MECCABrands has hired more data engineers, expanded its sales and marketing teams’ access to customers’ behavioural data, and plans to launch a new analytics platform in the next 12 months.
Head of data and analytics Dominic O'Halloran told Salesforce World Tour 2023 that the cosmetics retailer's main challenge at present is "right-sizing" its data operations to meet the needs of its rapidly expanding customer base.
“In the last eight years, we've actually grown the company tenfold and even since Covid, MECCA has doubled in size," he said.
Tech talent
The first step was bolstering MECCA's data and analytics team, O'Halloran said.
"I joined just over a year ago and at that time the data analytics team was very small. We just had about six data engineers who did everything."
Since then, the team has grown in size from six to "over 20" people.
The analytics team worked across three functions.
“We now have data engineering which looks after all that data warehousing, ingestion and transformation; reporting, analysis and insights; and then we have our analytics leads," O'Halloran said.
Self-servicing customer insights
In 2018 MECCA built a data warehouse on AWS and selected Tableau as its BI reporting tool.
The focus during the past 12 months has been enabling the sales and marketing team to self-service their access to 360-degree customer insights, O'Halloran said.
“Since I've joined, the real focus has then been to ... equip our merchandise team to deeply understand the customer right down to the product level," he said.
MECCA used Salesforce Commerce Cloud and MuleSoft to build a 360-degree view of customers by merging its retail, online, inventory and enterprise resource planning data.
“We now provide product-level customer insights that help brands understand the new customers, returning customers and churn," O'Halloran said.
“We're looking at things like switching and cannibalisation.
"Another big focus has really been to understand how our customers shop across our brands, across different departments, categories and so on."
In the last two years, the number of customers shopping both in-store and online has doubled; they make up 25 percent of MECCA’s customers, making it necessary for the sales and marketing team to have easily accessible data about individual customers’ online and in-store behaviour.
“We're looking to bridge that gap between the in-store experience and what you see online and really deliver personalised experiences to give you the right product at the right time and the right content," O'Halloran said.
"We really want to make online feel like you're in store as much as we can.”
Insights from viral campaigns
O'Halloran said that MECCA could also get insights into why some social media product promotions went viral and harness that insight to drive sales.
“I don't know how many of you follow beauty Tik Tok or beauty Instagram every day? Something can just [trend]," O'Halloran said.
“There might be some product that is featured and it just goes crazy.
"One of the things that we're using analytics for is so that when that happens we can understand who those customers are, what drives them, and what they are interested in?
“Then when they come into MECCA, we can then give them that experience and show them all the other brands and the other products that we have to offer.”
New analytics platform to launch soon
O'Halloran said that MECCA planned to launch a new analytics platform in the next 12 months that would drive its sales and marketing teams’ access to customer insights even further.
“We're bringing in a whole new data platform to MECCA so that we can really deliver the scale of analytics that we're being asked for, and part of that is to become truly self-service," he said.
“We use Tableau, and we've set up all of the data extracts and things like that. They're part of the new data platform that we're building.
"We're going to be looking at how we use a semantic layer to really open up the whole data warehouse and make that accessible through Tableau.
The platform will "abstract data away from what it looks like in the [data] warehouse and make it business-friendly and aligned to the business processes, rather than just [appear as] tables and fields."
More performance metrics
The data and analytics upgrade also included adopting more complex performance metrics systems, O'Halloran said.
"We've recently rolled out what we call the retail scorecard, a set of metrics that our stores can track every week to understand not just their commercial performance, but also their operational performance, customer NPS and sentiment at a store level, employment satisfaction, availability and conversion and all those things," O'Halloran said.
MECCA has started seeing an improved performance in the metrics that are tracked by the new retail scorecard.
“Within months of this being released we're already seeing metrics lift up. We're seeing improvements happening operationally in-store and it's now one of the most viewed reports across the business.”
The new metrics MECCA’s analytics team are tracking are being reviewed weekly by the C-suite, O'Halloran said.
“Every Monday morning, we have an all-in meeting led by one of the leadership team members.
“In those meetings, they take us through not just what trade was the previous week but also what was driving it.
“How does the average transaction value change and what does that mean for the outlook of the company in the week? These are all really closely followed [metrics]."
Investing in data governance
O'Halloran said that as MECCA rolled out analytics across more business functions it was complementing them with new data governance processes.
“in a really fast-growing business, data governance is really important but it's also quite hard to get your hands around," he said.
“We've got a number of different data governance initiatives, which I think are really changing the way that we're delivering work for the business.
“We're always discussing patterns and the right ways of doing things: What's best practice? How do we really ensure that what we're building is scalable and long-lasting and we're not duplicating on effort?”
