The most effective way to use data in advertising is to work through a 'Hierarchy of Metrics.' This framework, which echoes points James King makes on eCommerce Australia, moves from the most strategic business goal down to the tactical levers you can pull to improve it. It starts with a simple rule: your business only works if your Customer Lifetime Value (CLTV) is greater than your Customer Acquisition Cost (CAC).
Before you analyze a single campaign or ad set, you need to know these two numbers for your business as a whole. Justin Gecevicius emphasizes on The eCom Ops Podcast that understanding CLTV is critical for any advertising strategy because it dictates how much you can afford to spend to acquire a customer. If you’re spending $50 to acquire customers who only spend $45 on average, you’re losing money on every conversion, no matter how high your ROAS looks. The goal is a healthy CLTV to CAC ratio, ideally 3:1 or higher.
Once you have that top-level view, the next step is using other metrics to diagnose why your ratio is what it is. This is where you zoom in on metrics like Average Order Value (AOV) and Conversion Rate (CVR). If your CAC is too high, is it because your conversion rate is poor? Are you spending a lot to get clicks, but those visitors aren't buying? As discussed on Ecommerce Playbook, looking at funnel optimization through the lens of CVR can reveal friction points on your site. If your CLTV is too low, perhaps your AOV is the problem. You're acquiring customers, but they aren't spending enough on their first purchase to create a profitable long-term relationship.
Now you can apply this hierarchy to your advertising channels. This is where true data-driven decision-making happens. Instead of just looking at the ROAS of a Google or Facebook campaign, analyze the LTV of the customers it brings in. Run a cohort analysis, a concept Jake Cook explains on The eCom Ops Podcast, to track customers acquired from a specific campaign over time. You might find that one campaign delivers a lower ROAS but attracts customers with a much higher LTV. Another might generate cheap initial conversions that never lead to a second purchase. This is the kind of insight that allows you to allocate your budget effectively for long-term growth, not just short-term sales.
The place this framework breaks down is with attribution. Your ability to connect a specific customer's LTV back to a single ad is always going to be imperfect. As Judah Phillips discusses, marketing attribution is complex, and over-relying on last-click data from your ad platforms can be misleading. A customer might see a Facebook ad, get a text from a friend, and then search for you on Google a week later. Google gets the credit, but Facebook played a role. While the hierarchy of metrics provides the right strategic questions, the data is often messy, a point reinforced by the many challenges brands face with tools like GA4. You have to use the framework as a guide, not as an infallible truth.