome we

are interested in is the Quarter and not the result of each month

individually, this is the reason why the quarters (sum of the months)

are presented separately. On each column, the grey dataset represents

the actuals (zero in December and 355 € by the end of March for Q1)

which are the sales the company has effectively made over the first

quarter.

Below the grey datasets we have presented the Forecasts.

The yellow set is the initial forecast (often called Budget), the green

one is the sales forecast that integrates one month of actual sales. The

blue set is the best-forecast that integrates two months of actual

sales and one reforecast. On the last column we can compare the actuals

to the best forecast and see how they converged over time.

No alt text provided for this image


This

is the story the dataset is telling. The sales manager has initially

forecasted sales of € 350. After one month of work s/he had sales of €

90 (€10 below her/his initial forecast). Based on this first results,

s/he changed her/his strategy and by the end of February sales were € 20

higher than expected. As there was only one month to go and that

uncertainties had been reduced (compared to the end of December), the

sales manager has been able to reforecast March (now € 115 instead of

the initial €120) providing a best forecast of € 355 for the Quarter

which was in line with the Actuals that became available by the end of

March.

Indeed, by using this method, actuals were "known" one

month in advance and the forecast turned to be quite accurate. The other

positive outcome is that by structuring the information this way, the

sales manager was able to communicate effectively across the

organization reducing uncertainty over time.

For as much as this

methodology may be helpful, when it come to forecasts, here is the best

(and perhaps wisest) advice one can get:

"Hoping for the best, prepared for the worst, and unsurprised by anything in between." - Maya Angelou



Alain Rosenfeld,

a proud member of the BFS network.

Disclaimer:

all views expressed on this article are my own and do not represent the

opinions of any entity whatsoever with which I have been, am now, or

will be affiliated.






If you have a question, do not hesitate to contact us