What can Data Scientists learn from Sales Experts? - Part 1

During your data science life, have you ever had to pitch your work to someone else? Have you ever tried to "sell" your analysis to someone non-data related? If your answer is no, then don't worry this time will come sooner or later. If your answer is yes though, then you probably have already understood that there are times when we should be sales people and able to sell our work as a pro.

But let's start from the beginning. Most of the times we are assigned a project with a goal, i.e. a question to answer or a problem to solve. We collect the data, analyze them, extract some interesting insights and create some beautiful visuals to support our work. And then what? Is this enough? Does the assigner of the project automatically understand the work and most important its value? Obviously the answer is no. We have to present our work, explain what we did (but with not too many details because they don't care), summarize the most important results, but refer to some details when needed (but not too many details because again, who cares?) and prove that the time spent on analysis and visualization was worthy their resources (e.g. money, time, man-days etc.).

To prove this value, we have to "sell" our work, as if we tried to "sell" a product in a shop. We use the data as our weapon to convince them that there are actual and meaningful insights there and we are able to extract them Also it is important to have these key points very nicely summarized so that they do not have much work to do after the meeting. So, the big question of this post is how do I sell my work/analysis? If you have already worked as a salesperson, then some of the following things might seem familiar. If not, maybe you should try it. Professor Iordanis Ladopoulos (a well known professor in Greece) says that 'sales is the starting point for everything. What you learn in sales can open the path for anything else'.

Lesson #1 - Finding what intrigues them

In sales, the different reason individuals have for buying a product is called the hot button and is the leverage you can use to persuade them "buy" the product. However, in order to "press this button", the salesperson should first understand what drives the customer by asking all the relevant questions. In our case, the data analyst should ask questions to understand better the business need, what the customer wants to find out and what problems have not been solved yet. Then, they can proceed with providing the answers for the problem.

Lesson #2 -  Offer many "How to buy" options

Highly related to the previous lesson...when you need to prove that your analysis is valuable, you need to offer the most suitable visuals. However, if you have many alternatives at hand, e.g. a pie chart (many data science professionals are against this type of chart) vs. a bar plot (the best alternative to a pie chart) or a bubble chart. The same chart is not convenient for everyone, and even if it is you the expert to define the best chart, it is the "customer" who is going to use it in their daily routine. Offering many different ways may attract easier your "customer". So, be flexible and try to understand your "customer" needs.

Lesson #3 - Use Stories they can relate to

Given that you have already understood and solved the business problem, it is your work to explain the solution in terms that are understandable by the "customer". You should use their own wording and tell stories that they can relate to. For example, if you want to prove that your data analysis approach may track and identify sales problems, then you should explain it using previous sales problems which the "customer" had been facing. If they can relate your solution to their problem, then they can better embrace it.

Lesson #4 - Tailor your "Sales" presentation

Closely related to lesson #3, this lesson points out that along with the stories, the whole presentation should be tailored as well. For example, if you are going to pitch your work to the COO (i.e. Director of the Operations Department), you should talk about the problem they face and the final solution using the appropriate set of charts. A person at this level of hierarchy might not (and most probably won't) care about the algorithms that were used or how long it might take to refresh the results on a daily basis. On the other hand, if you are describing the same analysis to the company's Data Architect, then you should explicitly describe your approach, the algorithms, as well as their performance and how the daily refresh might affect the system performance.

Lesson #5 - Master the art of storytelling

No matter what you are telling and who you are talking to, there is one thing that you should master. This is the art of storytelling, which is how you describe the problem and most importantly how you describe your solution. When data analysts meet with their "customers", they are usually armed with plenty of facts and data. Although data may seem important, studies suggest that only 5 percent of people remember statistics after a presentation. However, about 63 percent of people remember stories after a presentation. So, just like sales professionals, data scientists need to arm themselves with stories that allow the customer project themselves at the solution.

The post is already long enough and it's better to take a break!. Another one is scheduled to introduce the next 5 lessons we can learn from sales experts. Stay tuned.... ;)

If you’ve found this post informative or interesting, let me know in the comments or like "My Datafication" on Facebook to stay up-to-date with the latest posts!