Businesses are often required to build Projection Models basis current / past performances to set expectations for the way forward. For such models, we typically use CAGRs (and not AAGRs).

But first – let us define the jargons:

CAGR stands for Compounded Annual Growth Rate and is, as the name suggests, the compounded rate of growth. CAGR formula is CAGR = (Ending Value / Beginning Value) ^ (1/No. of years) – 1. The CAGR smoothens out or diminishes the effect of the volatility over time. This is the standard method for building projections. AAGR stands for Average Annual Growth Rate and is, as the name suggests, the simple average of the annual changes (can be both growth as well as decline). AAGR formula is AAGR = (1st year change + 2nd year change + … + nth year change)/ n. It is a linear measure and does not take into account compounding.

However, since CAGR is dependent on the first and last years under consideration – in case there are massive fluctuations in either of these years, one can cross check using AAGR. In terms of stable data, one typically sees both CAGR and AAGR being quite close to each other. And then one can take a more informed call as to which to use for the projections.

Some ground rules to keep in mind. The more past data one has, the better the projections. If one has monthly data for the last 5 years, then it is recommended that you create more data points by 1st calculating the Moving Annual Trends (i.e. the summation of 12 months of data) to get 60 data points instead of 5. The models created are likely to be more stable.

And for any projection, it is important to also share a range where they may eventually play in – to accommodate for the best case as well as the worst-case scenarios. For variability one can use the “Realistic / Optimistic / Conservative” method of Trend Lines. The Realistic Trend Line will be the Base Line of your projections. For the Optimistic and Conservative Trend Lines, one needs to look into the variable at play and then tweak then basis market understanding. This can tricky, and one needs to have a thorough understanding to come up with logical estimates. But if done correctly, this gives a business not just one but three scenarios to visualize. What is even better is that these projections are not cast in stone, and one can always update new numbers as and when available to make future projections even better.

Final comment – this needs to be dealt with carefully. Stakeholder buy-in is strongly recommended before documentation.

If you found this interesting, please do reach out to us to see how we can help you drive data driven business growth.