How to make trusted decisions and get better results
We were invited to join leaders in Digital Transformation for a conversation on building business success with data. During this Fireside Chat, Fuse Data CEO, Dave Findlay touched on three areas for leaders to improve business with better decision making.
To help you pull some usable details, we’ve laid it out for you in these articles:
Facing Data Challenges
Listen in to Fireside Chat audio here
Do you find yourself playing that role of facilitator?
That's woven through the DNA of our engagements. It comes from this personal belief that we are in technology to help people achieve some goal. Without the people on the floor or the people that are taking business action, we wouldn't be doing what we're doing. We are enabling the force inside of an organisation and we need to have that customer-centric mindset.
I bring that in and coach the data teams and the technology teams up to maintain that point of view and their work. At the end of the day, I think that generates better results.
Even for the technology teams, understanding that your work has a business impact and seeing that is great. We get up and we go to work every day, it's really rewarding to see what you've done and how you've helped somebody save a couple hours of their time or save some frustration or made it so that they didn't have to work on the weekend. It's super rewarding and I try to bring that in when we do our work.
From your perspective how do you earn trust in the data?
Defining Data Quality
Trust is one of those things that when you lose it, it's really, really tough to get it back. Like if you put something out for a business group and the numbers are wrong, it's probably the last time, especially if it's an executive audience it's the last time they'll ever touch your work. But, trust comes from having good data quality and in order to have good data quality, you have to be able to define what good data quality is.
What does it mean to have, what is the data quality around a value of importer? Like what is a bad value? Should an order ever be zero, should it be negative? Should it ever exceed a million dollars? Like there's all these things that data teams we can't, weigh in on, and that's where that people focused approach comes in. When we're delivering data capabilities, we like to do it iteratively, by business unit.
As we're building out the underlying data infrastructure to support the business product we're building, we're working shoulder to shoulder, co-creating solutions with the business. It's not like the technology team goes off, does their thing, comes back for approval, we're co-creating a solution together with the business. By doing that, we’re working with them to define what data quality is. As we define what data quality is, we're reviewing the results of what we've defined and the rules we put in place.
After we applied some rigour and formally defined quality, we're starting to find challenges. So by working with the people - because companies don't trust data, people trust data. By working shoulder to shoulder with business and understanding what good data quality
means, that allows us to bake that into any of the data and analytics we build. So that when we release something, it's trustworthy, the business is able to adopt it and, and not have those nervous moments where they're like, “I don't know if this is correct. I don't know if I should make a decision”.
Assessing Data Quality
We seldom trusted the data without observing it. What we're seeing is our observation is the same as what the data is telling us. Now we can extrapolate and for the most part, trust the data. Is that which, is that something that you advocate or, or how, how are we going to get,
Sometimes it just takes time. It has to sit with the business for long enough that they start to become comfortable with it.
I always picture it like they just absorb this new capability and it takes time, you have to be patient. It's not something you can force. And a lot of the times I do, I advocate for running things in parallel. So if we're producing this, this new automated view reports, run it in
parallel for a quarter or two quarters with your current process.
I've done it before where we had people doing reconciliation between old world capability and new world capability and that went on for a number of months. It's not necessarily always a hard cut over to the new world. You have to be able to ease people in. And I think that, that comes a lot from focusing on the people, understanding the problems they have, understanding where they're at and what they need and the time they need in order to get from point A to point B. Even if they want a solution to the problem, it may not necessarily be something they can absorb right away. It takes time for change to take hold in an organisation.
What are the challenges around getting that one source of truth?
When you think about that person's job or the people's job and you multiply that it's probably not the only time that's happening across a company. If it’s happening in one area, chances are it's happening in other areas. When you have problems around synchronising the data, data quality and entry errors, there is a cost to that.
The biggest challenge I see is when you've got a person you're paying to do a job and chances are that manual syncing of information is noise. It’s not part of their core responsibility. When you have people biggest ROI in data and analytics is, freeing people up to do their best work. If you take them out of the muck and you take them out of having to perform manual data prep that they really shouldn't have to do in the first place, then they're going to perform at a higher level. If people are performing at a higher level, the company's going to achieve more.
The other challenge we ran into a lot, which is really interesting about this discussion, was having the same definition of terms. A lot of times we would have a calculation to define something we wanted to measure with a couple of inputs. If you didn't have the same definition, obviously, we'd tell you different things and that was the big challenge at times.
Work hard to do that to make sure there was a common definition, but we didn't always get there. So, common definition within the team, but as well as common definition cross-functionally.
At an enterprise level. you'd be surprised even in some of the most basic concepts it's tough to get the enterprise to align on something like even like, “what's an active customer?” Well, different teams may define active customer differently so you have to be able to sort of bridge those gaps and bring those people together to discuss
How do companies facing data challenges turn them into a competitive advantage and drive their business results?
That's the type of problem I really enjoy solving and the primary focus here is helping companies sort of establish that first data program or that first data and analytics capability. So I'd always start off with strategy and strategy is like one of those words that can scare people. It can sound like a big consulting firm coming in doing a half a million dollars strategic engagement.
I use the term strategy as much more like a pragmatic strategy, it's a people focused strategy. But really strategy to me is just a series of bets that we agree to make together. A series of decisions in order to achieve some common goal. So it doesn't have to be huge. But I think it has to be there in order to put everyone on the same page.And it serves as our guiding light as we iterate forward with the organisation. So you'll have your company strategy, like your corporate strategy, and your data strategy should really hang off of that. It should enable you to achieve the company objectives. But again, when I go in and that's an input for me, I need to understand where the company's going and how they want to get there.
But with data strategies, I like to say, let's start at the end. Sometimes you'll talk about building a data strategy and it's where are we today? Where do we want to go? And how do we get there?
I'm not super concerned about where we are today on its own. I want to start by saying where do we want to go? What is the future state? You guys have got these goals, you've set out this roadmap for you guys to accompany, and then go around every department and understand like, okay, this goal is attached to you or you're in it, you're you help achieve this goal, now how can data play a part in that? What information do you need? What insights do you need in order to be able to make the decisions that help you get that goal?
So if a company wants to enter a new market, or hit revenue expectation, the Sales Leader would have the information they need to hit revenue by entering this new market, what they need as you enter that market or if that market's viable.
Well, how do you track progress as you go through that? How do you know if you're heading in the right direction against that company objective? You start by working down, by understanding what the people need and to reach that corporate objective. .
You go through the department, you iterate through the business units. Once you understand where everybody needs to go, then you use that as a lens to focus on, what's there today? and what's the gap? If you just start with, “what do we have today” and I've seen a lot of people fall into this trap where they start profiling data and systems and they start doing all this inventory work and looking at stuff that’s not super relevant to getting to where you need to go.
So let's define where we need to go, like, we want to go to the moon, like, where you want to go, okay, now let's look at seeing what do we have available, and then what do we need to build in order to close that gap to get to where we need to go - the moon. Then you don't want a separate data plan, you want a plan that answers, what are my objectives? and then what data do I need to achieve it? rather than a separate data management plan?
Again, data teams are born out of technology functions, generally speaking, and, our job is to support the business. So if we put together a data strategy that isn't in line with where the company wants to go, then our data strategy is not the right one.
