Latest / Customer Journey with Fexingo: Touchpoints, Mapping, and Multi-Channel Marketing / How Customer Journey Maps Reveal Channel Cannibalization
Transcript
- Lucas: So you ramp up paid social, your overall conversions go up, everyone high-fives. But six months later organic search is down twenty-two percent and your blended cost per acquisition has actually crept higher. That's not a growth story. That's channel cannibalization. Luna: Cannibalization — meaning the paid channel is stealing conversions that would have happened organically anyway? Lucas: Exactly. And the scary part is most customer journey maps won't flag it because they're built around touchpoints, not incrementality. You see a paid social click leading to a purchase and you credit the channel. But you never see the organic click that would have happened if the paid ad hadn't been there. Luna: So the map is literally missing the counterfactual. How do you even begin to detect this? Lucas: Let me give you a real example. A mid-market apparel brand we worked with — let's call them 'Ridge & Co.' — had been running a pretty standard multi-channel strategy. Paid search, paid social, organic, email. Their customer journey map was a beautiful horizontal flow with sticky notes for each stage. Looked great. But their blended CPA had been rising for four quarters straight. Luna: Classic symptom. More spend, less efficient. Lucas: Right. So instead of building another map, we ran a simple cannibalization audit. We looked at segments where paid social was active, and compared conversion behavior to a matched control group that didn't see those ads. What we found: sixty-three percent of conversions attributed to paid social in those segments were also happening in the control group — just via organic or direct. Luna: So nearly two-thirds of what they thought was incremental was actually just replacing organic. Lucas: Exactly. And because their journey map only showed the attributed path — paid click, then purchase — it looked like paid social was a hero. But when you layer on a cannibalization lens, the map tells a completely different story. The paid touchpoint wasn't adding new customers. It was intercepting existing intent. Luna: This is why I've always been skeptical of last-click attribution. It rewards the channel that closes, not the channel that creates demand. Lucas: And even data-driven attribution models can miss it if they don't include a holdout group. They're still working within the observed paths. If the paid ad shows up and the organic click never happens, the model has no record of that organic intent. It's like a dark matter problem in physics — you can only infer it by its absence. Luna: So what's the practical fix? How do you build a journey map that accounts for cannibalization? Lucas: First, you need a cannibalization layer — a separate analysis that runs parallel to your standard map. The way we did it for Ridge was to identify all conversion paths that included a paid touchpoint, then strip out that touchpoint and see if the conversion would still have happened via another channel. Luna: That requires pretty clean data. Most companies don't have a unified ID across channels. Lucas: True. But you don't need perfect identity resolution. You can do it at the cohort level. For Ridge, we compared conversion rates between markets where paid social was on versus off. If a market with paid social shows the same overall conversion rate as a market without it, but paid social is getting credit for conversions in the on market — that's cannibalization. Luna: Right, because the total pie isn't growing. Paid is just taking a bigger slice. Lucas: Precisely. And this is where the journey map becomes a strategic tool rather than just a visualization. Once you identify cannibalized touchpoints, you can prune them. Ridge ended up cutting paid social spend by forty percent in those segments. Their organic recovered within eight weeks, and their blended CPA dropped twenty-eight percent. Luna: That's a huge efficiency gain. And it came from understanding what the map wasn't showing. Lucas: Right. The map itself isn't wrong — it's just incomplete. The problem is most teams stop at 'what touchpoints appear in the journey.' They never ask 'which touchpoints are truly additive.' Luna: So what's the one question a marketer should ask tomorrow to start uncovering cannibalization? Lucas: Pick your highest-spending channel. Run a geo-holdout or time-holdout test for two weeks. Compare conversion rates in the holdout against the exposed group. If the holdout group converts at more than seventy percent of the exposed group's rate, you have a cannibalization problem. That's your starting point. Luna: Seventy percent — that's a good heuristic. Easy to remember. Lucas: And it's actionable. You don't need a huge budget or a data science team. Just a willingness to run a clean experiment and accept that some of your 'best' channels might actually be parasites. Luna: Parasites is harsh. But I get it. Lucas: Look, the channel isn't bad. The way you're using it might be. The goal isn't to kill paid social — it's to deploy it where it's actually incremental. For Ridge, paid social still worked great for new customer acquisition in cold audiences. The cannibalization was mainly in retargeting and branded search terms. So they shifted budget there. Luna: That makes sense. Retargeting someone who was already going to buy is just paying for what you'd get for free. Lucas: Exactly. And that's the insight a good journey map with a cannibalization layer gives you. It helps you see where your marketing is actually creating value versus just intercepting it. Luna: You know, this kind of thinking — questioning what's really driving results — it's exactly why we keep these conversations ad-free. We don't have to sell you anything, we just dig into what works. Lucas: And it's why listener support matters. If these marketing conversations have sparked something you've actually used — a test you ran, a budget you reallocated, a map you rebuilt — and you want to help keep this show independent, you can find us at buy me a coffee dot com slash fexingo. No pressure, just a way to keep the signal strong. Luna: Yeah, it's a small gesture that goes a long way. And it means we never have to chase sponsors or dilute the content. Lucas: Alright, back to cannibalization. I want to talk about one more detection method that doesn't get enough attention — the branded search uplift test. Luna: Branded search — that's when someone searches your company name or a specific product term, right? Lucas: Right. And it's a goldmine for cannibalization detection. Here's the logic: if paid social is driving true incremental demand, you should see a corresponding increase in branded search volume. People see an ad, get interested, and later search for your brand. But if branded search stays flat while paid social spend goes up, those paid conversions are likely cannibalizing direct traffic or organic. Luna: So branded search acts like a proxy for demand generation. If it's not moving, you're probably just harvesting existing intent. Lucas: Exactly. Ridge ran that test. They looked at branded search volume before and after their paid social ramp. It barely budged. That was the second red flag. First was the holdout test showing no incremental conversions. Second was flat branded search. Together, it was conclusive. Luna: I love that it's a free metric. You don't need a fancy tool. Just Google Search Console data. Lucas: Exactly. Every brand already has it. And it's a leading indicator — you see the cannibalization signal before it shows up in your CPA. By the time CPA rises, you've already wasted months of budget. Luna: So the takeaway for marketers listening: run a holdout test, check branded search volume, and overlay that on your journey map. If you see attribution heavy in paid channels but no movement in branded search, you've got cannibalization. Lucas: That's exactly it. And once you identify it, you can reallocate spend to channels that actually grow the pie. In Ridge's case, they moved budget from retargeting to top of funnel content and influencer partnerships. Their total conversions actually increased, even with less spend. Luna: Because they stopped paying for what they already had. Lucas: Right. And that's the power of a journey map that's honest about cannibalization. It doesn't just show you the path customers take. It shows you the path they would have taken if you hadn't interrupted.