What management doesn’t get about
their tech teams - Part 2

It’s a strange paradox that though every company relies heavily on technology nowadays, most organizations fail their digital transformations. Is the technology to blame? Many analysts agree that the secret to success is in the right team, strategy, agility, and mindset. 

When things go wrong, an organization’s leadership can be blamed for lots of things: missed deadlines, miscommunication, lack of support, and micro-management. But could tech teams blame the C-Suite for too much?

“Executives have long resisted data analytics for higher-level decision-making, and have always preferred to rely on gut-level decision-making based on field experience,” wrote Harvard Business Review (HBR) experts as recently as March 2022. 

Earlier, in a 2019 survey, Deloitte researchers discovered that “67% of those surveyed (who are senior managers or higher) say they are not comfortable accessing or using data from their tools and resources.” 

It seems, then, as time passes the distrust of data stays strong. Yet, in the same HBR article, its authors suggest that once the C-Suite starts seeing reliable and consistent data insights and recommendations, this distrust will evaporate. 

And what’s their recommendation for the tech teams? Create reliable models by using good quality data (both structured and unstructured), fed in real time. Using the right tools can be helpful too.

“Take an extra hour or two up front to define business terms, demonstrate where you find the data in source systems, and what actions you can take using the data,” said Shayde Christian, chief data officer at Cloudera. “Ask management to define the common analytics themes they will be requesting of you, so you can assemble a robust dataset in advance of future requests; the Cloudera Data Platform (CDP) is an excellent platform on which to construct that.”

Overall, it’s not hard to convince business leaders: Executives do understand that data-driven insights are essential for business growth, as they can leverage the data they have to create solutions, improve customer experience, and respond quickly to changing market conditions and customer expectations. 

For the Cloudera Enterprise Data Maturity report, researchers have explored the correlation between the maturity of an organization’s enterprise data strategy and its business performance. Nearly 92% of surveyed IT decision-makers in telecommunications agreed that a modern data architecture will help them gain a competitive advantage. Likewise, in manufacturing, 65% of survey respondents already use data and analytics to improve the customer experience, 60% — to reduce risks, and 58% — to gain competitive advantage.

Yes, most executives do not understand the nuances of building a machine learning (ML) model, tech jargon, or programming in Haskell. They are focused on financial success, returns on investment, avoiding risks, and strategic priorities. So they may talk about the rule of 40, or annual recurring revenue, and other financial-oriented metrics that may be just as lost on the engineering teams as the ML talk is lost on many execs. 

It gets even more complicated in teams working with data science. “There are two sides to every story, but three sides to every data story,” Christian said. “Management understands ground truth, from which they generate an observable reality, while data practitioners believe in what the data says.

“Management might not believe the data because the numbers differ from their observable reality, while data practitioners don’t trust ‘management's anecdotal evidence’ over the data.”

The third side of the story? The data! “Phantom data, which neither management nor practitioners can see, is that missing data that if found would engender trust between management and tech,” Christian explained.

As a team, it’s important to understand each other — and drive success in a mutually beneficial way. It’s also important to understand each other’s point of reference. In data science — it means digging through the phantom data, uncovering all sources of data, and arriving at comprehensive and reliable data insights, together.

“Management must have or develop the business processes that generate the data required for analytics to be meaningful,” Christian said. “Tech should accept that although anecdotal evidence isn’t good enough, it might reveal gaps in data availability.”

While tech teams have to explain the technical aspects of a project clearly in business terms, management is also expected to leverage soft skills, listen, and learn relevant concepts — while striving to communicate information on the financial state of the business as plainly as possible.

After all, an organization’s success is dependent on how well all the teams within it work together. And that’s particularly true for organizations striving to grow, innovate, and compete effectively.