Dear Data Professionals,
You’ve done the unthinkable. You’ve analysed every nook and cranny of your data set, applied the best performing predictive models, designed the most appealing charts and graphs you can think of. Yet after listening to the insights you discovered, your project sponsor shook his head. "But that’s not what I want. Give me something useful that my team can act on. Don’t tell me things that I already knew."
Once being so confident about your capabilities, you’re bugged by the numbing guilt that you’ve let yourself down, and even worse, that you’ve let your project sponsor down. Will he ever give a crap the next time he sees you? Will you be replaced with someone he can count on?
Before you throw in the towel of defeat, know this – it’s never too late to get up and try again. But this time, you probably gonna need more than the technical skills of data cleaning, mathematical modeling or data visualisation. If you never wants to settle for less than making an impact, here are 3 non-technical questions (with answers included) that will benefit you not just at this moment, but for many years to come.
Why does it matter?
Here is an inconvenient truth that we should keep in mind when working with data. Our company, our customers and our business partners don’t care much about how we ingest data into the pipelines, what transformation techniques we apply, what model we select and how fancy our charts look like. They mainly focus on one thing: "Why does it matter?"
And when it comes to business, generally there are only 2 things that matter: time and money. Given a limited amount of time and money, who wants to waste precious resources on solving problems that don’t make a difference? Nobody! So before focusing on the data, the statistics or the software, let’s get a good grasp on how our project can either increase revenue, cut cost or save time for the organisation we serve, shall we? And if the problem can’t be clearly linked to either of these goals, I reckon it’s time to go back to the drawing board or prioritise other initiatives.
But how do we know what we don’t know? Well, if the following questions came from our project team, project sponsor, business users or even ourselves, treat them as a clue that we need to allow more space and time to understand why solving this specific problem matters more than others in the first place.
- What are we doing that for?
- Who are we are trying to help?
- Why does it matter to us? Why do we need it anyway?
- Oh, but what difference does it make to our bottom-line?
What is our scope of work?
If you have found a meaningful and important problem to be solved with data, congrats! Now comes the actual project planning. Just like any project, a data project needs to have a scope that defines why the project is required, what benefits are expected to be delivered and what success looks like upon completion. There are 2 benefits that we can gain from writing down the scope, and continue rewriting it as the project proceeds.
- When we put our thoughts into words, we can think more critically and deeply about the project and the problem at hand. This brings us closer to a thorough understanding of the people we work with, the shape that the work will take and the process of what needs to happen to arrive at the desired outcome.
- Writing down the scope helps us rearrange incoherent thoughts about the project into a well-built organised story. That way we can easily see the big picture and confidently communicate our ideas to anyone involved, thus allowing stakeholders to be on the same page and facilitating greater support.
Some of you might ask, "How can I write the scope when I don’t have much information to begin with?" Well, if you want answers, you have to ask questions. It sounds simple, but let’s be honest, probing for answers about existing problems isn’t gonna be easy. Nevertheless, we gotta do what we gotta do, which is to dig deep and obtain valuable information that helps us to solve difficult business problems and make better use of data.
I also have to admit that I don’t have the perfect answer to help you get through the process easily. But to guide you along on what questions to ask to define the scope of a data project, I have included a picture illustrating the "CoNVO" structure proposed by Max Shron, author of the book Thinking with Data. Also, I have added some useful probing questions that I have come across based on my own experience. Hope these will come in handy for you.
And if you are wondering how a CoNVO scope of a data project might look like, here is an example I wrote.
[Context] This major manufacturer supply mayonnaises, sauces, dressings and bespoke products for restaurant kitchens, food manufacturers and distributors. It makes money through selling and delivering products in commercial quantities. The person who asked for some advice is the CFO.
[Needs] This manufacturer does not know the right way to define competitive pricing to target at the customer and product levels. Account managers were making quotes and negotiate pricing with customers solely on gut feeling and personal experience. What is the right way to set better pricing to improve profit and customer negotiations?
[Vision] When this project finishes, the CFO will get a report outlining why the selected group of criteria is the ideal one to determine competitive pricing, with supporting examples; models for what-if scenario planning that connect the criteria to recommended price target and predicted impact on profitability.
