Whether you are an experience analytics professional, or you are just making your first steps in Analytics these are six habits that will definitely improve your performance.
- Planning vs implementation time distribution: The digital industry is definitely pragmatic and due its dynamism there’s normally no time for anything. However a common mistake is going from the idea to the implementation with no, or not enough planning. If I would have to put a number, I would say that it has to be 80% planning and 20% implementation. But why almost nobody invest 80% in planning and research? Because during planning and research you don’t see nothing moving, you have no deliverables and is even hard to deal with the (internal or external) client’s expectations. Once you have the first 80% done, the implementation stage in the digital era is outstandingly quick. If you don’t invest the 80% in planning, you will have to stop at every moment to evaluate and make a decision. At the end you will have a Frankenstein.
- Design necessary profiles: Once you design your project you have to define the skills you will need to carry on the project. It’s not about hiring “an Analytics” professional or a Data Scientist which is something completely broad, but to define what are the required skills and find the resources with that specific practical experience.
- Automation and quality assurance: There’s nothing worst that put a repetitive activity in hands of a human. A human can convert a simple excel report made of several linked formulas into a time bomb in no more than a couple of months. If something is repetitive is mostly possible that can be converted into an automated process. Then it is highly recommended that you put two “cross check” quality assurance processes together to be alerted in case something goes wrong. It doesn’t matter how amazing your algorithms are, if your input is garbage whatever you get as an output, is made of garbage.
- Build metrics systems fully connected by hierarchy: You have to develop your metrics system by level. Begin with the CEO asking the following questions:
- What’s the CEO main objective?
- To who does the CEO reports to and how is his performance being measured?
- Who reports to the CEO and how is his performance being measured?
Then you have to do the same with each area: Marketing, Logistics, Finance, etc. Each area has a main objetive related to the company’s objective. This way marketing has to generate demand that can be converted into sales, so it generates an income previously determined by finance that will generates a margin. Ergo, it will increase the EBIDTA (Earnings before interest, taxes, depreciation and amortization) which is the main objective of the company.
If you don’t have this relationship very clear could happen that, for instance, the marketing department define a metric like “engagement” so they will be focusing their activities with that purpose. But how can you relate “engagement” with EBIDTA? Unless you have information that supports that every X engagement points you get Y% in new sales or purchase intention, then that metric is not related to its superior metric, ergo, is wrong. And wrong means that is driving the marketing department to make wrong decisions that drives the company far from its objective.
5. Simplify the information consumption: We have to begin with a very important statement. Analyzing and Reporting are two completely different things. When you analyze information you count with all the premises, confirmed and not, you made during that process. The person that receives the report doesn’t count with that information, making very complex to understand that is being said.
So the main challenge is once you understand that the person that will receive the report and that has to make a decision, is in a disadvantaged situation and based on that reality build a report that is able to tell the story you want to tell, including all the necessary information that you count with because you went through the entire analytical process.
6. Write your questions before begin any analytics process: If you have the right questions you don’t need to worry about the answers. In a world with too much information, you wont have any problems on getting the answers. However if you don’t have the right questions, you will be drowning in data before you don’t even realize it. The most common mistake is login in a Digital Analytics platform without the right questions or no questions at all. Those platforms looks pretty simple and intuitive, however with the amount of data they have and without the right questions you can there for life and when you finish you wont even remember what was you doing there.
So ideally write your questions in paper if you can, before even turning the computer on.