Driving data sharing inside your organisation
It’s no big secret that I’m a fan of data sharing and of the opportunities it can create for an organization and its people. Last week I wrote about data sharing from a technology perspective and while implementing the right solution is essential for data sharing to actually work in practice, it is also paramount to bring your people along on the journey. That’s the topic of today’s article.
User adoption
I see a lot of parallels between having people adopt a technology and really embrace the possibilities it brings and many other aspects of life. Let’s take cycling, for example, where people spend a lot of money on components and parts that are a few grams lighter. If the willingness to train and to build strong leg and core muscles aren’t there, no ultra light carbon bike in the world will win the race for you.
Implementing the best available data and analytics technology needs to be supported by enablement for your people so they can make the most of the new solution, and they can recognize how it makes their lives easier, their jobs more fun and allows them to tackle trickier and bigger data questions than ever before.
What’s in it for them?
A crucial element for user adoption is asking ourselves the question: what’s in it for them? ‘Them’ being the analysts, data scientists and business users who will form the largest part of the user community around a technology.
So, what’s in it for them? Can you spell out not just the benefits of the solution but also how it addresses the specific pain points your data people experience in their daily work? Do you know what challenges they’re facing in analyzing massive, unwieldy datasets to generate insightful reports on time?
Most likely it is a combination of the following:
- Difficulty accessing the data necessary for answering the business questions — due to silos, data ownership, legacy systems struggling to cope with concurrency and general demand.
- Data quality. This is a big one. It impacts timelines as analysts and data engineers work tirelessly to tidy up data for analysis. Technology helps but won’t fix it all.
- Ambiguous business questions that leave too much room for interpretation when it comes to finding answers in the data.
- Performance problems, with legacy systems slowing down or grinding to a halt during resource intensive queries and high user concurrency.
- Interruptions to their flow of analysis. Often performance problems cause interruptions as analysts wait for a query to complete and a system to ‘catch up’. While they wait for seconds, minutes or hours, they might lose their train of thought or switch tasks, which disrupts their flow. This slows down work and keeps ideas from flourishing.
- Not being able to use the best tools for the job. The tool might be free/cheap but is it truly what will help analysts do their best work?
So how does the technology you implement address these common issues? Make sure to take this into consideration when selecting a vendor and tool to improve the chances of analysts actually loving the software they will be working with.
The user wish-list
You might shudder at the thought of asking a large group of people about the needs they have when it comes to analytics technology. And yes hundreds of unique requests are not helpful, but I’m confident that the above list plus a few extras will be the overarching theme you’ll find when talking to your data people.
Bring that list to your vendor and be as specific as possible to ensure the technology delivers what you need.
What does that have to do with user adoption? Well, it’s the precursor. Users won’t adopt a technology that doesn’t solve their problems and instead creates more work for them, so as a first step make sure that whatever you choose is chosen for the right reasons and with the target audience in mind.
Break down silos
As humans we’re creatures of habit and we like things the way they are (usually). That’s why we eat our favorite foods all the time and take the same route to the gym/office/park. We need a bit of rewiring when it comes to implementing bigger changes and to overcome our initial resistance.
Data sharing can be a game changer for any organization and can very actively break down silos between departments to foster true collaboration. We once again must ask the question: what’s in it for them — how do analysts benefit from changing their behaviors and how can we show that benefit quickly as well as long-term? Beyond that we have to build new habits around collaboration so that the technology that can unite us within a business has a good chance of being adopted.
My recommendation is to develop an action plan with specific activities and initiatives that introduce and build collaboration. Rather than scheduling more compulsory team meetings, however, let’s have internal champions drive these initiatives.
Step 1: get your champions on board
I have no doubt you have a handful of people in your organization who are eager to use the new solution, learn about it, implement it for their projects and beyond. Those are the people who will help you drive change internally, so get them on board early, enable them with the right training so they can build expertise and understand how the technology can have a positive impact on the business.
Step 2: give them space and time to drive the change
Once your champions have bought into the idea, develop ideas with them on how they can reach a broader audience. Better yet: let them develop those ideas, but depending on your structure and organization size, be as involved as appropriate. Set a clear objective for their initiatives, for example training their peers or implementing the new solution for a specific department or data source. Then give them time and space within their job to work on this. That means something else needs to be delegated.
Step 3: Incentivise them
At the risk of repeating myself: what’s in it for them?
Apart from people getting involved in implementation and change projects as part of their role, there should be additional incentives. They can be monetary or more creative. What they look like is up to you, but please ensure that people aren’t expected to take on extra projects, complex change management work and additional tasks just out of the goodness of their heart.
The work they do to support the implementation of a new technology can be instrumental for the success of the project and that is something that should be recognized accordingly.
As a result of mobilizing your internal champions to drive these change, education and enablement initiatives with their passion and enthusiasm, you will be able to break down silos around data and departments and lay a strong foundation for ongoing collaboration in your business.
Deepening your analyses
I remember very well the many situations during my time as an analyst when I thought to myself “wouldn’t it be amazing if we could get data from here, there and over there and bring it together with our own data to learn something entirely new about our customers?”.
I know such a wish is very common among analysts and most organizations today use external, third-party data to some extent. And it’s not the using it that is the problem, it’s the using it effectively and efficiently that many organizations struggle with.
You can already purchase all sorts of data from providers and receive it in all kinds of formats. That, however, is not a sustainable process when it comes to the daily work with data, because it involves countless hours spent on repetitive processes, such as data loading, data cleaning, matching, joining before any meaningful analysis can be conducted.
Wouldn’t it be much better if a data pipeline — once established — provided analysts with a constant supply of data that is current, clean, transformed and ready to be queried? Yes, it certainly would be.
So when you build your strategy around user adoption and increasing the sharing of data between departments and externally in your business ecosystem, make sure that your technology choice addresses the shortcomings of traditional ways of sharing data, namely the lack of efficiency, data freshness, security and speed inherent to sending data files from A to B or using API calls for large data volumes.
Support a focus on analysis and analytical work
When you remove some, many or all of the annoying interruptions that get in the way of your analysts doing their best work, you’re creating room for very focused analytical work. Work that involves not just data analysis in their tool of choice, but also research, building and testing hypotheses, brainstorming with others, engaging with the business to test ideas and to get input on how viable their recommendations based on data are.
Much like I am much better at writing these blogs when my phone is tucked away out of sight, analysts will get deeper into their work when they can focus on the job they were hired to do, the job they are good at and are passionate about.
Being curious, living this curiosity, being inquisitive and driven to find answers to questions and problems requires real focus and the removal of distractions.
As a leader of analysts you can help your people do their best work by reducing the barriers they face and by enabling them to build their analytical skills. Hiring experts is one aspect, but they can only keep their expertise fresh if you give them time to continuously learn and develop and tackle those tricky data challenges and complex business questions.
Helping analysts do their best analytical work will in turn help you grow the adoption of data sharing in your organization. Analysts will share their solutions and will seek opportunities to include new datasets — internal and external ones — into their analyses.
The connection between training, development and data sharing may not seem so obvious at first, but it is there. In my work with hundreds of data analysts over the past 5 years I have come to recognize that those who work in an environment that enables them to do their best work, their ideas thrive and they find ways to be innovative, adding value to their organizations in the form of new insights, identifying potential revenue streams and new products.
I encourage you to take the above ideas, in part or in full, and test them with your own data people. Driving change is not easy, but it’s worth it and helping people make changes in their work, for example, related to a new technology, means the likelihood of a successful implementation and user adoption are greatly increased.
