Sharon Flynn

How to Drive More Efficiency for Your Data Analytics Team

Sharon Flynn, Senior Manager of Digital Analytics at BMO Financial Group, breaks down her success tactics for driving efficiency across the data analytics team.

Diana Ellegaard-Daia
Sharon Flynn
Sharon Flynn

How to Drive More Efficiency for Your Data Analytics Team

Sharon Flynn, Senior Manager of Digital Analytics at BMO Financial Group, breaks down her success tactics for driving efficiency across the data analytics team.

By
Diana Ellegaard-Daia
Sharon Flynn
Sharon Flynn

How to Drive More Efficiency for Your Data Analytics Team

Sharon Flynn, Senior Manager of Digital Analytics at BMO Financial Group, breaks down her success tactics for driving efficiency across the data analytics team.

By
Diana Ellegaard-Daia
Sharon Flynn

How to Drive More Efficiency for Your Data Analytics Team

Sharon Flynn, Senior Manager of Digital Analytics at BMO Financial Group, breaks down her success tactics for driving efficiency across the data analytics team.

By
Diana Ellegaard-Daia

In this interview, digital analytics expert Sharon Flynn shares insights on how to drive analytics efficiency for your data analytics team, for less.

5 factors for running a data analytics team successfully

D.D.: What tips would you give companies to run a data analytics team for less? Are there any low-hanging fruits worth looking into?

S.F.: I have an acronym that I use and it’s called EARTH. So, running a data analytics team doesn't cost the EARTH. It’s E for empathy, A for automate, R for relentless, T for transparency, and H for harmony. This is really what it all boils down to. Let’s break them down:

1. Empathy

Empathy is a way of thinking about the organization or the data analytics team you're interacting with. If there's conflict or they're not doing something that's making sense to you, you have to approach it from an empathetic point of view. What incentives do our teams have that are clashing and how do we unpack that and get better?

2. Automate

Automate is being adamant. Get the machines to do as much as you can. If you're copying and pasting, stop doing that. That is nonsense, you’re way too expensive as a human. Human beings and human brains provide the insights and that is the most important output that we have. If you are doing a lot of data cleaning, ask yourself why are we doing that? How can we backtrack up the data manufacturing funnel to identify the problems? Use the tools that you have available right now and use them to your advantage. get to know your spreadsheets. I'm sorry, Excel is not the sexy tool, but it is what a lot of organizations already have that doesn't require a lot of investment and it can do extraordinary stuff with Excel and visualization.

If you're copying and pasting, stop doing that. That is nonsense, you’re way too expensive as a human. Human beings and human brains provide insights and that is the most important output that we have. If you are doing a lot of data cleaning, ask yourself why are we doing that? How can we backtrack up the data manufacturing funnel to identify the problems?

3. Relentless

Being relentless is being focused on the data analytics team. Are we relentlessly looking at the work we're producing and thinking about its value? I have another saying which is ‘every ad hoc is a failure’. Every time you get an ad-hoc request, think to yourself why. I'm a big fan of, what I call, bespoke requests. That's something that's unique that requires the subject matter expertise of the digital analyst - where their value for that output is explicitly acknowledged. But an ad-hoc request is most of the time something that people can pull data themselves from Google Analytics or Adobe - which have very user-friendly ways of pulling data. Why is our current report output or regular reporting not already answering that? Is there a skill set or math fear in that request? Every ad-hoc is a failure and they will kill you. Those things that arrive in your Jira board or your email that often sounds like ‘can I have the last 17 months of X’ will derail you and make you very, very inefficient. They're very low value to you, your digital brand, and they are not helping the organization.

4. Transparency

Transparency is being able and comfortable with explaining what we've done, how we do it, why we do it? And, when stakeholders ask, prove ow we do it. We need to be able to answer those questions and I welcome it.

