It has been, down the years, a salutary lesson in the data warehousing field that success is as much about people as it is about technology. Aubrey van Aswegen, MD of Knowledge Integration Dynamics (KID), provides a 10-step programme to ensure users commit fully and the organisation benefits accordingly.
As with any major business or technology initiative, people hold the key to success or failure when it comes to data warehousing. Clearly, you need executive buy-in, at various strata of the company; and you need a high calibre of resources deployed on the project. But above all, the users have to buy in to the goals, purpose and objective of the data warehouse. Otherwise you might as well cancel the project and save the money. Here are 10 steps to boost the success of the data warehousing project.
1) Do it with rather than for them. Involve the end-users of the data warehouse from the beginning, in terms of purpose, business objectives, design and process definition. Users who are involved from the beginning are more likely to be enthusiastic supporters of the data warehouse, and to pursue the business discovery that forms part of the its overall success. Obtain the buy-in, especially, of acknowledged specialists and leaders. They will in all probability be busy with other projects, where they are required to compensate for the loss of skills caused by staff cutbacks and skills shortages, and they will apply their energy where they see the greatest reward.
2) Involve the users in rollout and testing to promote ownership and acceptance. Users are most likely to find the inevitable bugs, performance bottlenecks and misalignment with business requirements. Involving them in the process will enhance the chances of yielding a data warehouse that is fit for purpose.
3) If the scale of the job seems beyond internal people, then engage the services of external consultants to boost overall efficacy. There are many specialists in the market whose skills could provide a different angle on the overall design goals of the data warehouse; and whose independence could help defuse any internal politics.
4) Provide appropriate training. Many users do the minimum with their IT applications and do not use them as effectively as they could. It also remains a truism that companies do not invest sufficient amounts of time, money and effort in training their staff. Under- trained staff will certainly not be able to maximise the potential of the data warehouse, thereby eroding the investment made in it.
5) Provide usability testing to ensure users can use the software in the manner for which it was designed. This applies in particular to the presentation layer, which is the user interface into the data warehouse. Screens need to be designed according to easily understood and interpreted templates, with a pragmatic blend of innovation and pragmatism.
6) You need to change behaviour patterns. Just because you build it, doesn`t mean they will come. This was the lesson painfully and expensively learnt by thousands of dotcoms around the world. The business needs to understand that the success of the data warehouse is as much about risk reduction and change management as it is about technology choice and deployment. Risk reduction involves identifying and countering the risk of failure; change management is about creating an atmosphere in which the goals and progress of the data warehouse are communicated visibly and regularly, with metrics towards progress.
7) Create a culture centred on continuous improvement. Excellence should be the only standard by which the data warehouse is measured. No matter how well you have progressed towards business goals, you should be striving for higher standards: in performance, reduction of batch window, boosting access and retrieval speeds, business discovery and provision of value to the business.
8) Conduct follow-up visits and audit regularly to boost success. The data warehouse is likely to stagnate unless management is seen to maintain an interest in it. Because of the magnitude of your investment in it - it is likely to run into the millions of rand - you will want to introduce an audit routine that examines use, success and user satisfaction and suggests corrective action. Engaging an external party for this audit is likely to boost its chances of success.
9) Introduce metrics that encourage the use of the data warehouse. For instance, you may want to incentivise user involvement by rewarding the departments with the greatest number of users.
10) Communicate consistently. Nothing will boost user buy-in more than honest and open communication; and lack of communication will inhibit its adoption. Users have seen technologies come and go, and they will be skeptical regarding the data warehouse. Keeping all informed as to delivery against objectives will ensure all are aligned.
Research indicates that few organisations obtain anticipated return on investment in new technology. Focusing on users, and increasing acceptance and take-up, will yield rapid returns, making this an area of rapid gain.
The bad news regarding data warehouses, as with other technology initiatives, is that end-users will need to be managed on an ongoing basis. The good news is that their negative reactions and apathy will ultimately fade out as they become accustomed to the new method of operation and the inevitable glitches are ironed out. Working with them to change their approach to the data warehouse will accelerate this process.
Share
Editorial contacts