Are you drowning in data? Do you feel overwhelmed – or underwhelmed – by the slew of options that claim to deal with your data problems? Particularly, in the area of pricing? It's the Era of Big Data, and many retailers are finding relief in new technologies like event stream processing or Hadoop.
In this post, I’m going to expand on the first of four steps to a successful pricing journey - data readiness. This step requires some determination, but it’s critical for building the best possible foundation.
Data readiness is all about preparation, and it will help you discover meaning and purpose at the heart of your data. I should also note, patience and hard work really do pay greater dividends here.
As you consider the first step in your pricing journey, keep the following opportunities in mind:
Why these four? They’re necessary for managing your data layers to produce data-driven insights, shape strategic decisions and ultimately, they prepare you for analytics and improved overall organizational adoption.
This step centers on the collection of your pricing history, and it’s the most tedious and laborious of the aforementioned opportunities. What are you collecting? How often? For how long? A rich and robust pricing history may sound like a no-brainer, but oftentimes it’s cumbersome to collect, store and retrieve as you begin your pricing journey.
Most will say that at least two years of history is necessary for ensuring accurate time-series and econometric forecasting to better understand trend, seasonality and life cycle effects. And that's true. But what's even more important is supplying promotional history, product attribution, competitive pricing, customer scoring and much more. And far too often, these reside in MS Excel spreadsheets on individual laptops across an organization.
The robustness of your data collection relates directly to your level of ease going forward. to your level of ease on the road ahead. Data insufficiency is the leading cause of failure in pricing projects, simply because the data is not "readily" available. When performing regression modeling to understand pricing effects, it's this data richness that will explain price sensitivity, serve as variables for lift effects, and be the causals to understanding why one model is a better fit than others. Companies that embrace this step wholeheartedly benefit from building enterprise data models that provide data access and readiness for more than just pricing, as the entire organization can benefit from this data.
Once collected, the investigation begins. What is your enterprise data model telling you? Are there competitive gaps? Are your customers responding in a new way? What data gaps prevent you from achieving success? Investigating your data profiling guarantees the enrichment process. Should you find the 80/20 rule applies to your data at the forefront, the price of storing it becomes staggering. I’ve witnessed shock and awe when a company embracing this step discovered that over 50 percent of their third party data capture was inaccurate. They knew they had a problem, but they realize its significance – and, they were paying for those records too! So, let your investigation begin and get your pricing data house in order!
Once you have the facts from your investigation, you're finally ready to explore your pricing data. Being able to see, understand and measure changes in pricing variance, for example, is crucial in this step. Is there enough variance so as to better understand your demand signal or observe elasticity effects?
There’s a greater discovery of new insights in this step; you’re discovering key attributes that drive pricing decisions, and models are also improved. Histograms, or the ability to plot trends for sales/inventory against pricing over time, and data mining techniques to better understand unstructured data, bring additional key insights.
In this phase, you’re also beginning your data-driven insights journal, which will quickly develop into your data library. They say knowledge is power, so go and explore to find new strength, sustained by your data
Collecting, profiling and exploring your data enables you to create a standardized and harmonized process. Creating a repeatable, sustainable data model ensures you can aggregate data at the right levels. Adding the right data layers reveals better, strategic pricing decisions and you can build new analytics hierarchies, embedded with ample attribution, while ensuring new data stores can be added to the process, easily.
Omni-channel engagement requires this kind of sustainable data model, as floods of data flow from your customers, manufacturers, multiple channels and more. Harmonization brings the data together in a way that is useable, insightful and purposefully meaningful to the end users. Whether they be a data scientist, buyer, pricing analyst or a C-suite executive, harmonization finalizes your data model, and will lead you to the next step in your pricing journey: pricing strategy!