top of page

How to build an Analytics system with reduced risks.

Updated: May 10

By Chelsea Lamb

Risk mitigation

Most companies these days are having to invest in Analytics and Data processing not only to gain a competitive advantage, but beyond that to survive in the marketplace. Without a data driven approach, it would be very hard to retain or acquire market share for their goods and services, especially with the amount of competition out there from different geographies as well as new entrants and startups, mostly because we live in a very information shared world reducing the barriers to entry at many levels. But Data platforms take money and resources – How do you build a system which will gain good adoption within the business, has successful outcomes as in being able to meet the objectives and can produce the actionable insights, yet make sure that this does not become one of those failed IT projects which we have all experienced , and could become a sink hole for money, resources and result in a loss of team morale due to the perceived failure. Here are some of the ideas and approaches, we take at Pacific Data to make sure this risk can be mitigated –

1. Have clearly defined objectives – this might seem very basic but is the key to the success of any project. Ideally it would be better to take a piece meal approach. Depending on where the company is in terms of the data maturity, the objectives should be achievable – does not matter if it is small – baby steps and walk before you run kind of approach is better than going on a huge initiative, which in and of itself brings about more of a size related risk.

2. Adopt Agile methodologies – If the company is using an Agile framework for other development projects – it would be easier to use the same for the Analytics platform as well. If not, we recommend our customers to start using an Agile approach, as it allows for the product to be seen as it is built and is flexible to change directions.

3. Avoid Scope Creep – When Agile is mentioned it goes without saying that there must be active barriers established to prevent scope creep within the project framework – reason being that Agile allows for faster development cycles and is a very flexible model but could quickly snowball into a different direction than intended, simply because of the speed of execution that is possible. Agile does not mean a lack of project leadership – This cannot be stressed enough. We need strong leadership and participation not just from the delivery teams but also the end customers of the project and the executive sponsors.

4. Building the platform – The goal of the analytics system should be to start delivering actionable insights as soon as possible. So, we need to start working with live data as soon as possible. We could do smaller subsets or divide the entire projects by geography, division, channel or one of the natural separations within the business and use the data to test the platform as it is built. The platform needs to be scaled horizontally and less vertically in the beginning and as the platform takes shape more features, functionality and reports could be added.

5. Technology selection – From our experience, we have learnt that we should always design the best possible system in terms of outcomes and not just the newest technology – It goes without saying that when we build something it should be "Built to Last" and should be robust to accommodate for future growth, but that does not necessarily mean the latest and greatest technology or the coolest technology out there, we should pick the one which is the most suitable for our goals and is most aligned with the existing technologies within the company. Good Technical people are one of the hardest to find on the planet and every effort should be made to leverage the existing expertise within the company. This is an area where Pacific Data is very strong in, and we help our clients by providing the support and necessary training in the beginning stages so that the client teams can quickly get ramped up and take ownership of the systems. Because only when the teams start taking ownership, ideas are more forthcoming and there is more accountability at every level resulting in a wonderful product which has been our experience.

In conclusion, Building an Analytics platform comes with its own risks not unlike any other new endeavor, but the risks can be mitigated through careful planning and adoption of best practices. We also believe it is very important to have a partner who has gone through these types of implementations many times – there is no substitute for real experience. So good Luck with your data platform projects and if you ever need any help, please feel free to reach out to us. We have been doing these sort of implementations successfully for many of our clients.


Featured Posts

Recent Posts


Search By Tags

No tags yet.

Follow Us

  • Facebook Basic Square
  • Twitter Basic Square
  • Google+ Basic Square
bottom of page