Answer the same question - again & again & again
There is a reason why expert negotiators become so after years of experience - typically taking 20 years or more before being considered masters of their field. Expert negotiators understand every subtle aspect of negotiation: every question is carefully posed, and every answer is parsed in detail and with skill and at speed. There are layers and layers of subtlety built into the situation that an expert negotiator navigates and gets both parties towards desirable outcomes. The higher the stakes, the better the negotiator has to be in walking the tightrope.
This deep art of asking and answering questions is woefully under-appreciated in the world of Data Analytics and Data Science where far too many graduates are given absolutely no training in this all too important aspect of this role. Business stakeholders often do not think deeply enough before asking questions, and the Data Analysts/Scientists are not trained to probe those questions - which results in answers that are often underwhelming or far too basic. The essence of the situation at hand is lost, and business leaders are often left wondering why not enough progress has been made in spite of investments in Data Analytics.
Possession of the technical tools of the Analytics trade is a far cry from being able to understand the situation the stakeholder finds herself in, and then delivering an answer that will solve that problem. To get to the heart of the problem and answer it in a useful way, one has to ask and answer the core question(s) more than once, like a skilled negotiator.
The first attempt at the answer will give something usable and useful, but it may not move the needle enough. The business must have patience to ask that question again, and the Data Analyst/Scientist must now take another fresh look at the answer, and seek out nuances that may have been missed previously, subtle details that were overlooked in the excitement of working out a solution. What unwritten assumptions were made during the first round of answers? What can be improved in this second go around? Is there additional data now available? Where did the answers from the first round fall short?
Recommended by LinkedIn
This kicks-off round two - an answer that builds on prior foundation of knowledge and either directly enhances the prior answer, or comes at it from a completely different angle. The second round will begin to answer the gaps in the first attempt. And often the use of both answers will give a better, well-rounded view into the solution.
But now must start round three - and this go around, perhaps surprisingly, certain flavors of the problem, which were previously completely invisible (yes, invisible - the human mind only sees what it wants to see), will show up quite clearly. And it would become quite obvious that solving for these hitherto invisible aspects is the one that will make the biggest difference to the business.
As an example, if you are looking to provide your customers with the next best experience/action - quite often, your problem statement will be conflated with questions about content performance, content investment, content ROI, channel/vehicle performance and so forth. In other words, it is rare for the business to ask a straight question that gets to the heart of the problem, and even rarer for the Data Analyst to take the time to really understand what is needed. This translates into the need for multiple rounds of answers before the nuances, subtlety and the right lever emerge that greatly benefits the business.
Thus, my recommendation is to solidify your questions (maybe your top 20 questions in any given area) and to keep asking them until you have answers that are absolutely hitting the mark. This takes time and patience and a multi-year investment program, but like an expert negotiator, with the right questions and right answers, the business will progress towards some very valuable outcomes!
Thank you for sharing Viswanath!
I like the examples you shared and I can see how this apppies the NLP project we are working on which derives insights from transcriptions. I will use your reccomedations to review our use cases and challenge our team to think more deeply about the business questions and answers we are mining from our data.
I have discovered that repeatedly asking simply “Why?” when a stakeholder says what they want to know will often drill in further and further to get to what they are trying to solve. All too often we fall into the traps of the “x-y” problem and miss the real issue, on either the side of making the request or answering. https://xyproblem.info/
Very rightly said Sri. Grateful to be working for leaders like you who develop a 360 degree view of these intricate, amorphous data science problems by asking the right questions and driving a true business impact for enterprise organisations.