Tenzo's Forecasting Module was built from the experience of Christian, our CEO, who having run a chain of restaurants had experienced first hand the challenges involved in creating reliable forecasting sales.
Using a minimum of 90 days of historical sales data, Tenzo can create a forecast for your future sales which is 30-50% more accurate than traditional finger-in the air or four-week rolling average forecasting processes. What's more, Tenzo also empowers your teams to be able to view, update and approve forecasts on their mobile on the go using the Tenzo mobile app.
How does Tenzo Forecasting work?
Machine-learning allows Tenzo to take into account many more factors and data points when building a forecast than a human forecaster could.
- Step 1: Using your historical sales data (the more data the better, two years is optimal to start) Tenzo is able to build a baseline forecast that incorporates your overall growth trend, as well as daily, weekly, monthly, and seasonal patterns.
- Step 2: Tenzo uses its AI forecasting engine to understand how historical weather and events such as Christmas, Valentine’s day, and other holidays, have impacted your business and adjusts the forecast.
- Step 3: The final layer of data needed is the weather forecast and a detailed schedule of forthcoming events and with this, your Tenzo Baseline Sale Forecast is created.
- Step 4: General Managers (GMs) take the Tenzo generated forecast and adjust this to take into account all local events happening near their store, for example, roadworks or a local street party.
Tenzo stores the Submitted Forecast from the GM along with the Original Tenzo Forecast enabling the accuracy of both forecasts to be tracked.
Through the continual process of forecasts being submitted and the accuracy of the forecast being tracked, the Tenzo Forecasting Engine continually learns and develops a bespoke forecasting model for your location.
Maximising the impact of Tenzo's forecasting module
Customers get the best results using the Tenzo forecasting module where this is part of a weekly process with their teams.
Here's an example weekly process:
- Monday - GMs and the Operations Lead receive their weekly KPI email from Tenzo, including a review of their forecast verses actual sales for the last week and their labour costs.
- Tuesday - Managers use the Tenzo Forecasting Module to review the auto-generated Tenzo forecast for the next week and update this based on their location insights e.g. unusual weather as can be seen below.
- Wednesday - The Operations Lead receives an automated forecast sales email and reviews forecasts submitted with Managers.
- Thursday - Forecasts agreed and locked in for all locations. Using Tenzo's recommended labour hours, Managers can also complete their labour rotas for the following week.
- Friday - Ops Lead received an automated email with a confirmation of forecast sale and planned labour hours for each location for the following week.
Throughout the week we would also recommend setting up a daily run-rate email, so you can track your forecast and actual sales combined outlook for your current operating week. This will help you:
- Motivate and empower your team to take action to achieve their weekly sales targets if it looks like they might miss it; and
- Look at how accurate your forecasting was on an ongoing basis and use daily qualitative insights from your team logs to identify ways that forecasting could be further improved.
Tracking sales forecast accuracy
Tenzo calculates forecast accuracy using the mean absolute percentage error (MAPE). The smaller the MAPE the better the forecast. As we store both the original Tenzo forecast and the adjusted GM forecast you're able to see the MAPE for both.
What's more, you can also receive automated forecast accuracy emails to track the performance of your forecasting process with Tenzo and to help you to work to refine your internal forecasting process. Speak with your Customer Success Manager to help you design your weekly forecasting process and to set-up the automated emails you need.