Restaurant Group: Increase Profitability with Mission Segmentation

Through Outra delivering a Custom Mission Segmentation, this restaurant chain was able to inform a hyper-local communications, promotions and pricing strategy.

Situation

The casual dining category is defined by a broad spread of customer types, a wide choice of competition and the widespread application of discounting and offers. Research by Deloitte in 2017 suggested that 38% of visits across the category involve some kind of “deal”, especially during the mid-week and lunchtime periods. This restaurant group understand they are in a vibrant, competitive category where customer understanding, appropriate pricing and excellent experiences are key to success.

Challenge

Outra was approached to help this company understand diner behaviour. The objective was to gain insights into all customers, not just those for whom they had personal details in an email CRM programme. This way we could drive real insights at a mission level that could be used in promotional offers, communications, experience and inform pricing decisions.

Solution

Outra had to work from the broadest set of data so we started by analysing 18 months of EPOS data. The multiple files of data were ingested into our analysis platform, Velocita, and matched with our proprietary data spine allowing us to understand the outlet, the competition and its catchment. We trained two years of data, with seasonality, bank holidays, car parks, changing offer partners and widely ranging local markets built in.

From this we developed a Mission Segmentation using deep learning to understand diner behaviour. We then built a model that automates decision making for 14 sessions a week across 157 restaurants, the optimal offers, cover turnover, spend per head, volume of customers and mix of order.

We chose to prove the approach across six restaurants prior to rolling out across all 140. The transaction level data included line level data across every bill giving us a highly granular view of both individual and ‘table’ level data.

Results

Two key aspects characterised the success of this project. The first was the speed of data manipulation and ingestion and the second was how the Mission Segmentation uncovered previously unknown patterns of diner behaviour leading to very different levels of return.

The restaurant chain has been able to show how they can deliver insight and actions tailored to each of the 140 outlets, driving differentiated communications and hyper-local activity, resulting in a high level of customer engagement and satisfaction, in addition to being able to service their customers more cost effectively.