In the previous blog, I looked at making the business case to exploit the value of data in your organisation. This naturally leads us to seek an understanding of where the value is, which is the subject of this blog and a cornerstone of any data strategy.
There are two types of value we can get from Data. Tangible value comes from increasing revenue and reducing costs, both combine to increase the margin of the business. The Intangible value comes from reducing risk and from improving productivity and automation. I have created, over the past 10 years or so, a mind-map of circa 120 business challenges where data can help generate business value. Typically, when you take the right data, combine it with operational activities, using the right tools, and data technique you can generate value. These are the key ingredients of your data strategy.
There aren’t that many types of tools you can use to exploit data – typically dashboards, data feeds / APIs, data science models, artificial intelligence, reports and alerts. Within each tool there are techniques used by a data consultant such as using a dashboard to rank customers by profitability, identify the least profitable customers, and then the operations team should put in place a process to move them to self-serve.
So let's look at an example of where the value comes from. Typically, an organisation would be divided one or more P&Ls and some enabling functions such as Finance, Human resources, Legal and Risk, Procurement and Information technology. It’s probably fair to say the most organisations are exploiting data in these silos today. For example, a customer CRM system will enable sales team and marketing teams to analyse and target customers. However, the real revenue opportunities come from the exploitation of customer data combined with products and service data. This data will give you the opportunity to identify how you can sell more of the same product to existing customers, how you can sell new products to customers, and generate new customers. The more datasets you combine the more insightful and valuable the data becomes. For example, when you combine customer data with finance data you can build a picture of the profitability of customers, and when you combine customer data the product data you can identify cross sell opportunities and potential ideas and new products.
So, whilst we understand a lot of value is obtained by combining multiple datasets this introduces new challenges. The biggest of these is the inconsistent definition of items such as customer ID, products, locations, and currencies, which are critical for combining data sets across the organisation and for cross functional reporting. This is where data governance helps by creating consistent definitions of business terms and common fields required to combine the datasets. The second biggest challenge occurs when you bring together multiple datasets because you are likely to start using data from another dataset in a way it was never intended to be used. This generally manifests itself as incomplete, inconsistent data or potentially duplicate records (e.g. duplicate client IDs). In my experience, a good data strategy will bring all these elements together.
Those people in the organisation who know what the critical business challenges are need help and you should prioritise supporting them with the data they need. However, the reality is that you cannot expect everyone in the business functions to know how to get the most value from data.
So, a good data consultant should supplement the knowledge of key people in the business by explaining the best practice of how other organisations have successfully got value from their data will get you the best outcomes. That’s why I have my mind map – it creates some great discussions and some unexpected opportunities.
In the next blog, I’ll start looking at “How do you extract the value from data?”
Chief Data Officer
If you need help understanding how to get the most value out of your data, get in touch with our experts by email or give us a call. We’re happy to help!