Commentary

A beginner's guide to analyzing restaurant data

Dave Bennett, of Mirus, offers simplified steps on how to analyze restaurant data provided by Mirus.

July 23, 2018 by Dave Bennett — President & CEO, Mirus Restaurant Solutions

Not long ago, things seemed easier. When you licensed a system, it included a set of reports that fed back to you the data that had been entered throughout the day or period. Back then you only had a few systems, and each had a good library of reports — so all was good. That is, until you needed data from all of them combined on one report, like a scorecard. Then it was cut and paste, cut and paste, etc. This process is time-consuming, error prone, but you probably already know that.

The other big change in the past decade or so is the explosion in the number of systems your company uses every day. We attempted to provide a list of all these systems, but no list can stay complete for long because it seems new ones are hitting the market each month. Today, it is likely you have eight, ten or more systems deployed, each of them collecting lots of valuable information.

The problem is that none of these systems have the job of pulling all the data together into one place where you can easily mix and match, and slice and dice the data. For years there have been Above Store Reporting solutions that typically worked with POS and some back of house data, but most of them fell short when challenged with reservation data, or customer feedback data. A new breed of solution was introduced at the beginning of this century that takes a different approach to analyzing all the data you are collecting, the Enterprise Data Warehouse.

This technology, originally developed in the eighties and nineties, can be used to provide a platform for combining data across multiple systems.

Which leads you to an inevitable question — do I build my own EDW using one of several generic EDW tools, or subscribe into one that is already running — like plugging into an electrical socket? The answer may surprise you. Do-it-yourself is an option for lots of software, but for your EDW there are some unique issues with DIY you should be aware of.

First, the skills needed to build an EDW are not common, so you will be challenged with finding someone with the track record of success you need. Second, the design of your EDW must be created; there are no blueprints for generic EDW like you can get when building your house. You have to hire an architect before you start construction, and it is crucial that the architect know the restaurant business intimately. Finally, your data warehouse is a living system that requires ongoing maintenance to remain accurate and complete. This is like putting additions on your home. If not done well, it may compromise the value of the entire house (data warehouse).

The heart of any EDW is the continual process of putting data into it, and that is known as Extraction, Transformation and Loading, or ETL. We have a series of blogs dealing with the importance of each of these steps. Extraction is the start of the data collection process, and its job is to get the data out of the source system and centralize it all. Transformation is a critical step that maps the data coming in to the data model designed by the architect. Finally, Loading is the step that takes the data from the staging area to the production EDW where everyone can see it.

While this is a simplified guide to analyzing restaurant data, this begins the roadmap to better understanding what is really happening within the four walls of a restaurant and within your company.

Business Strategy and Profitability