Published: November 21, 2014 | Comments (2)
When I think about this week’s topic: “turning data into actionable data,” I find myself going back some years to a moment when, definitely one of the most formative bosses I have ever had, explained how to deliver value-added reporting to clients using the DIKIA reporting model. In fact, it was so long ago that a cursory search of the term on google unearthed both the lyrics of a Greek song and conjugations of an old Swedish verb, but alas no explanation of the reporting model. That’s maybe because – after further searching - I now suspect he made it up (it’s actually based on the DIKW pyramid: http://en.wikipedia.org/wiki/DIKW_Pyramid if you’re interested).
Anyway, the principle is the same and DIKIA as a reporting framework for contact centers DOES work, so out of loyalty to my ex-boss I’ve attempted to recreate it here:
Of course, the higher up the pyramid you go, the better the quality of the reporting. In a nutshell, at the highest level, ‘action’, you are able to take informed, decisive actions based on the intelligence at your disposal, which is built on the knowledge you have gained from information derived from data. IT MAKES PERFECT SENSE!
So then, why is it that many contact centers today are still blitzing their clients (if outsourced) or management teams (if in-house) with vast quantities of basic data, accompanied by, at best, a summary from the information in the report? STOP!
To bring this to life, I’m going to give an example of DIKIA. First of all, let’s take your customer service agents that take all those calls from your customers. By analyzing call statistics (data), you will get basic level information on what happened to the call (information), how long it took, whether it was transferred, for example. This information will give you some of your efficiency metrics, but it won’t help you know why these things are as they are. If you dig a little deeper, you’ll be able to find out why, for example, some calls are transferred and others aren’t (knowledge), but you still won’t be (well shouldn’t be) confident enough to change established procedures based on this level of insight. For this you need to pinpoint whether these occurrences are statistically robust, perform ‘what if’ analysis to understand the drivers of positive and negative business outcomes (intelligence), and, well, if you know what those are you’re in a great position to take actions that will unlock significant business benefits. You’ll understand the things your agents are saying and doing that will have the most positive impact on call outcomes. You’ll be able to customize performance development and even be able to match best available agent to each customer. Ubiquitous personalization! I love it.
Now, as people are often the biggest cost in any businesses, I wanted to share some advice on how to get to the top of the pyramid quickly and deliver actionable intelligence to get the most out of your workforce-related investments:
1. Know what skills and knowledge your people have
Benchmark your existing employee base: knowing what skills and knowledge your employees have is crucially important before analysis can be used to understand their links to positive business performance. There are many ways for employees to be benchmarked depending on their role and sector and could be based on performance data sources, such as sales figures, quality management scores, length of service. The key is that they are values which can be analyzed later and are common amongst groups that require comparison.
2. Have a clear goal
This is very important as it drives the direction of any analysis and helps formulate the desired business outcomes, and without clear goals or objectives it is difficult to achieve a real performance gain since you don't really know what you're looking for or want to achieve! Once you have determined your goal, analytics should be used to find the causal factors that are driving the desired business outcomes.
3. Pinpoint which skills and knowledge are linked to the best business outcomes
And then make sure you triangulate this with the organization’s goals and objectives - a fundamental step in the effective use of analytics. Continue by analyzing your best performers but, remember, 'best' should not be a subjective view. Use available data and analytics to inform and agree to the definition of 'best' that most accurately fits the organization, its goals, and its objectives.
4. Find and fill gaps
Use analytics to pinpoint skills gaps and unearth trends in skill improvements. The outputs of this analysis should then be used to make sensible decisions about what development is required: specifically what, who and when, rather than the common ‘one size fits all’ approach to training and development. Use analysis to show where investment has the most impact as this is your best bet to achieve rapid performance improvements at the lowest possible cost.
5. Test and refine
Keep on top of the effectiveness of your performance improvement measures. Continue cyclic measurement and analysis and make sure you continue to take relevant, focused actions to maintain positive performance improvements. Analytics should be a continuous investment, not a 'one hit wonder', as used over the longer term it will mean that companies continue to invest in the right areas based on measured, focused, repeatable proven patterns of improvement rather than perceived areas of need.
So, thank you to the creator of the DIKW model whoever you may be (even Wikipedia can’t get to the bottom of it) and to my old boss for adapting and sharing it with me, because it was a very early lesson – way before the era of ‘big’ data – that actions really do speak louder than data.