With continual moving targets, shifting guidance, and volatile markets, hotel forecasting and budgeting has become borderline Sisyphean.
“Forecasts” and “budgets”—two words that create angst even in the best of times. In our current uncertain environment, these moving targets have never been so difficult and critical to predict. Budgets are typically just a longer-term version of forecasts with a bit of stretch built into them. However, the question of whether it is even possible to predict demand, and the resulting revenue, beyond the next 30-60 days is awfully risky to address with much confidence.
The reasons for even better forecasts today align with what we have historically been responsible for, but there is more on the line…at least for the near future. Owners and corporate/regional leaders have always needed some degree of accuracy with property-level forecasts, but now it’s about more than just figuring out projected bonuses—it is necessary to keep the doors open and people employed.
As a former hotel sales manager, director of sales and director of revenue management, I always felt responsible for creating and managing the demand to provide our team with hours to work and tips to collect. If I wasn’t successful at doing my job at the property, other people I worked with would feel the pain in their wallets. The massive decrease in travel has hurt us all and the volatility does not make it any easier to see the light at the end of the tunnel. What looks good one day may do a 180-degree the next, and vice versa. So, is it impossible to forecast better? Are these “unprecedented” times really that unprecedented?
We simply can’t throw our arms up and say “it’s too difficult.” We can do better than that. The hospitality industry has run into trouble before and we somehow waded our way through the muck. So, how do we do what we need to as far as financial forecasts and even our 2021 budgets?
Great question, but my suggestions are to not overcomplicate it and make sure you get consensus throughout the process. It is imperative that the revenue team and all consumers of the forecast are on the same page regarding the frequency of updates, the forecast window and defining what a successful forecast looks like.
From an above-property level, I would prefer an agile forecasting* approach that calls for weekly or even daily updates with an understanding that there may be shifts which would have been uncomfortable to see in the past. This way, I would have a more realistic expectation and major surprises could be avoided. Of course, a corporate, group-heavy hotel would be more difficult to forecast than a transient property, but this is where you need to get agreement on forecast performance and not necessarily forecast accuracy.
| *Agile Forecasting: to efficiently and continuously forecast (and re-forecast) in order to swiftly adapt to market changes |
I once worked at a property where the forecast accuracy was measured by using the mean percentage error (MPE)—which is easy to calculate and understand; however, the view of multiple periods using MPE becomes quite misleading since a -4.5% and +4.5% average to 0.0%. Using the mean absolute percentage error (MAPE) removes this illusion and provides a much better representation of how good (or bad) the forecasts were. This is a simple and common example of a lack of agreement on what is being measured, how it is done and, most importantly, why are we measuring it.
The “why” should drive the entire process. There are many different forecasts at a hotel, and they each provide their own set of data points. It may be strictly financial or perhaps a glimpse into expected check-ins/check-outs/covers or even the unconstrained demand for unqualified transient business. Each of these provide your owners, operations managers, finance folks, revenue managers and salespeople with the insights to adjust accordingly. Whatever you are measuring is, hopefully, done for a purpose. Understanding what that purpose is and how the consumers of it are using the forecast will allow you to provide a more accurate output (and save some time, too).
Now, if only the budget process could be simplified…
