Flonomics has been around for quite a while, and when people ask about the value proposition around counting, we are frequently met with a, “so what?” And it’s true. For most of our users, visitor counts don’t exist in a vacuum. It’s a single metric of modest utility, but when combined with other metrics, as ratios, count becomes a lot more powerful.

Our recent release of Flonomics 2.0 expanded the capability of our analytics platform to import other data like sales, transaction counts, employee hours, and so on. Now we can calculate things like staff coverage (ratio of visitors:staff), conversion rate (ratio of transactions:visitors) and so on.

Well, once you have this kind of power, the next step is to begin to set goals. This summer, following the release of Flonomics 2.0, we undertook a series of experiments with a very supportive client, to understand the nature of goal setting in the retail environment. This was a customer with good revenue history (we just worked with one location to make it simpler) and a well established business model so that they were not making changes to product, pricing, promotion or place – it was a perfect opportunity for a controlled experiment.

Our objectives in the experiment (with the language and process massively influenced by The Lean Startup and the idea of the Minimum Viable Product (MVP)) were to identify a goal setting methodology, test different concepts for how to communicate goals and results, and determine if any of these activities in any combination had an effect on the outcome. The client is a specialty food shop, so there was no question that employee performance has an impact on the outcome.

In our initial cut, we pulled data into Excel from their Square POS, and organized it in a way that we could test different machine learning techniques for forecasting revenue. With years of data, we could analyze trend, cyclicality, seasonality, and also test the effect of other variables like price increases, weather, special events, and so on. It sounds fancy, but the truth was that after trying lots of different approaches, it became clear that entrusting the forecast to the retail manager was the best approach. They could look at last week’s number, and the weather, and be within a couple of percent of the likely outcome. So much for fancy math.

The far more important lesson, however, was that it is incredibly time consuming to pull the data, create the goals, capture the goals, and then communicate them to the team. The process doesn’t really scale because it should be bottom up. If a retail location want to engage in the well accepted business process of setting goals, they will need lots of support from disparate systems like POS, Time and Attendance, one or more systems that have marketing information, and yes, their visitor counts. For the SMB, this seems like an unlikely place to put the focus, no matter what the business coach says about goal setting.

In summary, we learned a lot in our two month MVP about forecasting revenue, communicating goals and results, and what the implications are for the retail executives and managers.

1. We really doubt anyone is manually setting intra-day revenue goals
2. Therefore we really doubt anyone is setting intra-day revenue goals
3. Therefore no one is reporting intra-day progress against revenue goals
4. Therefore SMB retail employees and managers may have a general idea of what is expected, but lack the data to make decisions about what to do specifically.
5. Therefore SMB retail executives also lack the data to make decisions about how to allocate investment across all the possible things they could do to make revenue goals (or any other kind of goal
6. Dunno, it kinda seems like an opportunity