There was a time when the ability to use a computer was limited to a select few, and when most organizations had just a handful of computers in each office. Fast forward a couple of decades, and digital literacy is now a prerequisite for just about every job.
As we move slowly but inexorably towards a future where data literacy is required know-how, how can organizations build a strong data culture that will stand the test of time? Here are three keys to a successful data culture.
If the data seems wrong, it might just be
Data-driven decision-making is a laudable goal that can give organizations the insights and clarity they need to innovate faster than their competitors. But should all decisions be based on data? What if the data isn’t good, and overenthusiastic employees attempt to extricate insights from data when there is none to be found?
A speaker at a recent event organized by CDOTrends spoke about this recently by cautioning against “forcing” data to fit a given narrative. After all, the data can hardly speak out to defend itself against an erroneous interpretation. But how can an organization tell when this happens?
According to this speaker, data analysts and CDOs should go with their “gut instinct” for the times when the data appears to be wrong. In his view, if it sounds wrong to employees with ample experience and deep expertise in the business, then it is probably time to exercise additional caution.
The computer engineering concept of “garbage in, garbage out” or GIGO is probably an apt parallel here. GIGO is the idea that the quality of output is determined by the quality of its inputs, and that feeding erroneous data to a software program will simply give you incorrect outputs. In a nutshell, give yourself the flexibility to disregard the data.
Ditch the top-down or bottom-up approach
Should culture change be top-down or driven from the grassroots? According to Keith Ferrazzi, the founder of a global consulting firm, successful culture change requires both pushing and pulling.
After all, a centralized, top-down model typically results in business units that are slow to take up new data initiatives, with executive sponsorship that might wane before things take off. On the other hand, a bottom-up or decentralized model might result in success stories that are dismissed as insignificant – or irrelevant – to other parts of the business.
Ferrazzi instead advocates both a push-and-pull strategy where a core team of data scientists and experts serve as internal consultants. But rather than hold CDOs accountable for showing financial results, he suggests putting the onus on business units to demonstrate financial results from the use of data analytics.
So how can organizations foster a culture to support this push-and-pull strategy? Ferrazzi suggests bringing data leaders from various business divisions into a central data analytics group while embedding data leaders from the central office in business units. The organization should also work on improving data literacy and promote push-pull collaborations by rotating assignments of data support teams.
Finally, to demonstrate the value delivered and to ensure accountability, an advisory council should be established to keep the CEO and board of directors updated on various data-centric initiatives.
Focus on generating a return with data
Finally, data analysis must demonstrate a clear return on investment to work. This means the business must find new ways to connect and use the data with the right marketing platforms or decision-making tools. Only then can the data be leveraged towards new marketing campaigns to realize new sales or other tangible business outcomes.
And as noted earlier, organizations need to start small to win big at data. But don’t just stay small, says Jeff Beck, chief growth officer at Leaf Home. In a contributed piece on Smart Business, he noted that rolling out changes to larger audiences can help businesses to quickly learn and “not spend time in an endless cycle of misleading results.”
Scaling up will require a clear and structured system for collecting feedback to help businesses learn and adapt at a faster pace. This means identifying KPIs and long-term goals and setting them up early to avoid being left scrambling when data starts to flood in.
Only when businesses make a concerted effort to apply data to technology can they unlock better marketing results to positively impact the bottom line.
Paul Mah is the editor of DSAITrends. A former system administrator, programmer, and IT lecturer, he enjoys writing both code and prose. You can reach him at [email protected].
Image credit: iStockphoto/ChakisAtelier
