By Kaitlin Dunn, Writer, Hospitality Sales & Marketing International (HSMAI)
Last June, hospitality revenue professionals from around the globe tuned in to HSMAI’s ROC@Home, a day full of presentations focusing on revenue in the age of COVID. Heidi Gempel, managing partner and founder of HGE International, presented on “Forecasting Rooms in the Age of the Crystal Ball,” sharing her thoughts on how to forecast even though past analytics aren’t relevant and everything is unprecedented.
After nearly a year of lockdowns and restrictions, Gempel’s advice is more relevant than ever. Here are key takeaways from her presentation:
SHOULD WE FORECAST?
It sometimes can be difficult to keep up with ever-changing trends and try to forecast when there isn’t a lot of data. Is it even worth it to try?
“My answer is a resounding yes,” Gempel said. “We need to forecast because we need visibility. Although you may be staring at a very foggy landscape right now, keep going and don’t give up. Your business needs this. The organization can adapt to the changing trends that you as a revenue manager can share with them. We need to be able to respond when business comes back. It is a business necessity.”
PAST AND PRESENT
Even though historical data may not be the most important thing to factor in, it still provides an important baseline, Gempel said. “Historical data is useless to predict the future,” she said. “But it’s still important to understand how your business operated. You can use it as a baseline and measure against it.”
Gempel recommended looking at past data to tell how much money was spent trying to attract new guests compared to trying to bring back past guests. “On a property level, we don’t spend a lot of time understanding why guests come back,” Gempel said. “You can look at data to reach out to guests that may have a reason to come back, and it may be less costly than trying to find new guests to come for the first time.”
When it comes to more-current data, Gempel suggested looking at the correlation between when restrictions were lifted and travel demand for a specific location.
If you only build one forecast, it is likely to be incorrect no matter what you do, Gempel said. To combat this, she suggested looking at multiple models with various assumptions.
“My recommendation is to build multiple models,” Gempel said. “Come up with a model that has multiple assumptions and break it down and make sure to record it. It is critically important to record what you base decisions on. Then monitor what legislation is changing and restrictions that are being lifted to determine the implications.”