Maximising Value and ROI from Data

By Harry Cruickshank

If you think about it, data is a lot like food. There’s a huge choice. Some of it is good and much of it is mediocre. Your enjoyment of it, and the value you assign to it, can be impacted by many factors – how well it has been sourced, the quality of the various elements, the care with which it is handled and prepared, the ultimate presentation to you and how it makes you feel after you’ve consumed it. OK, enough with the analogy.

We’re informed incessantly that the future will be “data-driven” and that businesses lacking data will be at a competitive disadvantage. Whilst that may be broadly true, for an organisation to derive value from data there are many obstacles to be overcome for the data to be insightful, actionable and used for decision making. In a 2021 report by Alation¹ 97% of data/analytics leaders said their company had ignored data when making decisions and this had led to negative consequences.

What c-suite teams want is not raw data, nor fancy dashboards, nor complex reports. They want valid insight derived from data analysis that can be translated easily into action – action that will positively impact their target KPIs and help them achieve their objectives. That is what will underpin a decent ROI. Anything less won’t cut it.

As an example, the CEO of an £800m company recently said to us: “I have spent a lot of money on researching all the customers and employees I have in many countries around Europe using NPS and do not know what value I got.”

There first hurdle that gets in the way of insightful data is relevance. Gathering data is usually the starting point to derive insights that will direct necessary business decisions within an organisation. And yet how often do we see mountains of data collected that measure or monitor the wrong or largely irrelevant things? In other words, are you focussing on gathering the right data, rather than the data you think you should have or have habitually been collecting?

The next hurdle is acquisition, i.e. where are you getting your data from and is it reliable? If there are issues here, they degrade the quality and value of everything that comes after. Integrity is key. According to a Dimensional Research² survey of data professionals, 90% said numerous data sources were unreliable over the previous 12 months and 86% said they used data that was out of date.

Of course it depends on the nature of the data. A simple transaction list from an EPOS server is likely to be accurate. But what about a customer or employee survey? The quality of that data will be strongly influenced by what information is being sought (relevance), the appropriateness of the place from which the data is extracted and the rigour of the methodology used to collect it, including any potential risks of bias or gaming the system.

Even if the data is relevant and its acquisition is solid, the handling and analysis of that data can add further kinks in the wire. One common problem is seeking so much simplicity that you dumb down the good stuff. At the other end of the scale, too much information means companies fall victim to the “fire hose syndrome”. Without effective analysis from people who have the necessary skills and a clear understanding of the end users’ needs, which enables them to extract relevant and useful insight, we’re back to square one.

This is one reason why businesses are investing millions to improve their data architecture and training up employees to be more “data fluent”. Data analysts and scientists are in high demand and the war for talent in this area has been intense.

In the end, the true value of data is the ultimate effect it has on strategic and operational decision-making. The gorilla in the room here is trust. Unless senior executives have confidence in the quality and consistency of the information they’re presented with, they won’t rely on it. CEOs, CFOs and COOs accept a level of complexity and ambiguity in their data and insights, but if their trust in the data is not sufficient they’ll default to making decisions based on gut instinct, historical experience or political & budgetary debate, which may bring greater risks.

There’s a cost associated with acquiring high-quality data, processing it effectively, presenting it well and ensuring that valid conclusions are drawn and acted on. That cost can be presented as a good investment with a strong ROI. The alternative, where raw data is flawed and its ultimate value consequently compromised, will prove more expensive and degrade the ability of any organisation to improve performance and grow sustainably.

In the world of data, the gap between intent and reality is wide and deep.

So here are a few guidelines:

• Make sure you look for the right information and collect it from the right place. For example, survey questions in customer or employee surveys that are created by the management of the company seeking feedback are of low value, because of the bias and possible irrelevance built in. However, that same survey focused on areas that are most important to the customers or employees is of higher value.

• If your data collection method is not intentionally diagnostic, then it is of low value. C-suite executives need insights that tell them, reliably and without bias, where and why they need to focus efforts and investment. Make sure your data goes beyond simple ‘stake in the ground’ datapoints & scores. It must be able to answer a “so what?” question.

• If the insights you gather are about your company alone, you may be missing a trick, even if you partake in common leader boards or market comparisons. It is often the case that your customers and employees can tell you a great deal about your competitors too. If you don’t ask them you won’t get that insight.

¹ “State Of Data Culture Report”. Alation. September 2021.
² “Data Analysts: A Critical, Underutilized Resource”. Dimensional Research. June 2020.