[Webinar] Streamline your web hosting managementRegister Today

  • Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 1255
  • Last Modified:

Annualised Percentages?

I'm having a bit of trouble finding a good explanation of what an Annualised figure actually is.

For example, annualised staff attrition.

Within a 1 month period, there were X amount of staff employed, and Y amount left the company.

What is the formulae for working out the Annualised Attrition %, and what does it mean, and what is it good for?

I've come across several ways of working this figure out, but i've never got a good explanation of why a person believes theirs to be better than the next.

All help appreciated.
  • 4
  • 3
  • 2
  • +4
1 Solution
It's very simple:

if you're in the company Public Relations department, the best formula is whatever one gives the lowest number that you can truthfully put out in the press release, so you can write "Amalgamated Sludge International undertook a Hominid Resource Reallocation Process which resulted in just 0.9166% fewer employees on an annual basis."

If you're the guy in charge of streamlining the company, the BEST formula is whichever one meets your promised goals, so you can truthfully say "Annuallized, I cut the payroll as promised by 36%."

You can get EITHER FIGURE by using different assumptions:

(1)  We cut 11% of the employees this month, but won't cut any the other 11 months of the year, so that's 11/12% annualized.

(2) We cut 3% of the employees this month, if present trends continue, that's 36% annualized.

Numbers are such wonderful things.

niceyAuthor Commented:
So, it's an assumption of attrition over the course of 12 months (if going by month figures), based on the current month?

I've seen the following formulae used for figures given in February :

((no. of leavers this month)*2)*6) / no. of employees in month

Or the leavers and no. of employees maybe the other way round?

Apparently, multiply the figure by the month number (feb. is the 2nd month), then multiply that figure by whatever makes the 2 up to 12.  So, 12 divided by 2 = 6,

Is this a correct way of doing this, or is it just rubbish?

I've also seen that you just do somthing like, (employees * 12) / leavers.

I am getting confused still.
niceyAuthor Commented:
Ok, let's say that we have 180 people employed within 1 month.

5% of them leave.  That's 9 people.

Would the Annualized attrition be 60%?

And would that be saying to us that over the next 12 months, if this months trend continues, we'll have lost 60% of our employed staff?

300 points for the person who helps me here.  Once i've cracked this, i'll take a look at how it works in the financial world. :-)
Free Tool: Subnet Calculator

The subnet calculator helps you design networks by taking an IP address and network mask and returning information such as network, broadcast address, and host range.

One of a set of tools we're offering as a way of saying thank you for being a part of the community.

Like most business analysis, it's basically bogus because:


That's an irrefutable fact, but commonly ignored, because it makes about 60% of business-like analysis totally irrelevant.

You can get around this by taking one of several positions:

(1)  "The future will be like the immediate past".  In that case, you'll lay off 5% next month, and so on.  Total layoffs, 60% per year.

(2)  "The future will be like last year".  

  (a) Last year your company laid off a bunch of people one month, none the rest of the year.  With this assumption, you apply it to THIS year, no more layoffs, annualized layoffs: 5%.

(b) Last year your company laid off a bunch of people two months, none the rest of the year.  With this assumption, you apply it to THIS year, assume another month of  layoffs, annualized layoffs: 10%.

(3) "The future will follow the Dow Jones Averages".  Last year the Dow gained 20%, your company hired 10%.  This year the PREDICTION is for the Dow to go up 10%, so you predict the company will lay off 5%.


You can get ANY number you want out, just state your assumptions and your method, and you'll get at least a C+ or B grade.  Nobody can say with certainty what will happen, all you can do is make a logical argument.

niceyAuthor Commented:
Fair points.

But if sombody asked for an Annualized figure, based on attrition rates in a month, it would be :

(no. of leavers in month * 12) / no. of employees in month.

"annualised" simply means that you estimate what will happen in one year on the basis of partial figures relative to the current year and any other trend-figures (i.e. historical figures of the previous years, estimated relationships between the indicator you are annualising and other indicators and parameters, seasonal fluctuations, ...)

Basically, you have your experience plus partial figures relative to a year and estimate something for the whole year.

That's why grg99 [rightly] insists on saying that you can annualise in a way that supports your message and [hopefully] convinces your audience.

If x% of your workforce has left between Jan and Feb, you could say that the annualised figure is x*6.

But you could also say (for example) that over the past years (or in the industry, or in your geographical area, or for the avergae age of your employees, or for the sex or ethnicity or what-have-you of your workforce, ...) the rate in Jan/Feb is twice as high as in the other months (or any other figure that came out of any single statistics, even if all the other statistics disagree!).  Therefore, as each following month will contribute half as much as (Jan+Feb)/2, the annualised figure is
  (x/2)*(2 + 10/2) = x*3.5

Alternatively, you could say that the resignations in Jan/Feb are due to the fact that there was a rumor of take-over (or a change of management, or any other event you can convince your audience of) and that so-and-so many more people left (or so-and-so many fewer people)...

