Coles looks to set up full stack observability

Coles is working to establish a more end-to-end view of IT service delivery, bringing all its telemetry together to resolve issues faster and shift internal perceptions of network reliability.

Coles' Andrew Greaney and Karen Sue.

Speaking at the Cisco Live 2023 conference in Melbourne, the retailer’s head of workplace experience and support Karen Sue said the company had “just started” with a move towards full-stack observability.

“A really good customer experience is not just made up of one thing: it’s 100 different things that have to work together in harmony, and from our point of view it therefore doesn’t make sense to look at those 100 individual things in isolation,” Sue said.

“What we’re trying to do at Coles, and have just started I would say, is really switch our way of thinking so that we are looking at things more from an end-to-end services perspective, so bringing all of those things together into one view.”

Head of core technology services Andrew Greaney said Coles had invested heavily in Cisco tools already, for monitoring specific technology domains - individual components of service delivery.

"We've got some really smart people doing some great things in isolation, so we’re looking at the wi-fi Cisco Catalyst Center [previously DNA Center or DNAC], we’re looking at our devices with ThousandEyes, and we’re looking at the application with AppDynamics,” Greaney said.

“We’re getting individual telemetry from those, but how can we extract the maximum value from all three, so rather than being told where the problem isn’t - it’s not wi-fi or the device or the application - we’re able to triangulate that data and actually work out where the problem is.”

The company is particularly keen to use the capability to resolve any technology-related issues that might arise in stores that could prevent it from replenishing stock.

“If you think about Coles as an organisation, the only thing that we need to do is to help feed the nation ultimately and ensure there’s food on the shelves so when customers come in, we can provide them with what they want when they need it,” Greaney said.

“Part of that ecosystem, if we look at it really simply, is we need the network to work, we need the devices that the team members use to make sure we’ve got stock on the shelves to work, and we need the application at the backend to work.”

Greaney said he was also personally invested in shifting internal perceptions that the network was always to blame.

“Maybe I’m over-sensitive because I’ve got networks within my accountability, [but] what we were finding is the network was a really simple thing to blame,” he said. “It was always the network’s fault, which I was reasonably upset about.”

Still, he noted the overriding driver was employee and - by association - customer experience; neither cared which technology domain an issue existed in, only that it could be resolved.

“Our only measure of success was whether that team member experience was really good,” Greaney said.

The retailer had worked to pull its internal monitoring teams together and had partnered with Cisco to “create some dashboards” to enable the root cause of a service delivery issue to be uncovered faster.

“So rather than waiting days or weeks to see where we might’ve had a problem - or never finding out - we’re now in a position where we can actually identify [problems] really quickly and a team member may not even know about it anymore,” Greaney said.