Revenue Teams Are Adding Complexity Faster Than They Can Prove What Actually Works

Some retail media networks pull nine figures in revenue and still run the operation on a spreadsheet. It points at something most revenue teams would rather not look at directly. The stack gets bigger. The workflows get more complicated. And the underlying question, whether the system actually works, goes unasked while the complexity piles up.

Nick Hinsley, Chief Revenue Officer at Vantage, works inside that problem. In his category, operational pain usually gets treated by adding something: a person, a platform, a fresh data source. The reflex often skips the step that would tell you whether you need it at all.

"If you can't measure it, you can't improve it," he says. "I'm a big believer in measure, act, learn." Most operators would treat that as a truism. Hinsley treats it as a rule he follows: he would rather refuse to buy or hire until the measurement says he has to.

The data is new, the operating model isn't

Ask Hinsley what has actually changed in how revenue teams make decisions, and his answer is deflating in a useful way. "I dare say that most aren't doing much different operationally to what they were doing one or two years ago," he says. New signal is everywhere, but it arrives in pieces. Onsite advertising came first, then offsite across Meta, Google, Pinterest, and connected TV, and now in-store, which North America has been slower to adopt than the UK and parts of Europe. Each channel carries its own ad types and its own data, and stitching them into one live picture is the part most teams still haven't cracked.

"It's really hard to have a consolidated view that's live," he says, "as opposed to the static view in a BI tool that exists in most retail media networks today." The spreadsheet creeps back in right there. Someone reconciles the channels by hand, so the shared source of truth is only as current as the last person who updated the tab. In a late-2025 Forrester Consulting survey, only 12% of networks reported to be able to truly orchestrate end to end.

Teams have more to work with than they did two years ago. What they can actually act on hasn't kept up. Revenue teams rarely lack technology. They tend to add complexity before knowing whether the system underneath is working, when accumulating data was always the easy part and building something that can act on it is where the difficulty lives. That's where Hinsley starts.

The command center is only as good as the data

Hinsley starts with treating the CRM as the thing everything else reports to. "We try and live and die by our CRM," he says. Vantage runs on HubSpot, wired into Slack and inbound and outbound email, and he only half-jokes when he calls the daily discipline "HubSpot hygiene." The CRM matters less than the discipline behind it: keeping the underlying data current enough that decisions don't need another round of reconciliation first.

This is where his sequence does the work. Strategy sets the direction, that shapes the structure of operations, and structure sets the process. Put measurement in first, establish a baseline, and watch whether it moves. If structure and process are already improving the numbers, the case for onboarding another tool or another head gets weaker.

"When scale comes and you've got all of that in place, then that's when you start to look at whether it's tooling or human or a combination of both," he says. The decision to add headcount or tooling comes after the numbers. The goal is to stop using people as filler for a system that was never fixed.

Measure before you add

Hinsley has run the experiment everyone is curious about. Feed a few inconsistent spreadsheets into Claude or ChatGPT and let it reconcile them. The insight it hands back is impressive until the limit shows up. "You can only do that once," he says. "It doesn't typically update on a nightly basis or on a frequency that you need." AI can make bad data easier to interrogate. It doesn't make bad data good.

So his sequencing is clean, centralized data first and AI layered on top only after, whether that home is the CRM or another platform. He describes a model pointed at fragmented data as something that tends to automate the confusion faster. He sells a platform that is part of his own answer, and he frames how he sells it this way: "I'm a big believer in serve, don't sell." He applies the same test to his own pipeline. Vantage runs on the SPICED sales methodology, which gives that measurement a framework, and the indicator he watches most closely is inbound RFPs. An RFP from an organization Vantage has never met, or one it has been building toward, tells him two things at once: marketing is landing, and the company's positioning is traveling far enough that prospects are coming to sales. That's a more useful signal than counting activity.

Before the next hire, platform, or data source, Hinsley would name the leading indicator that tells you whether the system already in place is working, establish the baseline, and watch it move. Then the evidence decides whether the answer is technology, a person, or nothing at all. Adding anything means the system earned it first.