5. Harmony

Harmony is all about the team that you build. At BMO, we have a very elaborate hiring process. When someone starts, resigns, or moves on, or is promoted, we're very clear on what gaps we want to fill and what's that team member going to look like in two or three years’ time. We also make sure that they are a good fit in terms of personality, and they challenge us. I ban invisible work and I hate overtime. Overtime hides the ad-hoc tasks and the inefficiencies that burn people out, especially now.

If you think about these 5 principles, that approach allows us to run an enterprise-level data analytics team with just nine people. We have over 900 end-users in our enterprise tool who self-serve and we manage 2000-3000 various sorts of dashboards and data visualizations. It sounds like a lot. But because we serve so many different teams, each team would have maybe two or three, but you start expanding that out to an organization as large as ours. That's how we know our worth, but it's not from a place of arrogance or a place of embattlement. It's being responsible - with the resources that our shareholders and their customers have given us, maximizing every dollar, and leveraging the tools that we use to the maximum.

Right now, we can no longer on the ‘I’ve seen this before’ approach. I think that data is absolutely vital to inform or place context around that. Any initiative with a data objective helps us keep focused and have an agile rather than a waterfall approach.

Importance of analytics for data analytics teams

D.D.: Why is data analytics important, especially during these challenging times?

S.F.: If anytime we needed analytics and data, it's now. Because for the vast majority of companies, this is their first global pandemic of this scale. Typically, in analytics, you pull numbers, insights, or research and compare them, at the most minimum, week over week, year over year, quarter over quarter. None of those comparisons make any sense right now. We don't have a previous period or a lookalike model to build. We can make estimates, but, in reality, we're in a 30-day ultimate agile environment now. So, all we can understand is how is our work today, did it make the impact of the objective, and if not, what should we do.

Right now, we can no longer on the ‘I’ve seen this before’ approach. I think that data is absolutely vital to inform or place context around that. Any initiative with a data objective helps us keep focused and have an agile rather than a waterfall approach. Organizations are trying to pivot to a digital plan that they were going to do over the next two to three years, and they now have to in the next 60 days.

We will see a definite delineation between those who are going that extra mile by including data in the decision set and those who discard data because it is too costly or too difficult to integrate. Organizations that take a deep breath and look at facts are going to do much better in this environment than those who push out tools or products without evaluating whether that was the right thing to do.

Data helps us keep grounded in our decisions. The cognitive load on everyone is very high right now, so in order to keep the decisions based on fact, rather than emotion, data is extremely important. We don't necessarily have to abandon long-term planning, but we may look at our current skill set and our current marketing structure, think very hard and ask ourselves if we are deploying this the most effectively for now? How do we move in a sort of, what I call, a strategic triage? How do we allow our colleagues, who are subject matter experts and may vary in seniority, have a voice at the table? How do we bring them to help us identify pain points in our current market stack or skillset, so that we can effectively decide what we can do with the tools we have.

In the current environment, I am concerned if organizations are making decisions absent of any kind of evaluation. I think we will see a definite delineation between those who are going that extra mile by including data in the decision set and those who discard data because it is too costly or too difficult to integrate. Organizations that take a deep breath and look at facts are going to do much better in this environment than those who push out tools or products without evaluating whether that was the right thing to do. MVP and agile are very good mindsets to have in these this kind of times and they both require data.

About Sharon Flynn:

Sharon Flynn is a Digital Analytics Pioneer with over 15 years’ experience driving business decisions with data. Experienced presenter, data advocate, model builder from excel to SQL and RStudio. She specializes in Digital Analytics-Adobe Analytics (Omniture,) Adobe Audience Manager, Test & Target, A/B Testing Comscore, Google Analytics, SEO/SEM, and Customer segmentation. She is the Lead Digital Analytics Consultant at Infotrust, and formerly worked as a Senior Digital Analytics Manager at BMO Financial Group in Canada.

newsletter

Get the latest marketing analytics insights straight to your inbox

No-nonsense marketing and analytics best practices from industry leaders, straight to your inbox.