Self-service BI: The driving force of data adoption?

Every modern company wants to be data-driven. The idea that we should incorporate data into our decision-making processes is today an important component in the ‘common sense’ of modern business. One of the reasons for this is the rise of business intelligence (BI), which has made it possible to structure and organise the massive amounts of data produced today by digital activities in ways that, in the first years of the internet, would have been unthinkable.

But even if BI can ‘tame’ big data and produce valuable insight, there are still a number of obstacles that can hinder its success. The biggest of these is adoption. If business intelligence isn’t being used by the various facets of an organisation that could stand to benefit from it, it will become almost completely impotent. You can have the best tech architecture in place, but without widespread adoption, it will be pointless.

The trend for self-service business intelligence has emerged to tackle this problem and make it easier for teams and individuals - sometimes with little to no programming or analytics experience - to leverage data in a way that is relevant and impactful to their specific context. In this article we’ll take a closer look at BI and self-service BI, and explore what’s driving the trend for the latter.

Business intelligence: history and context

Business intelligence — the use of data analytics to support business decision making — has been around for at least 20 years. However it is only in the last few years that big data has come to be its main driver. In doing so, it has made BI a critical part of digital transformation for many businesses.

One of the key reasons for this change can be located in the growth of cloud computing. Cloud makes it easier to collect and store data at scale, and the extensive ecosystem of features and tools on public cloud platforms like Azure and AWS have made it much easier to leverage data quickly.

However, although cloud providers offer BI features and tools (many of which can be very useful!), the increasing heterogeneity of data formats, and sources (financial, website, social media) has only led to a proliferation of solutions and platforms, all attempting to cater to specific business needs. These range from the purely descriptive - tools which are, ultimately, really reporting platforms - to more multi-dimensional and sophisticated tools that offer greater depth and the capacity for exploration.

The emergence of self-service business intelligence

The challenge of a multi-dimensional approach is that it is very difficult to cater for the specific needs of different teams and individuals within a given organisation. The more use cases you have across your organisation — marketing, product, finance, sales — the more customisation will likely be required. What the marketing team wants and needs from data, after all, will be very different to what a finance team needs. Ultimately, this will lead to low adoption. As powerful and fast as your centralised BI project may be, if it’s not relevant to the needs of particular individuals or teams, it will probably just be ignored. It’s for exactly this reason that standard BI projects fail.

It is from this context that self service BI has emerged. It allows individuals who may not even have a background in engineering or analytics to make use of data in ways that are pertinent to the challenges of their own roles. It means that they can be data-driven in their decision making without technical know-how or, indeed, the bottleneck of requiring support from engineering or centralised analytics teams. It is in this sense that staff can become what has been termed 'citizen data scientists' - someone who is not part of a data or analytics team but can still make use of analytics tools and models.

Centralised business intelligence vs. self-service BI

Standard BI relies heavily on IT or BI teams to create data infrastructures and analytics processes to support various lines of business. Data users will typically submit their requests to the BI team who will then generate relevant data reports. This does have some advantages: there is greater control over a company's data, for example, which is particularly important in the context of issues such as data quality and privacy. However, as mentioned, it can also create bottlenecks. Cross-team dependencies and constraints are forms of friction that delay the process of turning data into insight.

Self-service BI, however, is a user-centred approach. End-users should (in theory at least) be involved in a given BI project from design through to implementation and evaluation. If done effectively, they should be able to access data directly, create data models, and generate their own reports.

This doesn't mean the BI or analytics team suddenly becomes redundant, however. Instead, they no longer control the process in a centralised and cumbersome way, but rather govern and facilitate it in a way that allows end-users to work more effectively.

Managing the trade off between quality control and time to insight

Companies seem to face a trade-off between control over data quality and data assets versus a high level of BI adoption and a faster way of transforming data into business insight. However, this doesn’t mean that companies considering shifting to self-service BI should train all their staff to become data specialists. As we've seen, the emergence of the citizen data scientist provides a useful model for thinking about how team members should use and interact with analytics and BI platforms. However, it will nevertheless require a different approach to data security, training, and governance. Without proper data use and data security training, business users can misuse and misinterpret data.

A product-based approach to BI

These issues and challenges need to be consolidated into a product-based approach. Rather than seeing analytics and BI as a business function, it should be seen as an internal product that is managed and owned by a central team for the good of other stakeholders throughout a business. Doing so will minimise security issues and misalignment by ensuring that any self-service BI platforms are deployed with the needs, expectations, and skill levels of end users in mind.

This isn't easy and will often require a huge shift in organisational mindset and culture. However, it can be incredibly powerful and bring long term benefits if done properly.

Talk to us about how we can help you make BI a powerful driving force within your organisation.