In the face of global disruption, data teams have found they must pivot rapidly and deliver results fast. Here, we explore how agile project prioritization, alternative data sources and DataOps are helping them do it
Successful innovation involves adapting to emerging situations. In a year that’s been a whirlwind of twists, turns and setbacks, this ability to respond to unexpected changes has been a key data leadership trait.
ACS International Schools Head of Data Strategy and Analytics David Furlow is one business leader who found himself putting his skills to unusual uses in the age of COVID-19.
When the pandemic first began to spread, his team helped the organization use parent feedback and third-party data sources to provide analytics to inform policy decisions at each of its schools. But when the International Baccalaureate Organization announced a plan to use a ‘black box’ algorithm to assign students grades for exams they couldn’t sit, the priorities changed again.
“We had to be in a position to respond to parents who were saying, ‘What do you mean my kids didn’t get the grades that they had worked to achieve?’” he recalls. “We were in the position of saying, ‘We don’t write the algorithm for the grades, but we’re certainly going to advocate on our students’ behalf, on the basis of data.’”
“We assembled detailed, linear and textual arguments, supplemented with pictures of data, to appeal, ‘These don’t look right. These students have received incorrect evaluations, on the basis of the evidence’”David Furlow, Head of Data Strategy and Analytics, ACS International Schools
Lobbying the International Baccalaureate educational body to change its exam results policy became a top priority for ACS teams during the exam results period. Ultimately, they succeeded in achieving improvements on a wide scale, based on timely provision of evidence.
“We got news that about 40% of students had at least one of their grades raised,” he says. “Some students had increases of up five or six points out of 45, which is very significant if you’re trying to go to med school.”
Of course, organizations in other regions and sectors have been affected very differently by the pandemic. But data and analytics leaders across the continent have all been grappling with constantly shifting goal posts in 2020. Our research has identified three common techniques that have helped them meet business needs in these uncertain times.
Customer Engagement has been a Priority
In response to COVID-19, many organizations have refocused their strategies on their core propositions and engaging their customers.
For Furlow, that has meant informing and supporting executive decisions around things like distance learning and finding efficient ways to communicate policy changes to parents. But other data leaders have been tasked with helping their businesses retain customers or fill gaps in services.
“With all this market uncertainty and COVID-19 and the drop in demand in many markets, you need to be more valuable for your customer,” argues Manuel de Francisco Vera, Global Head of Analytics at eBay Classifieds Group. “If they have to choose between using two or three companies and they need to drop using one service, that could be us!”
“We generate insights for sales. With analytics, we are trying to empower the salespeople with a really good data-driven approach”Manuel de Francisco Vera, Global Head of Analytics, eBay Classifieds Group
The ‘insights matrixes’ de Francisco’s team creates for dealerships that list cars on eBay eCG Motors are one example of this approach in action.
“We also identify the main drivers for each dealer,” de Francisco adds. “We have many projects that are around using data to empower the sales account managers to go out and speak with the dealers.”
At other organizations, data teams have played a key role in measuring the pandemic’s impact on customer needs and digitizing parts of their offerings that can’t always be provided in person at present.
Alternative Data is Providing New Perspectives
Models that base predictions on historical data can be less accurate in eras that are unlike any that have come before them. At the same time, many data-focused executives have been faced with requests for totally new insights to inform business decisions around COVID-19 this year.
In both cases, alternative or external data sources have proven to be valuable tools. While many such datasets still provide insights into historical trends, they provide new perspectives on the past, giving a more complete overall picture on which to base decisions.
Integrating these datasets into existing data architectures has become a key priority for many European data leaders.
“[Data and analytics] helps a lot, especially during this unprecedented time,” says Rani Piputri, Head of Automated Intelligence Investing at NN Investment Partners. “Using alternative data, we are able to get a sense of what is really going on right now.”
“With this data and these new tools that are available, we can get a sense of what is really going on right now faster than before”Rani Piputri, Head of Automated Intelligence Investing, NN Investment Partners
Meanwhile, ACS International Schools is using third-party data to visualize the spread of COVID-19 in the regions it operates in and to inform decisions about in-person teaching.
What’s more, representatives from vendor-side companies have reported increased interest in alternative data sources from their clients over the past 6-9 months.
“We’ve seen a little bit more openness to it,” says Mackenzie Hargrave, VP, Product Strategist at data company FactSet. “I think this is out of a need to navigate the uncertainty around the pandemic and respond to increased adoption of alternative data across financial firms and their competitors.”
Embracing DataOps and Agile to Boost Productivity
Our 2020 State of Data and Analytics Europe survey revealed that many data-focused leaders have been under pressure to do more with less this year. In fact, 41% of our survey respondents expected their budgets to decrease in 2020 because of COVID-19.
We’re now seeing a renewed focus on methodologies such as Agile or DataOps to ensure that teams can deliver the right insights and capabilities to their colleagues as quickly as possible.
MoneySuperMarket is one example of an organization that has greatly improved the efficiency of its data team using Agile and DataOps. As MoneySuperMarket Group Data Director Harvinder Atwal says, these methodologies are great for aligning everyone around common goals and ensuring end-users receive new insights and capabilities regularly.
“We used to have long data and development cycles around models,” he recalls. “We looked at that end-to-end to say, ‘Where are the bottlenecks?’”
“It’s about flipping everything on its head. We’ll take an outcome and say, ‘What kind of data product can we build that will deliver that outcome?’”Harvinder Atwal,Group Data Director, MoneySuperMarket
He continues: “We broke up the entire model scoring pipeline into lots of different steps and the model training pipeline into lots of different steps, and then introduced automation and reproducibility.”
At the same time, breaking larger projects up into smaller chunks that can be delivered one-by-one makes it easier to continually update and reprioritize plans in line with changing business needs. With the impacts of COVID-19 likely to last well into 2021, data and analytics leaders should carefully consider how adopting practices like these might help their teams deliver value more consistently and efficiently.