I have always believed in the old adage that, “if you cannot measure it, you cannot manage it”.
I have spent my career defining metrics to assess and drive performance improvement in business. But before developing a new program, I always ask myself the same questions:
- Who owns the metrics?
- How many metrics do you need?
- Do they reflect performance?
- Are they complementary or conflicting?
- Do they drive action?
- Can we afford to measure?
- Can we afford not to?
The reality is a metric horse for assessment course. So, what makes a good measure? I use my own 5-Cs to build the basic framework:
- Comprehensive, yet simple
- Correlated to business performance
- Connected to the system, with diagnostics that drive process improvement
- Compensation-based for success through total reward plans
To explore this challenge, I want to consider retail product availability – that is, getting products on the shelf at the right time for customers to buy. Many of our retail clients face this metrics dilemma – striking the optimal balance between the creation of a comprehensive governing metric for assessment and action.
Who owns, and would benefit from, an availability metric?
This is not a straightforward consideration. The measure is broad and its impact spreads across multiple functions, – from Store Operations through Merchandising and Supply Chain to the relatively nascent Online Operations departments. Aligning these departments is difficult and can create confusion and political divide across an organization if you do not manage it effectively. In my experience, you must define one owner for a system or initiative to succeed.
How many ways can you measure product availability?
You need a metric you can apply across multiple store formats, banners, and items. Once you have that, the method of measurement ranges from simple item gap-scan by store associates and sophisticated near-real-time store level data-driven models to computer vision technologies.
- Weekly gap scan score – what the associates can see as out-of-stock on the shelf
- Online first-time pick rate – personal shopper pick performance
- Image capture – video and photo of what’s available either continuously or at set times of the day
- Sales, demand, and inventory data-based score – item ranged, in-stock, and selling as expected
How do you make the right choice?
There is nothing to stop you from using more than one model to measure on-shelf availability. But simplicity is key, so I evaluate the possible choices against my 5-Cs:
1. Weekly gap scans
Weekly gap scans are simple, but they are expensive. They capture only a small sample of items and time, and do not reflect overall performance. That is to say nothing about stores, in effect, marking their own homework with this model – particularly when they can influence reward – result, 1-C.
2. First-time pick rate
Online first-time pick rates are a strong output metric, which you can include in your reward plan. But I really want to influence the inputs that drive this metric. It can be cost-effective, but typically only covers a limited number of items (albeit this has increased in the last year) and only for stores that offer online pick. The fact that most retailers pick very early morning also assesses at a time when the store should be set up for success – 3-Cs
3. Image capture
Image capture can mimic the human eye to capture gaps, but comes at significant cost to install, power, maintain, train the detection models, manage physical store layout changes, and item-set changes over time. While detecting gaps, they can miss obvious shelf plugs (no gap, but incorrect item in location) and does not capture the value of each gap, what to fix first, or the diagnostic causes of a gap. A strong additional signal all the same – 3-Cs
4. Sales, demand, and inventory
You can build sales, demand, and inventory data-based scores from data held by any modern retailer at a relatively modest cost. A well-constructed store reflects in-day trading dynamics and the reality of the shopper experience. It also powers systems and processes of corrective action on labor scheduling, in-stocks, inventory records, and shelf capacities. It correlates well with overall business performance, according to researchers. And, with the returns it can generate, makes complete sense to be part of the formal reward structure – 5-Cs
Product availability in multi-channel environments
Product availability is a critical consideration for any retailer that wants to succeed, in today’s multi-channel food and grocery environment. Admittedly, we have a dog in this fight, but we also have a healthy level of dissatisfaction in what we do. Our objective 5-C evaluation ensures we stay focused on delivering the most relevant metric that not only measures what you are doing, but also directs the business on what you should do to improve overall performance.