Democratising Data with Self Serve Analytics

What is Self Serve Analytics and democratising the data?

Self Serve is an initiative which helps to open up data access to the entire organisation and educate the employees on how to work with the data, regardless of their technical and data background. In other words, data is democratised and used by everyone within the organisation.

Self Serve Analytics: How did we get here?

At Yubi, we have a centralised Business analytics team to serve the business users for their business intelligence and data insights-related work. Being a central and lean team, we felt that some business users have to wait considerably longer to get a quick and easy analysis report either from the Business analytics team or processing the raw data from the IT customer support team. These processes created a roadblock for business users and prevented them from being a data-driven function. To overcome their dependency, the business analytics team developed a self-serve platform where they could analyse their numbers and draw their insights using an easy-to-use tool. 

This idea got momentum from the entire organisation, and the business analytics team created a self-serve platform for different users within the organisation. The business analytics team was able to scale inorganically in the whole organisation within just a couple of months.

One of the apparent concerns from the senior leaders was about the security issues related to the data, as they were worried about the misuse and mistreatment of data, especially sensitive personal data about customers. However, we were able to block these concerns by deploying various security measures which didn’t allow the leakages to flow from the users. In addition, our security team has also trained the employees to use the data correctly to make a data-driven, informed decision for their business use case. Hence, that became a win-win from all aspects for all the concerned teams involved in this process.

What are the use cases of Self Serve Analytics ?

While the business analytics team has already published different internal dashboards for various internal business functions, we realised that every individual has different views towards their problems, and they would like to solve their problems without getting constraints on their thoughts. We could see some creative analysis done using the Self-Serve platform by unlocking the users’ thoughts. A few are listed below:

1. Turnaround time: This is critical for a digital lending business; every second counts to improve the end-user experience. Our loan accounts team was able to drill down to the last second of the entire loan process and identified the areas where they could reduce the turnaround time.
2. Nudging the users to be active: Another problem that any marketplace faces are identifying inactive users and how you make them functional by transacting on your platform. In the same context, our business users could locate the list of inactive users by day/ hour and help nudge them to transact on the platform by providing them with the right offers and services.
3. The other use cases are related to the overall market share and revenues. How this data has been distributed across various cohorts has helped the team understand user behaviour and preferences.
4. Past behaviour and trends of specific products and customer cohorts to understand where the sales team can intervene and help to grow the overall business.

What steps can an organisation follow to get on the Self Serve Journey?

1. Know your user: Understand the requirements of your power users, and bifurcate these requests into repeated or one-time activities. Understand whether these requests can be converted into dashboards for repeated use or have to be a stand-alone analysis. What complexity is involved? Is this request going to influence strategic or tactical decision-making? Who are users going to look into the data and various details associated with the users? 
2. Know your data: Understand the current inventory of your data. Where is the data stored? How is it accessible? Which software tools and technologies are you currently using to capture, store, and analyse data? Who is in charge of processing the data? What level of data is expected as part of Self Serve? What are the security measures you can deploy to avoid misuse?
3. Know your tools: Once you have understood the users’ requirements and the current set of data present in your system, the next important task is to choose the right tool that can be easy to use and can be easily trained. We have researched many BI tools, such as Power BI, Tableau, Looker, Zoho, etc., but decided on Quick Sight primarily because of its ease of use and seamless connectivity with the AWS servers. Since most users wanted to avoid getting into too much backend technical learning, Quick Sight was a good match. 
4. Prepare the data layer:  Gauge how data-literate your employees are. Depending on their needs, you can have various layers of data which can provide the granularity they expect. This layer should be thought through and well-defined by the data analytics team so that the end user can quickly get this information at their end. 
5. Data Dictionary/Cataloguing:  Describing data is as important as enabling data access to self-serve users. Having data dictionaries in place will help them understand the data better and help them do the correct aggregations for measures at every granularity.
6. Train your users: Invest in proper training and ongoing education for all users who would like to embark on a self Serve journey. Once you’ve implemented a new self Serve platform, educate the people to use it by showcasing the demos and the potential benefits of data-driven strategies. Speak about this platform in your internal forums and encourage more users to join and use this platform.
7. Measure the Usage: Monitoring usage metrics is vital in many aspects, like, measuring the success, governing for any data leakages, auditing the features being used, and so on.

What are the advantages of Self Serve – Democratising the data?

1. Better Collaboration: Different teams and departments can speak the same “language” around data and address business concerns faster at scale. 
2. Innovative thinking: Scaling this to many users will lead to more creative and diverse problem-solving.
3. Time saving: Employees can focus on more critical initiatives with a reduction in the manual effort required to obtain and disperse data.
4. Ability to scale faster: Though the Self Serve platform requires an investment upfront (such as the right tools, technologies, and security), it ultimately leads to more scale across the organisation as more people can get trained and become an extended team of the business analytics function.
5. More data-driven decision-making: Data democratisation makes it easier to focus business efforts in the right areas and justify strategic plans.

How did Self Serve help us at Yubi?

We have seen a positive response to Self Serve from our internal users since its launch in October 2022. Around 4% of the organisation is on self-serve as of January 2023 and has created around 180+ analysis/dashboards so far. This has reduced a total of around 900 man-hours for the analytics team and reduced the TAT for our internal users by at least 1-to-2 weeks. As we move towards data-driven decision-making, we expect these numbers to grow exponentially in the coming months.