People-centric BI

Tania Armstrong

I'm a career sales women and during my career I've been fortunate to be exposed to the wonderful world of market intelligence, networking, telecommunications and all things LAN, WAN and other three letter acronyms.  Never-the-less, prior to working in software, BI wasn't a term I was at all familiar with.  The outputs of Business Intelligence were something that I used - unknowingly.  The process of how my monthly sales reports were created were a complete mystery to me and of limited use once I felt the data could not be trusted.  Back in the day much of what we now call BI was in-fact manually mashed reports created from extracting data from a myriad of line of business systems by some poor person who liked pain and spreadsheets. It was time consuming and prone to both human and data quality errors.

Today's modern BI landscape provides tools to replace the manual operations and streamline processes like never before, however true BI doesn't start with a technology stack, it starts with the people and processes it supports.  Business Intelligence is about understanding the data landscape but more importantly working out how to get that data into a clean, clear and concise format to the people who can use it to make pertinent business decisions.

Business Goal

Step one is to understand what the over-riding business goal or the business case for change.

To understand how to reach the business goal, we need people from both operational and management levels as well as IT professionals to talk and tell us through workshops and interviews about what they are trying to achieve, what they need from their data, how they want to use it and where it will be used in the business. Getting to this point is critical to ensuring a good old fashioned ROI but also it makes peoples jobs easier, helping with eventual end user adoption.  (Note: Its so important to talk to everyone needing to use the information. A consultant I worked with in a previous job talked about HIPPO - or the 'highest paid person’s opinion’,   In software development it was often the financial stakeholders paying the bill for the project, who got their requirements met.  This approach can lead to reduced end user adoption post 'go live' as the finished project met the stakeholder and not the end user’s needs).

Once people are talking, focussing on documented and undocumented processes is key. This essential piece of business analysis helps us work out two things is it a process issue or the technology issue?  Regardless of how big an organisation is, standardised processes wax and wane through the lifecycle of a business and need updating and reviewing.  It makes sense to change a faulty processes wherever possible instead of utilising unnecessarily technology to fix it.  

Functional Requirements

The end result of our workshops is a list of functional requirements and corresponding technical deliverables.  Understanding the investment helps sharpen the mind on what is going to deliver the maximum value and priorities can be set based on ROI, urgency, cost, timeframe / effort, resourcing.  It is often at this stage where a steering committee can be worth their weight in gold, allocating priorities based in accordance with the over-riding business case / goal.

The delivery is usually the easy bit.

We tend to take the approach that will help our clients to consolidate their needs together, and prioritise their work. We end up with a flexible delivery based on getting key milestone requirements across the line incrementally.  Piecemeal delivery can appear a little underwhelming but it comes with sound reasoning.  While large software projects can be sexy, they are often flawed due to the inherent time lags and ongoing changes in the business itself.  A lot can happen in six to 18 months - buyouts, restructurers, market downturns, changes in product mix - all of this will impact on what you are looking to achieve out of your BI solution.  It also defeats the purpose of BI. 

Getting better information quicker is key

Streamlining your business intelligence is essential to supporting your reporting, business performance and predictive analytics goals.  Although ‘waterfall’ and road-mapped approaches have merit to individual stakeholders needs, the downfall of not taking an incremental piecemeal approach is that on the people side, those stakeholders who supported the large initial project may get frustrated as their expectations are a 100% perfect solution on first launch.  Additionally, the project may have difficulty getting traction on end user adoption.  Consolidating and having an incremental piecemeal approach is not only BI seems sensible but it allows for the stakeholders to see a faster ROI and feel more engaged in the project and its progression

Summary

We use a structured methodology to deliver the functional requirements of a BI solution.  Presenting these back to the business by prioritising the deliverables to meet budgets, road-mapping the technology investment, meeting internal time frames while delivering 'early and often' to end users is a balancing act.  But once we get to this stage, we can be confident that we are in a position to deliver a BI solution which is fit for the people using it.  After all – it’s the people that make your business and we are here to support that.