This post concludes our two-part series, on the questions that insight leaders raised, at DataIQ Discussion 2017 event.
I hope you found part one, helpful. There we focussed on topics including:
- measuring your impact
- breaking through cultural barriers
- best organisational design
- tools needed by insight teams
- executing your insights (driving action as a result)
Those certainly were hotly debated topics, that appeared relevant to everyone around the table. All those topics were raised by first two tables of insight leaders.
Now, I will conclude this series by sharing two more questions raised by second group of Customer Insight leaders, as well as the three questions from my final group.
It was interesting to see the consistency of topics of interest around these tables, so I hope that means this content feels relevant for you too.
Back to those tables of insight leaders. Their next question was…
Are you using Social Listening & Voice Analytics software?
Broadly, the answer was, yes. However, there were several leaders who expressed that their social listening was mainly limited to being a qualitative research source, or for customer service teams. Few around the table were seeing the ‘Big Data‘ benefits of integrating social data to enable new profitable actions.
One, more sophisticated, company was supplementing social data with network analysis – in order to identify influencers for targeting.
As we’ve suggested before on this blog, more ‘Big Data benefit‘ was being realised by use of wider internal datasets. A number of leaders are using Voice or Text Analytics, on call centre data, to guide further insight work. One delegate has a ‘big screen’ dashboard for all staff to see, featuring: live Twitter feed, social listening brand tracker & key themes from searches on their own website. A nice example of data visibility internally.
How can you make life easier for your customer insight analysts?
The goal here, or course, was to free up your analysts to be more effective – not just put their feet up in some ‘chill-out room‘. Most leaders around the table were frustrated with how much their analysts were bogged down with data problems or not used to best advantage by internal customers.
Given that, a couple of leaders identified the need to improve the questioning skills of their analysts, to ensure they are working on the real business problems that matter most.
A theme also emerged, around developing analysts to consider the “whole value chain of data”. This phrase meant encouraging analysts to build relationships with both consumers of insight (e.g. marketing teams) & the providers of raw materials (e.g. data teams). It was acknowledged that analysts know why data items matter. For that reason, they should also help data teams make the business cases for improving data quality or filling data gaps.
As well as this (and wider encouragement for analysts to know more about data ownership & governance issues), another tip was to use graph databases. These were found to be helpful in handling the complexity of different business teams defining customers differently & needing different views to answer different types of questions.
How do you define Customer Insight?
This, first question, of the final table of insight leaders, built on earlier questions around definitions. This time the focus was more forensic, picking apart both how others defined customer & how others define insight.
Organisations with multiple uses of the word customer struggled the most with this clack of clarity. Different types of ‘customer’ included the ‘end-customer’ (consumer of service/product), distribution and/or service intermediaries (sometimes multiple layers), B2B customers and some variants of ‘internal customers’. Most agreed the simplest solution was to define end-customers as the ‘customer’ and find other terms for other parties.
There were a number of definitions for ‘insight’:
- Some included only analysis or only research (I was pleased to see that more leaders included both)
- Several excluded Predictive Analytics or Data Science (with the interesting perspective that “modelling follows a script, but insight is more like jazz”)
- A number of allusions to the definition of insight as ‘actionable’
- Others put this as “Needs to answer a business question” or “To be valuable to the business”
- All agreed that job titles can be confusing (it is more important to know what you are there to achieve & technical skills at your disposal)
- Reference was also made to emotional insights & getting the ‘why’ to explain behaviour identified through analytics
- All agreed it is more than BI & reporting (or even just analytics)
Another interesting point, raised during this conversation, was the need to define what is a helpful level of granularity. What is relevant when seeking insights on customers (e.g. some interested in an individual, but a number more interested in segments, or level of ability to action)?
What is Best Practice? What has given you most ‘bang for buck’?
These questions prompted an interesting, roundtable, sharing of what had worked best for each leader. Some interesting case studies were shared, each slightly different from each other & the previous ‘biggest bang for your buck’ analytics advice I’ve shared.
Some of the examples, insight leaders today shared, were:
- Value-based segmentation (after a history of not getting value from an attitudinal segmentation, that could not be overlaid onto database)
- A Recency/Frequency/Monetary-value (RFM) segmentation from data
- Improving Customer Retention (with a suite of different predictive models for exit at different stages for different reasons)
- An Optimisation engine for a Next Best Action deployment, that was avoiding the frequent challenge of being seen as a ‘black box’, it did this by allowing business to visually control priorities for optimisation
- Another delegate sang the praises of continuous improvement instead (lots of small incremental improvements – like British Cycling), in aggregate driving a big change in terms of value-add
- Other examples included more transparent marketing effectiveness measurement & enabling C-Suite to see for the first time what their customers are actually like (overcoming inaccurate stereotypes)
What are the cultural barriers that get in the way of making a difference?
This, final question of the day, partly revisited the discussion in part one on cultural barriers. However, this time the focus was more on what the customer insight leader & their team can do about it.
Barriers identified included:
- Stereotypical thinking (data & analytics viewed as just for Finance or ‘nerds’) – leaders need to demonstrate relevance
- Egos tied into previous decisions (e.g. sunk cost on a segmentation that doesn’t work) – sometimes you need to work with the imperfect until you can pilot value of improving
- Analysts not really clear on the real business need (just what they have been asked for) – another reminder of the need to improve questioning skills & wider softer skills for analysts
- To move insight team beyond being seen as a ‘report generation team’ – part of the challenge for leader to build the right team reputation whilst actively managing stakeholders
So, interestingly, most of these ‘barriers‘ were actually perceptual or educational challenges for leaders to influence. Another reason that customer insight leaders need strong leadership skills too.
Thoughts on hearing questions that insight leaders shared
Hope those memories proved helpful & perhaps inspired some relevant ideas for your own role & challenges.
In the spirit of the debate, that produced these questions & answers, if you disagree or have other examples to share – please do. You can use our comment feature below.
Have a productive 2017 in your customer insight leadership role. This event certainly reminded me how important such a role is to the growth of businesses today.