The mindset for digital transformation

In another article, I cited that the digital transformation has to do with mindset, business models, value proposition and technology. All well balanced. Let′s follow an example.

Situation "A" – The traditional approach

A country hotel forecasts its occupancy rates. The demand planner uses time series models for the forecast. The 12-month historical data considers seasonal and cyclical factors. It also considers the day of the month, day of the week, holidays, etc...

• The accuracy of the occupancy rate forecast is 71%.
• The average occupancy rate of the rooms is 74%, but fluctuation between 45% and 92%.

 Situation "B" – The new mindset

The hotel – as a business – decides to understand the customer journey.

When do the customers decide to stay? By asking its customers, this hotel knows that the decision takes place, on average, 5 days before the check-in date. In other words, the demand in this hotel has a strong impulse element.

What factors do the customers consider just before booking the room? The same survey signalled that the weather forecast is a determinant factor. Specifically, three factors: temperature, presence of winds and possibility of rain.

Common sense leads us to believe that the highest occupancy rates will occur on sunny warmer days, with no presence of wind. And that, on the other hand, the lower occupancy rates occur on cold, rainy and windy days.

But the hotel opted to manage its data in a strategic way. Captured records of these three meteorological factors at the beginning of the week and compared with the occupancy rate data in the following weekend. They did it this for the last 50 weeks and used machine learning decision trees (artificial intelligence) to “see” the consumer behaviour.

The highest occupancy rates occurred when:

• Sunny and temperature not hot (nice or cold): 95%
• Cloudy and cold: 92%

The lowest occupancy rates happened when:

• Rainy with the wind: 40%
• Sunny and very hot: 55%

From this moment on, this hotel has been monitoring these three meteorological indicators. When the forecast combination for the next 5 to 7 days indicates the least advantageous conditions for occupancy rate, they automatically announce special offers and benefits. For example:

• Forecasted rainy and windy weekends: indoors activities at no extra cost (massages, games, live music shows)
• Forecasted sunny and very hot weekends: activities in heated areas, activities in the pools, activities on the lake, etc...

In contrast, the flexible tariff system can promote higher values on sunny and not very hot days, when of course the average occupancy rate is high (99%).

With this strategy, the minimum occupancy rate went from 41% to 55%. The average occupation went from 74% to 83%. The revenue was 6% and the profit grew 4%.

Summary

The change of mindset began with the understanding of the consumer's digital behaviour, who consult the weather forecast via mobile a few days before the scheduled date for the check-in. It follows with the processing of data with machine learning techniques (no more traditional time series approach). It ends with a relationship strategy capable of offering a value proposition for each (meteorological) situation.

What matters: Improvement in the occupancy rate forecast (that allows for better asset utilization, including people management and procurement), improved occupancy rate and increased revenue and profitability.