Working in a big organisation often means that you need buy-in from quite a few different people when you want to make an impact – especially a technical one that not all major decision makers understand. I’ve found that spending time putting together a proposal to explain your idea with the right context to the right people is an effective way to get something off the ground. Here’s how I did this when my company was looking to understand how consumer behaviour led to our product’s growth.
At Absa, we have Card & Payments divisions that look at current needs, as well as the future of payments. A payments business generally grows with the increased use of its payment mechanism. As a newly formed strategic data analytics team that was still finding its feet in the organisation, we were tasked with understanding what drives an increase or decrease in this use.
Coming from a technical background, it sounds like a fairly straightforward modelling problem: "We want to find the Xs that best predict Y, right?" Building the best model with the data I had on hand could potentially have produced an answer – in theory. Having actual business impact, though, was a whole other story. Often, there are underlying reasons for the business problem that need to be addressed. Decision makers need to consider your solution, amongst many other ideas, to be on board with it.
In my experience, answering these business questions with the first ‘data answer’ that comes to mind doesn’t always work.
Data can be hard to understand, which leads to another question, and another, before finally resigning to rushing ahead with an ill-advised ‘quick fix’.
When it came to answering the question of ‘what drives the increasing use of our payment mechanism?’, I realised that I would have to map out my proposed solution in an easier-to-understand way that would lead to a quicker implementation. To do this, I took the following approach:
- Reached out to key stakeholders to understand the problem’s context
- Sought advice from other professionals in the organisation to cut back on analysis work time
- Presented the solution in a way that was easy for a non-technical audience to understand and give feedback
Here’s what this looked like in more detail.
Get in the room with the right people to understand the problem’s context
On receiving the ‘what drives an increase in the payment mechanism use?’ question, my first step was to follow the trail in locating the original source: Who asked this first? My thinking here was that only the asker of the original question truly knows why they asked it – in other words, they are closest to the problem that needs to be solved. Each person conveying the ask after that adds their own interpretation and assumptions.
After doing some investigating, I found out that it was the head of business who was looking to understand the product’s growth. This was pretty daunting but I embarked on booking a one-on-one session to discuss why it was so important to solve this problem.
In that one-on-one session, I posed these questions :
- "Where does this ask come from?"
- "What triggered it?"
- "Is it new or has it been lingering?"
- And, most importantly, "What do you think the answer is based on your experience?"
From the start, this got the number one stakeholder involved in the solution – making buy-in down the line much easier – whilst also empowering me to focus on solving the root cause of the ask.
Through this discussion, I learnt that the business growth had been fairly consistent. There was, however, no clarity on whether this was good or bad compared to the industry and market growth. If it was bad, the business would need very practical guidance on turning things around. If it was good, the business would need to make sure they understood the key drivers needed to maintain performance.
At this stage, the temptation was once again there to jump into finding a solution in the data, but it was still critical to suppress that urge. With only my own ideas and the thoughts of the head of business, I still didn’t have enough information to solve the problem as efficiently as possible.
Seek advice from other professionals in the organisation to cut back on analysis work time
Armed with a bit more context on the challenge, the next step was to ask more questions and listen more.
I spent the next two weeks meeting with people from all areas that could possibly have information or experience in driving or analysing business growth within the organisation. The finance department, for example, generally has a high level view of what influences their business forecasts. Economists have information on the state and movements in entire industries, whilst risk departments understand customer credit health and the appetite of the business to provide further credit.
Importantly, this showed me that being a data analyst does not mean that you need to create all of the facts from scratch.
After meeting with a variety of people, I now had clear guidance on the following:
- The head of business needed us to understand whether there were drivers of growth that we could practically influence.
- Other external views had given me information on the impact of what we could not control and how we compared to the market.
- Existing internal reports supplied an expert breakdown of key internal drivers.
Thereafter – and only then – did I start diving into relevant internal raw data sources to further supplement the wealth of learning I had already gained. Much of this entailed just filling the gaps in the reports and views that were already available or drilling down a level deeper to understand a certain driver in more practical detail.
Building up sufficient history gave me everything I needed to assess the impact of each driver over time on the levels of growth.
Towards answering the root question, I further identified those practical actions we could take as a business going forward, and simulated what the impact of each of these would be on future growth.
Present the solution in a way that is easy to understand and give feedback on
Even with context and supporting inputs, I found that a great solution is still likely to get lost when communicated poorly to people who haven’t spent as much time as you understanding the problem – especially when you work in a big organisation where many stakeholders are involved.
After compiling the relevant information, I had to start getting my ducks in a row. This entailed starting from that one key message I wanted to land (i.e. how does our growth stack up and what do we practically do about it) and then thinking about what ‘information storyline’ would be required to give the audience just enough context that they could grasp the proposal without being overwhelmed by information.
Going back to my original discussion with the head of business, I framed my proposal as getting to market leading growth. To land this one key message, I worked on crafting one visualisation showing the new growth target, with a breakdown of how much each key initiative would contribute.
Working backwards, I then included only the relevant information required to give context as to why this should be the new target, alongside evidence of the potential contribution from each initiative, making sure each insight followed logically from the one before.
Getting feedback
With a first draft completed, I set up small sessions with stakeholders to present and get feedback. To get diverse perspectives, these sessions were with departments like finance and risk, and one-on-one sessions with people like the head of the business.
I found that it was important to not make these sessions too big, as you want to take in each question or piece of feedback and really unpack it. During these sessions, I paid very close attention to any comments, questions, frowns, silences or loss of interest. Thereafter, I went away to rework my proposal, set up more sessions, got feedback, and repeated until I had received some good buy-in.
Getting feedback is hard. Especially when you're passionate about your work and the bulk of it is negative.
I learned that you need to maintain a strong mindset by understanding that this is a process of development – zooming in on any positives and keeping criticism in context.
From initially dreading them, I've come to really enjoy business presentations and the creative challenges they provide. It's become clear over the years that business data analytics does not start with data work, and it also does not end there. It's definitely a team effort.
So, in summary, my advice for successfully presenting a technical idea to a non-technical audience, is to always:
- Make sure you understand the root business challenge
- Gather input from both technical and non-technical sources to put together potential solutions
- Take the time to build a coherent story and constantly improve it by incorporating feedback you receive
A lot of this might not come naturally, and the learning process can be painful, but the amount of growth is immense and once you start to get the hang of it your work can become exponentially more fulfilling and impactful.
After finishing his Honours in Mathematical Statistics, Michiel van Staden joined the newly formed Customer Value Management team at the Absa Banking Group. Over the following decade, he learnt the practicalities of data analytics across fraud prevention, collections, legal recoveries, credit risk & campaigning, while also exploring the root cause of complex business problems as a strategic data analyst. Venturing into the realm of management, he now leads campaign analytics efforts, and has a passion for communicating data effectively and sharing his learnings. He also has a keen ear for listening as an aspiring performance development coach.
