From dashboards to decisions: closing the gap between insight and action

The gap between seeing a problem in your data and doing something about it is where most finance teams lose their edge.

Finance teams have invested heavily in visibility over the past decade. Dashboards for cash, spend, revenue, payments, FX, forecasts, month-end close. The data flows in from every direction, and yet most finance leaders would tell you the same thing: by the time they've reconciled what happened across four or five systems, the window to act on it has already started to close.

The problem has shifted from "can we see the numbers?" to "can we trust, interpret, and act on them before they go stale?" That gap between insight and action is where finance teams stall, and it's widening, because the volume of data keeps growing while the systems holding it remain stubbornly disconnected.

Dashboards show variance, not the next move

A dashboard can tell you cash has dropped, processing costs have climbed, or supplier payments have slipped. What it can't tell you is whether to delay a payout, adjust FX timing, chase a receivable, revise a forecast, or tighten card limits. That decision still travels through humans, spreadsheets, approval chains, and cross-functional debate.

The bottleneck sits between systems. Payments data lives on one platform, FX rates in another, card spend in a third, accounting in a fourth. Before anyone can pull a meaningful insight, someone has to reconcile data across all of them, often in a spreadsheet held together by formulas no one fully trusts.

Data teams can't really shoulder all the blame here. They're bandwidth constrained, with a perpetual queue of requests and follow-up times measured in weeks. The information exists across the organisation, but it's the stitching it all together that stalls everything.

AI changes the conversation, but only on connected data

AI can summarise variance, detect anomalies, forecast cash, and suggest next steps. McKinsey found that 44% of CFOs used generative AI for more than five use cases in 2025, up from 7% the year before. The adoption curve has steepened, and finance teams are at the front of it, applying AI most to process automation, financial forecasting, and risk assessment.

AI on top of fragmented data, though, produces fast wrong answers instead of slow wrong answers. If the data feeding a model is incomplete, contradictory, or stale, speed only accelerates the problem. The prerequisite for useful AI in finance is connected operational data, where transaction context, ownership, approval rules, audit trails, and human oversight all travel together. Without that connective tissue, AI surfaces the same fragmented picture your team was already wrestling with, only faster. Confident-sounding outputs built on shaky foundations are worse than slow reports, because they invite action on bad information.

Forrester research found that while 74% of finance leaders state that integrating AI into finance workflows is one of the top priorities for them in the next 12 months, 65% are struggling with fragmented data as a top technology challenge holding this back.¹

Agentic AI pushes this further. Models that trigger actions rather than surface reports could transform how finance operates, from flagging a cash shortfall to adjusting FX exposure to routing an approval. The opportunity is real, but it demands foundations that most finance teams are still putting in place.

Connected financial operations close the gap

When your payments, FX, cards, expenses, payouts, and cash management data sits in one connected system, you spend less time stitching together money movement data across platforms and more time acting on what that data reveals. You spot payment signals in a new market before competitors do. You catch a better logistics alternative and act on it the same week. You reallocate spend toward a high-performing channel as soon as ROI surfaces, rather than discovering it buried in next quarter's review. You close books faster because reconciliation happens continuously rather than in a frantic sprint at month-end.

That connected data layer also gives AI tools something trustworthy to work with. Finance AI built on a unified platform can surface genuine anomalies, flag cash flow risks, and accelerate approvals with the context needed to make those actions reliable. Airwallex consolidates these data streams into a single platform, giving both AI tools and finance teams a complete operational picture to work from rather than fragments that need stitching before anyone can move.

The shift is from reporting accuracy to operating cadence. Traditional finance asks "what happened last month?" Connected finance operations ask "what changed, who needs to act, and what controls should govern the action?"

What Hex is building to close the analytics gap

We recently sat down with Barry McCardel, co-founder and CEO of Hex, to explore this theme from the analytics side. Hex is an AI-native analytics platform used by companies including Anthropic, Reddit, Figma, and HubSpot, and McCardel's core observation lines up with the argument above: the traditional analytics workflow, where a request takes three weeks to move through the data team, breaks down in a world where decisions can't wait that long.

Hex puts AI directly into the analytics workflow so non-technical users can interrogate data in natural language. McCardel's favourite example: an executive chef at a quick-service restaurant became a power user, analysing oven timing, ingredient freshness, and supply chain patterns. Those questions would never have reached the data team because the queue was already overflowing with higher-priority requests. With AI-native tools, they don't have to. McCardel is candid about the limitations too. Models are probabilistic, generating plausible answers rather than verified ones. Hex runs an ‘agent behind the agent’ reviewing every AI-generated thread, scoring accuracy, and flagging incomplete context. The accuracy loop is continuous, never finished.

His parting idea deserves wider airtime: decision metabolism. You can have pristine data, gorgeous dashboards, and AI agents answering questions in seconds, and still be a slow-moving organisation if the culture can't convert insight to action. The bottleneck, he suspects, is organisational.

To really thrive in this environment you’ll audit where your decision latency lives. Sometimes it's in the data, sometimes it's in the systems, and often it's in the culture, in the gap between seeing a number and trusting it enough to move. The tools to close that gap are here. The question is whether you're compounding your advantages or letting them erode.

¹ "Building An AI-Ready Finance Function", a commissioned study conducted by Forrester Consulting on behalf of Airwallex, May 2026.