6 Steps for Data Confidence in a Digital World

Companies today are critically dependent upon their data.  It is the lifeblood of operations and is relied upon for strategic and tactical decision-making.  Data doesn’t just support the business – it is the business of a growing number of companies such as search, social media, integrators, and marketing companies.  With such a heavy reliance upon data, it is vital for companies to have confidence or trust in their data.

I have been listening to several IDG Tech Talk podcasts lately on the subject of data strategy and data trust.  These have been very informative on the subject and were the basis for my six-steps here.  You can find links to these podcasts at the end of this article.  With that, let’s talk through six steps for data confidence in a digital world.

  1. Develop a data strategy

The first step is to develop a data strategy.  The data strategy outlines how your company will use data to develop your business, drive innovation, and further your digital transformation.  Essentially, it takes business problems and defines how data will be used to solve them.  The data strategy starts by outlining each of the business strategy elements such as identifying people who are likely to buy our product, creating our product at the lowest cost and highest quality, and providing customers with reliable support.  You then look at the processes that feed into decision making in those areas and define the data that is required to deliver value.

  1. Data governance

The second step is to implement data governance to establish rules and guidelines for how the data will be used.  Governance should be defined to execute the data strategy.  Many often see governance as the antithesis to agility, but I think, like Graeme Thompson, CIO of Informatica, that governance provides freedom to use data because employees know what is allowed.  Without governance, data may be unused because employees do not know if it is allowed or not and unused data provides no value to the company.  There is a lot of sensitivity around data use and data privacy so it makes sense to clearly define the organization’s expectations for how it will be used.

  1. Perform a data inventory

Once we have defined how data will be used to transform the business and how that data can be used, we must assess what data we have.  This step happens third because our strategy is created independently of what data we have.  There may be additional data the organization must obtain to enact the strategy.  However, we need to conduct the data inventory before we can move to further steps since these steps will rely on know where the data is being held.  The data inventory is also essential for security because you need to know where your data is to protect it.

  1. Validate inputs

The fourth step is to consider how data is entering your processes to ensure that data is standardized, validated, and in the right format.  As the old saying goes, garbage in, garbage out.  An entire system can be tainted if bad data is allowed into the system.  Essential data may not be found, or it could be used inaccurately if it is coded incorrectly.  Data should be allowed only from trusted sources or validated with a standard process to ensure that data meets reliability requirements.  No company wants to make decisions based on bad data because those decisions will likely be the wrong ones.

  1. Clearly define terms and questions

It is important to look at the inputs, but equally important to consider the outputs of a data system.  The questions and terms that are used to query a system will determine how effectively it is used.  Companies must ensure that those interpreting the data have a good understanding of how to phrase their queries to achieve the best results.  For example, you may search for the most popular destination if you are a hotel service, but what you are really looking for is the most popular destination during non-peak times so that you can increase bookings through the rest of the year.  Without those additional parameters, the searcher will need to do much more work to try to find the answer, or they may not be able to find the solution at all.

We must also consider the terminology that we use.  The same term does not mean the same thing in every context, and so we must ensure that the data we base a decision on is actually related to the queries we use.

  1. Automate the process

The last step is to automate the process.  Standards are great, but employees make mistakes.  Once standards are translated into automation, a company can have the assurance that the operation is performed the right way every time and the resources required to operate the system are reduced.  This has a double benefit for the company.  Many times, self-service portals can be created to service data requests and streamline the process of taking the data strategy and turning it into a core competency for the business and one that fuels digital transformation.

 

For more information:

IDG Tech Talk: The Big Pivot, Ep. 11: How to Execute a Successful Data Strategy

IDG Tech Talk: The Big Pivot, Ep. 12: How Trusted Data Enables a Data-Driven Culture

IDG Tech Talk: The Big Pivot, Ep13: Trust Never Sleeps

Leave a Reply