Your Facebook ROAS is 1.8x. Google’s ROAS is 4.2x. You cut Facebook. Six months later, total revenue drops 30%. This scenario plays out across WooCommerce stores every quarter—and it’s almost always traceable to the same root cause: ROAS rewards the channel that closes the sale, not the channel that builds the customer.
Here’s the thing: 97% of Google Ads conversions still use last-click attribution (Google Ads research, 2025), which means Google gets credit for sales that Facebook, TikTok, or email started. The closer wins. The opener gets cut. And you make budget decisions based on a number that’s structurally designed to mislead you.
Why Your Attribution Model Is the Problem, Not Your Channels
Last-click attribution works like this: whoever touches the customer last gets 100% of the credit. A customer sees your Facebook ad, visits your store, leaves. Returns via Google Search three days later. Buys. Google Search gets the sale. Facebook gets nothing.
That’s not just unfair—it’s operationally dangerous. 92-95% of first-time ecommerce visitors don’t purchase on their first visit (Peel Insights, 2025). The entire customer journey that precedes the final click is invisible to last-click ROAS. If your customers need 5-7 touchpoints before they buy—typical for stores with average order values above $100—you’re measuring only the last fraction of the journey.
GA4 does offer data-driven attribution, which sounds like a fix. But it requires 400+ monthly conversions to activate (Google Analytics Help, 2025). Below that threshold, it silently reverts to last-click. Most small-to-mid WooCommerce stores never hit that volume. They think they’re using smart attribution. They’re not.
You may be interested in: GA4 Last-Click Attribution Is Hiding Your Best Marketing Channel
What ROAS Can’t See: Customer Lifetime Value by Channel
ROAS measures one thing: the revenue from a customer’s first purchase divided by what you spent to acquire them. It has no memory. It doesn’t know if that customer bought once and vanished, or bought five more times in the next 12 months.
LTV—customer lifetime value—is what ROAS misses. And the relationship between LTV and acquisition channel is almost never what ROAS suggests.
The typical WooCommerce pattern: Google Search closes well because it captures customers who already know what they want. Facebook and TikTok introduce customers who were not actively searching—they discovered you. That discovery-to-purchase journey is longer, so ROAS looks worse. But those customers, once they buy, often become repeaters. They have a relationship with your brand, not just a transaction.
A LTV:CAC ratio of 3:1 or higher is the accepted benchmark for sustainable ecommerce growth (Saras Analytics / Wicked Reports, 2025). ROAS alone cannot tell you whether you’ve cleared that benchmark. You need purchase history linked back to the original acquisition channel—who came back, how often, and how much they spent across their lifetime.
Referral traffic, for example, can show significantly higher LTV than organic search despite generating far fewer sessions (WooCommerce / MailPoet case analysis, 2025). Channel volume doesn’t predict customer quality. Channel ROAS doesn’t predict retention. You need the data to find out which is which.
The Trap: Feeding the Closer, Starving the Opener
When you optimize ad spend purely on ROAS, you’re doing something specific: increasing budget for channels that close sales, and cutting channels that build awareness. The closer keeps getting fed. The opener gets eliminated.
Short-term, this looks like efficiency. Your ROAS improves. Revenue holds steady—for a few months. Then it starts drifting downward because the top of your funnel is narrowing. Fewer people discover you. Google Search has fewer warm audiences to close. The metric that looked good is quietly consuming the system it depended on.
40-60% of digital marketing spend is wasted due to poor attribution (Commerce Signals, 2025). That waste isn’t random. It flows systematically toward channels that look good on last-click ROAS and away from channels that create the customers those last-click channels are closing.
This is the ROAS trap. And the only exit is data that ROAS can’t provide.
You may be interested in: Every Ad Platform Is Claiming the Same Sale
How to Actually Measure LTV by Acquisition Channel
The measurement you need isn’t complicated to define: for every acquisition channel, what’s the total average revenue generated by customers acquired through that channel over 12 months?
Three inputs are required:
- First-touch attribution: Which channel brought this customer to you the first time?
- Order history: How many times did they buy, and for how much total?
- Time window: Measure over 6-12 months minimum, not just the first transaction.
The challenge is input one: first-touch attribution. When a customer visits via Facebook ad and then returns via direct or search to buy, GA4’s last-click model records the second session—not the first. The Facebook touchpoint is gone before the sale registers.
Server-side tracking changes this. Events captured on your server—before they touch the browser where ad blockers and ITP restrictions operate—preserve the full session sequence including first-touch channel. With that data streaming into BigQuery, a SQL query joining session events to WooCommerce order records surfaces LTV by acquisition channel. No data team required. No expensive BI platform.
Transmute Engine™ is a first-party Node.js server that runs on your subdomain (e.g., data.yourstore.com) and streams every WooCommerce event to BigQuery in real time. Because it operates server-side, it captures sessions that browsers would otherwise drop—giving you a complete dataset to run LTV analysis against. The WooCommerce Looker Studio dashboard templates make the channel comparison visual without writing a single query yourself.
Key Takeaways
- ROAS measures first-purchase revenue only. It has no memory of repeat purchases, where most ecommerce profitability actually lives.
- 97% of Google Ads conversions use last-click attribution—meaning Google takes credit for sales that upper-funnel channels originated.
- GA4 data-driven attribution silently reverts to last-click below 400 monthly conversions, catching most small WooCommerce stores unaware.
- The channel with the worst ROAS may have the best LTV:CAC ratio. You need 6-12 months of linked purchase history data to find out.
- Server-side tracking + BigQuery provides the first-touch and full-session data required to calculate LTV by channel without a data team or expensive tools.
Last-click attribution credits the last touchpoint before a sale. Google Search often catches customers who already discovered you through Facebook—so Google gets the credit even though Facebook started the relationship. Your Facebook ROAS looks low because it’s measured by a model designed to ignore the work it does.
You need raw order data linked to first-touch acquisition data. WooCommerce order records contain purchase history; Transmute Engine streams first-touch and multi-touch event data to BigQuery. A SQL query joining order_id to session data gives you repeat purchase rate and average order value by channel—the two inputs that define LTV.
A LTV:CAC ratio of 3:1 or higher is the accepted benchmark for sustainable ecommerce growth (Saras Analytics, 2025). Many stores discover their ‘low ROAS’ channels actually deliver 4:1 or 5:1 LTV:CAC ratios once lifetime value is factored in.
Not based on ROAS alone. ROAS measures first-purchase revenue against ad spend. It ignores repeat purchases and customer lifespan. A channel with ROAS of 1.8x but a 4:1 LTV:CAC ratio is more valuable than a channel with 4x ROAS and one-time buyers. Always check repeat purchase rate before cutting a channel.
If your WooCommerce store runs paid ads on more than one channel, you almost certainly have a ROAS vs. LTV gap right now—and the only way to close it is to own the data that ROAS can’t see. Seresa gives you the server-side infrastructure to collect it and the BigQuery connection to query it.
