Why your best customers don't look like your best campaigns

The campaign that looks expensive might bring your best customers. First-purchase ROAS only tells you about the first order. If you're optimising solely on first-order ROAS, you may be defunding the channels that build your most valuable long-term customers.

This is the LTV:channel problem — and it's one of the most common structural mistakes in e-commerce marketing.

Why first-purchase ROAS misleads

Consider two campaigns running simultaneously:

• Campaign A has a first-order ROAS of 3.0×. But customers acquired through this campaign come back on average 3 more times in the following 12 months. Their 12-month LTV is 4× their first-order revenue.
• Campaign B has a first-order ROAS of 4.0×. But customers acquired through this campaign almost never return. Their 12-month LTV is their first-order revenue, and nothing more.

Campaign A looks worse at acquisition. Campaign B looks better. But at 12 months, Campaign A is generating 3× more value per customer acquired. Optimising only for CPA can defund channels that build loyal customers and scale channels full of one-time buyers.

Real LTV data by acquisition channel

One DataMaster client showed the following 12-month LTV by acquisition channel across 11,633 customers:

Google Ads customers were worth nearly 3× more at 12 months than Facebook Ads customers — while costing half as much to acquire. The LTV:CAC ratio: 11.4:1 for Google vs 2.1:1 for Facebook. These numbers change every budget conversation in the business.

64.7% of customers placed a second order. But that repeat rate varied significantly by acquisition channel — Google Ads customers returned at a materially higher rate than Facebook Ads customers. First-order ROAS didn't show any of this.

How to act on LTV by channel

The framework is five steps:

1. Calculate 6-month LTV by channel. Pull Shopify customers acquired 6 months ago, segment by acquisition channel (using UTM data), and calculate their total revenue over the 6 months since first purchase. This is your 6-month LTV per channel.
2. Divide by NC cost → LTV:CAC per channel. Divide each channel's 6-month LTV by its new customer acquisition cost. This is the single most important number for deciding where to put next month's budget.
3. Compare channels on LTV:CAC, not first ROAS. If Google Ads has a lower first-order ROAS than Facebook but a higher LTV:CAC, allocate more to Google. You are buying customers, not conversions.
4. Adjust budget allocation accordingly. This doesn't mean abandoning channels with high first-order ROAS but low LTV. It means understanding the full picture before deciding. Some channels are good for first orders; some are good for building long-term customers. Your budget mix should reflect your current strategic priority.
5. Use NC ROAS for prospecting decisions. When evaluating prospecting campaigns, calculate new customer ROAS (revenue from first-time customers only, divided by spend) rather than blended ROAS. This removes the distorting effect of campaigns that simply convert warm, existing customers who would have bought anyway.

The email paradox

In the client data above, email is the most striking finding. Email brings 45% new customers at zero ad spend, with a 64.7% overall recurring rate and the highest LTV of any channel.

Yet most e-commerce brands exclude email entirely from their ROAS analysis. Email doesn't have a CPC, a CPM, or an obvious "ad spend" — so it doesn't fit neatly into a ROAS calculation. The result is that the highest-LTV channel in most e-commerce businesses is also the most undervalued in every budget conversation.

If you're investing in email list growth and email flows, include the estimated cost of that investment (ESP fees, design time, offer discounts) in your email "spend" and calculate email LTV:CAC as seriously as you calculate any paid channel. The number will often be your best channel — and it won't show up in any dashboard unless you build it yourself.