Business intelligence (BI) should be seen as a business initiative and a business ownership venture. It is - and always should be - about making correct business decisions, not only in a faster and better manner, but also to ensure it adds value to all business processes.
Previously, IT departments operated as part of the company, delivering solutions to the business as soon as possible and with user requirements as the key driver. Today, however, IT departments have started to operate like independent companies within a company, ones that want to make as much money as possible, stretch the work out as long as possible and have business pay for extras that have no immediate benefit to the BI initiative. The truth is that IT departments are taking over the BI initiatives, or rather, and in some cases, they are hijacking it. Although this may be a harsh statement, the reality is that data warehouse projects are failing because they are not delivering business requirements on time and within budget and are not end-user focused anymore.
While new applications developed can - and most usually are - “delivered” to a client, to be truly successful and focused on the end-user, a BI solution needs to be “implemented” rather than “delivered”. Delivery can be seen as developing in isolation, installed on a client site, and then handed over for use by the client.
All-inclusive
With implementation, however, it is of great importance that the client is part of every process within the life cycle of building a successful BI solution. When outside companies undertake the development, it is especially important to ensure the solution remains user or business focused, as the solutions are in some cases delivered to the information systems (IS) department, which in turn sends it through to the business. It is important to remember that a successful implementation in a BI environment must be done in parallel to the IS department and the business users, rather than in isolation.
Why? For the reason that a successful implementation is measured by the success of the user acceptance testing (UAT) phase. If the client was not involved throughout the life cycle of the BI solution, it is only at this point - the UAT phase - that the users will realise what it is exactly that will be “delivered”. If the solution was “implemented”, however, the UAT phase becomes just a formality. So how do developers get the UAT phase accepted from both a technical and business perspective?
Below are a few guidelines that will assist such approval:
* It is essential that the business requirement definition documentation, source-to-target mappings and report specifications are understood, accepted and signed off by both the client and the solution provider.
These documents form the backbone of the solution to be implemented and ensure that the client understands and agrees to key performance indicators, within the relevant business processes, that will be included and /or excluded.
* Setting of the UAT boundaries
It is important to have the timelines and deliverables set for the UAT phase. This must be agreed upfront between the client and the solution provider. The first step is to decide on what portion of data the testing will take place. Will it be for a particular month, week or day? This all depends on the availability of testing data and amount of the data to be used.
* End-to-end reconsolidation
Before even attempting UAT, basic reconsolidation needs to be successfully performed and handed to the client. This must include any flat file loads, direct database extractions, etc. Before loading any data into the data warehouse (DW) mart, this process must be done to ensure data and extract, transform, load (ETL) process correctness. In many cases, it is difficult for the source systems to give accurate reconsolidation figures that enable it to match the source system to the DW staging area. One method is to obtain some structured query language code from current reports and then transform these to work on the staging data. By doing this, one can now accurately compare “apples with apples”.
* Resources required
To be truly successful and focused on the end-user, a BI solution needs to be “implemented” rather than “delivered”.
Ren'e Muiyser is principal consultant at PBT Group.
The following three types of resources are the bare minimum that is required to successfully UAT a BI solution.
* Implementation manager - must have BI project management skills as well as technical knowledge regarding ETL and databases.
* Business/data analyst - needs to understand the business processes and relevant data to be able to assist the client in understanding the data and representation thereof.
* Technical specialist - must be able to hand over all of the ETL processes to the first line support team and assist the BA with any data issues.
These are just a few of the key considerations that need to be taken into account to ensure that the UAT phase is accepted from both a technical and business perspective.
It has taken years to prove to business how important BI is to the successful running of a company. Let's not go back to the dark and unintelligent ages, but rather ensure the IS department and the end-users understand each other, work together and that the BI solution is implemented, handed over and most importantly, accepted. By doing so, great successes can be achieved and business intelligence can grow from strength to strength.
* Ren'e Muiyser is principal consultant at PBT Group.
Share