You get the idea:  you can honestly try to determine how many resignations you expect in the current year on the basis of everything you know or you can basically achieve any figure you want if you have enough imagination.  Only one condition must be respected: you must refer to the current year and must also include partial figures for the current year in your calculations, otherwise you cannot speak of annualisation but only of projection.
niceyAuthor Commented:
I'm getting the picture clear here.  Nice going.

But, just how did you come to "(x/2)*(2 + 10/2) = x*3.5" in your example?
Well, to make an example I had said that the rate for Jan/Feb was estimated to be twice the rate of the other months.  It was just an example, but if you have x resignations in Jan/Feb, it means that the monthly rate for that period is x/2.  If the rate were the same the whole year, to have the yearly figure you would simply multiply the monthly rate by 12:
But I had said that the monthly rate between March and December was half the rate of Jan/Feb.  Therefore you get:
  (x/2)*(2 + 10/2)
1 for Jan, 1 for Feb (which makes 2), and 1/2 for each one of the remaining months (which makes 10/2).

I could have also said:
  x + (x/2)*10
where the first x corresponds to the people who left in Jan+Feb, x/4 is the rate of the remaining 10 months.

But obviously nowhere is written that twice as many people leave in Jan or Feb than in any other month.  It was just an example.  In fact, in reality (just as a curiosity), more people leave at the beginning of Spring, because that is when the budgets are allocated but the money has not yet been spent.  Therefore, there are more job offers.  At least that's how it is in western Europe.

You have had some good input already...

" I'm having a bit of trouble finding a good explanation of what an Annualised figure actually is.For example, annualised staff attrition."

In the most simple case if  3 people leave in january you could annualize that by
multiplying  3 * 12 = 36 .

Usually annualized things are related to forecasts which use historical data to refine
or lend some credibility to the forecast. If january is not a typical month for people
leaving it could be misleading.........

So in short annualizing is to provide a number which is to represent a 12 month period.

Usually it is to take data for part of a year and multiply to yield a 12 month number.

I suppose you could also use an average to employ as an annualized rate.
Let us suppose you have data that is a single number for a 5 year period.
You could divide that number by 5 to give an annual rate approximation.

It is important to recognize that not all monthly data are typical. If you were to
annualize church attendance by a month such as December ( Christmas) or
the month for Easter you would vasterly overstate the collection basket proceeds.

The word attrition makes me recall an interesting use of the word. I had the job
of running a PC next to a battle board for a computerized war game for the US Army
in Japan. We employed a random number technique to show the results of  some
actions. For example a blue army commander might order artillary fire of some
magnitude .... after we calculated the result on our PC the standard question was:
" How many people were attritted ? " 

I have been away from EE for a long time .... this is my first contact for a year or more.



google      define: attrition

a reduction in the number of employees through retirement, resignation or death

A natural reduction in work force as a result of resignations, retirements or death. Most unionized companies cannot unilaterally reduce their employment levels to cut costs, so management must rely on attrition to provide openings that they, in turn, do not fill.

google     define:annualized

Short term or current period data used to project results for a hypothetical full year that filters out seasonality and abnormalities.

mathematical manipulation of a statistic for a period shorter than one year (usually a quarter or a month) that tells you what would happen if the amount or the rate of growth (or whatever it is) were to be sustained for a year

google     define:rate

a magnitude or frequency relative to a time unit; "they traveled at a rate of 55 miles per hour"; "the rate of change was faster than expected"

Above are common usages of the terms although the meanings can vary considerably.

Using the above definitions, the RATE of ATTRITION would be the number of people lost divided by the number you started with divided by the amount of time over which the measurements were taken


Beginning of month:  100 people
End of month:            96 people
Number who left:        4 people
Rate of Attrition = 4 people/100 people/1 month = .04/month = 4%/month
Annualized Rate of Attrition =.04/month * (12 months/year) = .48/year = 48%/year

Do they replace the people who leave? If they don't the model would look like this:

Within a 1 month period, there were X amount of staff employed, and Y amount left the company.
So after 1 month there are (X - Y) people left
So There are (X - Y)/X of the people left  - Call this Z

After 12 months you would expect number of people remaining to be:

X*(((X - Y)/X)^12)
Well, I do believe to have given a good contribution, and I am not the only one who made useful comments.  I think that the points should be assigned.  From the comments of the author it seems clear that we were providing good input.
PAQed with no points refunded (of 300)

Community Support Moderator

Featured Post

Learn to develop an Android App

Want to increase your earning potential in 2018? Pad your resume with app building experience. Learn how with this hands-on course.

  • 4
  • 3
  • 2
  • +4
Tackle projects and never again get stuck behind a technical roadblock.
Join Now