That experience hung a lantern on a hard truth about data-driven marketing. It’s not simply about collecting swaths of loose data; it’s about cultivating insights from that data and making decisions as a result.
Across industries, the gap between data collection and decision-making continues to widen. According to Forrester, only 48% of marketing decisions today are based on quantitative information¹. McKinsey research shows that companies that successfully integrate analytics into marketing and sales outperform their peers by 15–20% in ROI². Yet despite the investments, most firms still operate with heaps of unsorted, unevaluated data, and inconsistent accountability.
We’ve seen this pattern play out repeatedly. Organisations invest heavily in analytics tools and reporting systems but fail to redesign the operating model around the insights, often failing to draw value from the data entirely. The result is impressive dashboards that produce little to no business value.
Here are five practical lessons that we’ve gleaned from that project, and from many others we’ve led before and since.
Here are five practical lessons that we’ve gleaned from that project, and from many others we’ve led before and since.
Every metric must be anchored to a potential decision even before you begin to collect the data. If no one changes their behaviour when a number moves, that data is entirely ornamental. Gartner found that 87% of organisations have low analytics maturity, meaning they collect data without embedding it into real workflows³, rendering it largely useless.
Define ownership for every metric, along with rules governing what happens when it changes.
Marketing teams often compete for budget rather than share accountability for growth. The Data and Marketing Association found that 45% of marketers still rely on siloed, channel-specific metrics⁴. True attribution starts when leadership aligns incentives and KPIs across teams, not when a new platform is deployed.
By looking at ways data can offer insights to multiple teams within an organisation, you can begin to better collect data in a way that makes it more actionable once collected.
Dashboards should trigger actions, not decorate meetings. In a 2023 survey, only 22% of businesses reported tracking campaign ROI correctly⁵.
Establish a closed feedback loop: insights leading to decisions, then measurement, then refinement. This ensures reporting becomes an operation with outcomes, not a folly designed to tick a box labelled “reporting”.
Short-term data wins look good on slides but rarely scale, and often retroactively identify themselves as vanity metrics. McKinsey found that companies using data-driven strategies see 8% higher profits and up to 10% lower costs over time².
Focus on metrics that compound efficiency - like improving lead-to-customer conversion rates while lowering cost per acquisition - rather than single-quarter gains or a short-lived spike in ROAS.
Data systems should evolve like any core product, with owners, maintenance, and user feedback. Harvard Business Review notes that 59% of firms struggle with data quality, and only 17% have a dedicated data governance lead⁶. A “set-and-forget” analytics project quickly becomes outdated.
Build data governance as a living capability, not an exercise to answer a single proposition after the fact or, worse, to prove an existing hypothesis.
The core lesson here is simple: Value doesn’t come from collecting more data, but from sutrcturing data exercises in order to make better decisions more quickly and confidently.
Being data-driven isn’t a state of technology; It’s a top-down culture of clarity, ownership, and sustained diligence.
