The six-point guide to data-driven transformation.
by Jeff Rajeck of Econsultancy
Nearly every business now has a company-wide program to help it capitalize on recent advances in digital technology.
Commonly known as ‘digital transformation‘, this process could (and perhaps should) be thought of as ‘data-driven transformation’ considering how important data is to the whole process.
Reason being, unless data is captured, shared and utilized, any increased adoption of digital technology will be in vain. A competitor with the better data will most likely win in the end.
With this in mind, how can companies transform both digitally and with data? What are the issues that need to be considered?
To find out, Econsultancy recently invited dozens of marketers to discuss this and other topics over roundtable discussions. At a table hosted by data experts David Brigham, Analytics Director, Mirum, and Zak Agus, Sales Director, SEA, Tealium, brand marketers revealed their thoughts on what was driving data-driven transformation at their organizations. The main points from the discussions are summarized below.
The six-point guide to data-driven transformation.
Before we start, though, we’d like to let you know about upcoming training which may help you with your data-driven transformation. Econsultancy is offering an Advanced Mastering Analytics course on April 8th, 2018 in Singapore. Click here for more information and to book your spot.
So what do organizations need to consider when driving data-driven transformation?
Breaking down data silos
The first thing participants asserted was that there is no ‘one size fits all’ approach to getting data flowing through an organization. Every organization is different and so each data-driven transformation project needs to take the organizational structure into consideration first.
Yet despite these differences, nearly everyone said that departmental data silos are perhaps the biggest barrier for companies aiming to harness the power of data.
In order to get over this hurdle, participants argued, the transformation team must examine all of the company’s data assets and determine which department has ownership and who already uses the data. Then, they should ensure that transformation has buy-in at the highest level so that management cannot unreasonably stand in the way of future data requests.
Breaking down insight silos
Data silos, however, are not the only problem faced by data-transformation teams. Another issue which attendees brought up is that companies also have ‘insight silos’.
An insight silo occurs when one department has data analysis expertise which is lacking in other areas of the organization, and is unwilling or unable to share.
According to attendees, this situation occurs quite often at airlines. Airlines have large teams of pricing analysts and separate teams of marketing analysts who work independently of each other, with little sharing of insights.
So, for organizations to benefit from their talent, the data-transformation team should also identify where the analyst talent sits and find ways to get the teams to collaborate.
Yet despite these differences, nearly everyone said that departmental data silos are perhaps the biggest barrier for companies aiming to harness the power of data.Jeff Rajeck, APAC Research Analyst for Econsultancy
Finding industry ‘data pools’
For data that does not exist within the organization, the transformation team will have to look elsewhere.
Existing solutions, such as data management platforms (DMPs), help but often they are expensive and may actually offer too much data.
A new alternative to DMPs, according to participants, is for companies in the same vertical to share data between each other so that everyone benefits from having access to relevant and relatively inexpensive data.
Named ‘data pools’ by our subject matter experts, these new ways of obtaining data inexpensively should also be researched by the data transformation team.
Resourcing the transformation
In addition to identifying the talent already present within the organization, participants indicated that data-driven transformation often requires people with new skills, such as data scientists.
The six-point guide to data-driven transformation.
As finding the right people for these roles is often time-consuming and difficult, one suggestion was that an organization going through transformation should first hire a ‘data guru’ who will be responsible for upskilling existing staff. In this way, expertise can be built-up in-house at the same time new talent is being recruited.
Proving data-driven ROI
Proving return on investment (ROI) is now expected in most marketing departments, but it is a relatively new topic for analysts.
The six-point guide to data-driven transformation.
For a small project costing a few thousand dollars and lasting 3 months, ROI may not be a big issue but for a long-term data transformation costing a few million dollars, ROI should certainly be a consideration.
To manage this requirement for ROI, data-driven transformation teams should be prepared to defend their investments in technology and resources. The team will need to demonstrate how their approach will either increase revenue or decrease costs, something which most analysts have not yet given much thought, said one participant.
Communicating the limits of data-driven transformation
Finally, attendees said that the transformation team should set departmental and management expectations about the potential and the limits of data-driven transformation.
For example, a marketing department who wants to invest in personalization data must first understand the risks of becoming too intrusive through excessive use of personal data.
The six-point guide to data-driven transformation.
Additionally, there will come a point when additional data and analytics will only achieve incremental results, and business heads must be made aware when that point is reached. As one participant put it, “you can’t continually invest in data and expect to get the same results every time.”
So, while teams are talking up the potential of a data-driven transformation they must also be careful not to over promise and subsequently under deliver.
