Restaurant Forecasting in Distil Diagnostics

Introduction to Forecasting

Forecasting in Distil Diagnostics helps teams make proactive decisions in operations, reservations, marketing and finance to achieve cover and revenue growth. It helps answer questions such as: What services need attention? Where can we increase bookings? Where can we reduce controllable costs such as labour or stock?

Forecasting is part of the engine that powers Distil Diagnostics. You can find out more about its companion, Availability Scanning, here. 

How Forecasting works in Distil Diagnostics

Forecasting in Distil is built on our own machine learning technology and data made accessible through our partnership with OpenTable. 

It incorporates four key data sets:

• Historical data from each site in the group
• Trends across your group portfolio, not just one site
• Market-wide benchmarking from OpenTable’s 60,000-strong restaurant network
• Live demand signals from people searching for availability in the future.

Forecasting is available for both established and new venues.

• For established restaurants, the algorithm uses up to one year of historical data and then continues to learn from new information over time
• For new venue openings, the algorithm draws on group benchmarks and wider market trends to generate an initial forecast, even before the site has operational data.

Each restaurant gets its own forecast algorithm, which continues to learn over time and runs every single day.

Currently, Distil Diagnostics Forecasting is available to restaurants on OpenTable. Forecasting for restaurants on other booking platforms (such as Seven Rooms or Triple Seat) is available through the Distil Group Reporting suite.

Forecast: Summary View

The Summary view is the first screen displayed in Distil Diagnostics once onboarding is complete. 

It shows group-wide revenue performance for the next 5 weeks, split by day and a cumulative view, plus a cumulative view for the full year. 

This key metric be toggled between revenue and covers performance with the buttons located just above the charts.

Key:

• Blue lines and bars - expected performance (custom targets set during onboarding).
• Green lines and bars - actual performance
• Black lines and bars - forecast performance. 

By hovering your mouse over any of the charts, the exact numbers of expected, forecast and actual performance are shown.

Forecast: Weekly View

At the top of the window, each of the coming 5 weeks is shown in a horizontal menu.

Selecting a week updates the display to show cumulative and day-by-day data for that individual week. Again, you can hover over any point on the charts to show the exact figures, and switch between covers and revenue.

Forecast: Individual Venue Services

To view forecasts for individual venues and services, select Forecast from the sub-menu. 

Each service is displayed showing Forecast vs Expectation, indicating how well the service is booked against predicted performance. The Expectation relates to the targets set by the user who completed the onboarding process.

These forecasts highlight services that may require operational, reservations, or marketing action. For example, the marketing team may adjust their promoted inventory campaigns in OpenTable, create targeted offers in their promotional channels or activate outreach to guests identified through the Single Guest Profile (additional Distil package required). 

In the example shown below, most services look like they will perform well, lunch and dinner services at the Gridiron Burger House look like they will perform well against Expectations, however Midtown Oven appears to need some assistance to improve bookings. 

Forecast: Monitoring Performance

When a service is identified that needs attention, the user can set up a Monitor to track its progress.

• Click ‘Create Monitor’ from the right-hand link next to the service to track
• The Monitor tracks changes in the forecast daily from activation
• All active Monitors are stored in the Monitor section, accessed from the vertical menu on the left-hand side of the screen.

This feature supports proactive, data-driven optimisation at every site, updated daily.