[Outcome] If the CFO signs off on its finding, the selected criteria and the models will be incorporated into the Sales process across the entire organisation. Account managers will leverage price targets generated by the models and quantitative insights about the products and services to create quotes and negotiate with customers. A follow-up study will be conducted in six months to verify the impact of the new pricing solution on boosting revenue and profit for the organisation.
How do we talk about business?
Assuming we have done all the hard work, wrangled all the data set and extracted insights that could make or break the business, how do we talk to our business stakeholders? Well, I have to warn you that things may get a bit more complicated. But this is the final stretch, so let’s persevere.
In 1914, before coding and computers exist, Willard Brinton described a familiar problem in his book Graphic Methods for Presenting Facts: "Time after time it happens that some ignorant or presumptuous member of a committee or a board of directors will upset the carefully-thought-out plan of a man who knows the facts, simply because the man with the facts cannot present his facts readily enough to overcome the opposition… As the cathedral is to its foundation so is an effective presentation of facts to the data."
Fast forward to 2021, this gap between business and technology still holds for more than a century. Business stakeholders complain about how much money they invest in data analytics that doesn’t provide the guidance they desired whereas data professionals whine about how decision-makers misunderstood their laborious analysis or unreasonably expected them to pull magic from thin air. There must be a better way to make everyone happy, right? And I believe a better way to communicate lies in 2 simple principles.
Empathy is King
Imagine someone who doesn’t speak French and only knows how to cook instant noodle, how would he feel if he was asked to make croissants from a recipe written in French? The same logic applies here. How can business stakeholders to see the tangible results if the results aren’t communicated in their language or linked to the topics that they don’t care about? In this case, a little more empathy goes a long way. And this means trying to answer the following questions before talking about business with stakeholders.
- What do stakeholders have in common and where they differ?
- Considering key decision-makers’ prior beliefs, assumptions as well as personal interests, what do they care about? How would they react to the insights?
- What is your core message?
- What do you need them to know/ to think/ to feel or to act on?
- How can I translate the language of my work to the language of my stakeholders?
PEEL = Point + Evidence + Explanation + Link
How can we convey important insights that actually convince people? The PEEL method (borrowed from essay writing) is here to guide us. When presenting insights from data to business stakeholders, understanding how to structure our argument based on the PEEL method gives us the special power of persuasion, which eventually help our stakeholders to understand complicated ideas and be more confident in our results.
Here is how you can package your insight into a persuasive argument based on PEEL method.
- Point: an insight that could be reasonably doubted but we believe we can make a case for
- Evidence: a data visualisation, a map, a model that helped us translate raw data into findings and make a case for the point
- Explanation: justification of why business stakeholders should believe in the evidence we presented, what method we have used and how we have reasonably validated all assumptions and disclaimers
- Link: a clear statement to specify why our findings matter to stakeholders, how our insights generate potential business impact and what they should act on
Oh, wait! But where do these arguments come from? Like everything else in a data project, it’s an iterative process. At the beginning, we might have some points that are poorly backed up by the evidence, or we might end up with lots of evidence without any conclusive points or linkage to the business value. Don’t fret. A good approach to validate your argument is to have someone else listening and questioning your argument. If possible, I highly recommend running through the argument with your stakeholders. In doing so, your own understanding of the subject increases by fine-tuning the arguments while your stakeholder would be more informed about what you are working on. Who knows? They might be kind to offer extra domain knowledge to point you towards the right direction.
Ready to Move Forward?
In 1962, President John F. Kennedy said, "We choose to go to the Moon in this decade and do the other things, not because they are easy, but because they are hard; because that goal will serve to organize and measure the best of our energies and skills, because that challenge is one that we are willing to accept, one we are unwilling to postpone, and one we intend to win, and the others, too."
In 2021, given the greater need to derive meaningful and useful insights from data, let’s ask ourselves, "As a data professional, what challenge are we willing to accept?" How hard are we willing to strive towards becoming the magical unicorn who rocks at technical delivery while creating a long-lived business impact?
Originally published at http://thedigitalskye.com on January 9, 2021.
