What is Data Strategy?

| Zain Daniyal |

What does data strategy mean? What does it cover?

Data Strategy is the problem understanding step before embarking on the solution.

There is a famous quote that Give me six hours to chop down a tree and I will spend the first four sharpening the axe. The point here is the preparation of doing a task is extremely important.

Data Strategy is exactly that. It is the preparation of doing a data project with a well-thought-out plan such that we can avoid surprises and anticipate the solution and know what we are likely to get.

Data strategy involves the following points:

1. Understanding business goals – what is the business trying to achieve in a data project as well as everywhere else.

2. Scoping out problems and business pain points – What is the problem we are tackling and how important is this problem.

3. Analysing internal datasets and evaluating the need for external data sources – What data does the company have already and what else does it need?

4. Validating we have or can get all the data we need. Given the data we have, could we solve the problem? Does the data contain the right variables or can those variables be derived? Do we have enough data? Does the data cover all aspects of the problem?

5. Creating a simple data model to verify the solution for the problem – This is to check we have all the variables in the right place and that we can bring the data together.

6. Checking if we need to use AI and to what extent – Does the problem require the usage of AI? What kind of AI algorithm do we need? Not all problems require AI and not all problems require the deep end of the AI too.

7. Finding the best technologies to solve the data problem – Which cloud tech do we need? Do we need high speed read and write type database or do we just need a serverless data warehouse which is easily accessible.

8. Creating a list of human resources requirement – What kind of talent pool do we need?

9. Calculating the cost of data acquisition, data storage, data transfer and any other data costs. Are we likely to transfer data between clouds? Listing all the costs of data storage, data transfer and cost of data engineering. I.e. getting data to how it needs to be before you start your analysis.

10. Creating a road-map of how to implement the strategy in terms of time, technologies, costs and human resources – How long will it take, what the procedure looks like. What technologies are we going to use and what kind of people do we need.

11. Validating the full cost of the data project against the ROI and making sure its worth doing – Does the timeline and the cost makes sense for the business? Are we sure we will get better return on investment by doing this data project?

12. Implementation of the data project, I.e. data engineering and data science and AI – Lets get the project done. Let’s do implement what we just prepared for.

If you want help with your data strategy, book a free session now!



Copyright © 2022 ML Sense. All Rights Reserved.