.svg)
One of the cool things about this VC job is that you get weird chunks of insider knowledge about other industries. For example, I recently learned that there's a cursed phrase inside large complex retailers: "It was the weather."
This phrase will often surface in the "Weekly Trade Review", which is the beating heart of a retail operation. It’s where teams come together to pour over data to understand last week’s performance and how they need to shift their tactics for the week ahead.
The stakes are huge, and small improvements matter enormously. Why did sales go up in this city but down in that one? Where are we discounting too heavily? How is our sales mix impacting profitability? Do we need to make range changes? Price changes? Promotion changes? And what trends are competitors taking advantage of? Answers to these questions could mean literally billions in market share.
Everyone is scanning tables of data, searching for signal in the noise. However, the data often come from different data sets, and they say different things. If you want to make a case that X was driven by Y, you can probably find data to support it.
Then there are the delays. It can take hours just to pull the data. Properly matching it all up and interpreting it can take days. Months-long projects are often required to get all the information you need to properly review your range.
This is time retail teams just don't have. By the time you work out you’ve been discounting too deeply; you’ve already poured millions down the drain. And what if, like many fast retail chains, you want to change your tactics on a daily basis to meet the immediate needs of your local customers? That's impossible with today's technology.
So, how do retailers make these tactical decisions? The answer is that a bunch of the pros in the room have a sixth sense of what shifts in a certain metric might mean. When they see a pattern they know they go with their gut. But what happens when those patterns don’t emerge? Someone will probably say, “It was the weather.”
"It was the weather" it's a white flag of surrender. It means, “We don't know why things moved the way they did last week.” Maybe it literally was the weather, but the point is we don’t know. We can't see it. We have no source of truth.
This is a pretty wild thing to hear in an industry that spends $6B on everything from data warehouses to BI tools and consultancies. If you ask retailers about these tools what you'll hear is a list of what they don't do. My favourite quote: “Retail plus analytics does not equal retail analytics.”
The truth is the tech industry has largely failed to deliver on the promise that "big data" would transform retail.
Enter the founders of GoodWork.ai. Three people who bridge the gap between retail and tech. Matt Smith was a Partner in BCG’s Global Retail Group who has spent a decade embedded in the guts of the world's largest retailers. Alex Bergsåker has built AI systems for 10+ top global retailers. Ric Porteous is a PhD in Engineering who was known as one of the top data scientists at BCG.
After years of frustration, the founders realised that, for the first time, it was becoming possible to give retailers their dream: unlimited, accurate, real-time analysis.
Now, if you're in retail, hearing about yet another "insights" tool might feel depressingly familiar. But this is a different animal. What GoodWork.ai is doing has never been possible before.
They are building their own proprietary models to replicate the brains of the world's best retail analysts. They're using LLMs to provide an interface to those models so that real humans on the team can work in partnership with the AI analyst to dig into the "why".
What this enables is a lightning-fast, unified source of truth combined with an "always on" analyst that deeply understands retail and can provide answers in seconds or alert the team to issues before they even know to ask the question.
GoodWork.ai is already hard at work with some of the world's largest retailers as design partners. Every week, their models leap forward with shocking new capabilities. They've made me a believer that retail data is one of the major frontiers for the development of truly game-changing AI.
