- Report finds that only 15% are happy at work, here’s 8 ways to be happier in yours
- We fell out of love with Holden because it fell out of love with us
- The pointlessness of social media’s partisanship
- Science thinks you’ve already peaked, sorry
- Sacking volunteer firefighters, limiting free speech: Who the bloody hell are we?
Unless it’s carefully navigated, the burgeoning data landscape has the potential to become the new Wild West.
Business intelligence is designed to inform strategies for competitive advantage, and that relies on access to the best, most relevant information and the ability to properly interpret this data to develop actionable insights.
This hasn’t changed since Business Intelligence was first defined by Hans Peter Luhn in 1958. What has changed is digitisation driving business change at every level and it means we have more data than ever (sales data, loyalty card data, media consumption data, tagged up digital journeys, search data) all passively collected across a wide range of data management platforms at scale and at speed. Advancements in the fields of neuroscience, consumer psychology and behavioural economics also mean we can better understand and quantify the impact of, for example, brain activity and “system 1 thinking” (that is, fast, automatic and subconscious thinking) on decision-making.
But what’s our goal? Having lots of data – big, small, digital, direct from the brain – does not necessarily translate to business intelligence or help a business become customer centric. It does not tell us how to grow.
What enhances business intelligence is understanding the strengths of different data sets, how they complement each other, and how to fuse them to get the smartest, best data and insights to inform our decision making. The further opportunity is speed. Not real-time, every time – that can lead to overload, but at a faster pace that fits with the speed of a business and their operating rhythm.
So firstly how do we bring together this explosion of data for competitive advantage?
Passive vs active data
We can think of data in two broad buckets: “passively collected” and “actively collected”. “Passive data” includes the data mentioned above; your digital data, stuff that is a bi-product of behaviour. “Active data” is that data we specifically collect, including surveys, consumer panels and qualitative insights. It is necessary to understand the strengths, weaknesses and limitations of both types of data.
Typically passive data can help us understand many consumer behaviours such as sales, loyalty and repertoires. It can also be great for understanding the impact of marketing actions such as communications exposure (reach, frequency, depth and duration). But as much as this can tell us, and as much as it can help reduce the load of actively asking people for information, there will always be a need for direct dialogue or conversations with consumers if we want to really understand the “why” behind the “what”; that’s when active data is needed.
Currently the other use of active data is to help us “join the dots” where there are gaps in passive data; for example non-customer views, competitor understanding and cross-media exposure. Concise, focused and engaging studies are the key to getting great data through active collection methods. This type of data collection needs to be contextually thoughtful and help us validate real world outcomes.
360 business intelligence
So the data is all out there, but it becomes truly powerful when we can connect it together for one worldview. The question is “how?”
Enter the nerds – or “data scientists” as their business cards might read. Analytics can help us join the dots between data sets. Market mix modelling (MMM) seems old hat these days, but fundamentally it brings together data sets to understand business outcomes and help predict future scenarios. Attribution modelling, long and short term ROI modelling, decision tree modelling and choice modelling are all great weapons in our current armoury. These different approaches operate with both passive and active data – but where next? Analytics is becoming increasingly important in answering all sorts of business issues, and computer learning and artificial intelligence can potentially deliver a quantum leap in capabilities. Imagination may be the only barrier, once data availability, ownership, integrity and transparency are givens.
We once referred to the importance of robust “single source data”. As ambition and expectation have grown, it is now more of a case of “single journey data”. We combine data from one source that tells us what individual people think (from surveys), what they are exposed to (from media consumption data), and what they do as a result (from sales or other behavioural data). The picture no longer needs to be cropped; fusing with broader data sets such as holistic financial transactions, or broader online behaviour through analytics, can help us understand an ever-broader set of potential influences on that journey or ways in which to profile and segment the consumer.
That sounds like a utopian dream for business intelligence – an end-to-end view of consumers that gives us the ability to measure, test and predict business outcomes, supplemented by a bunch of other data we can incorporate to build an ever richer view. This is now a reality – but perhaps a new reality and a new battlefield for researchers.
What does this mean for research?
Will the next five years see us replaced by an army of data scientists and their AI robots? No. But if you don’t like change, it’s time to find another income stream.
For all the talk of the “rise of the nerds” in research, marketing and business in general, newer skills will always need to be underpinned by the core values of a rounded researcher who knows the craft of strategic marketing. There is greater need than ever for those who can:
- Guide clients and colleagues through this increasingly complex data-rich landscape
- Understand the complexities at play; the strengths, weaknesses and relationships between different data sets and how they can be structured to deliver learning frameworks
- Determine the strategic importance and relevance of different hypotheses and how they are best tested.
Perhaps, most of all, there is a need for people who are not solely driven by data and maths, but understand its value and can bring a real worldview to problems.These skills will always be fundamental to the way we get to the best business intelligence.
What does this mean for clients?
For clients it means new skill sets and capabilities will be needed, both in-house and from external partnerships. In-house, clients will need to consider:
- How will you manage your data to ensure you get the best data for that 360 view?
- What do you retain in-house versus outsource?
- How do you store your own data in a way you can readily integrate and interrogate (and afford)?
- How do you future proof from an ownership and transferability point of view?
In addition, clients will want to pick partners wisely by considering:
- Whether they are prepared for the future with the right tools, skills and mindsets?
- Do they understand what it will mean for your business and how you can take advantage to drive business growth?
- Do they have the integrity to help guide you through a data landscape that has the potential to become the new Wild West?
If you’re embracing the change we’re already seeing and it excites you as much as it excites me, the next five years are likely to be the most fulfilling and transformational times both you and the market research industry have ever seen.
Good luck and enjoy the ride!