What the Flaconi case study actually shows
Flaconi is Germany's largest online-only beauty retailer, carrying over 850 brands and more than 55,000 products, per Mapp's data-streams case study. At that scale the automation looks nothing like your store. The mechanics underneath it, though, are the same ones a Shopify store already has access to.
Most write-ups of this case study are vendor stories. They walk through the platform Flaconi bought and stop there. What they skip is the part an operator cares about: which automations paid, and why owned-channel revenue beats paid revenue on margin.
This article covers the same subtopics those pages do — the lifecycle flows, the real-time data, the mobile strategy — and then adds the profit math they leave out. If you want the wider framing first, our store automation playbooks guide maps where each of these automations sits in a real operation.
The numbers behind Flaconi's automation
Here is the sourced record, so you can separate the results from the sales pitch. Every figure below comes from Flaconi's published case studies, not from us.
Per SAP Emarsys, Flaconi runs more than 1,000 automated campaigns along the customer journey, and those campaigns generate roughly 60% of its CRM revenue. The same source reports a customer retention rate up more than 80%, push-notification revenue growing to three times the year-to-date average, and about 90% of sales flowing through the app and mobile web.
The efficiency numbers matter as much as the revenue ones. Emarsys also reports a roughly 20% faster technical go-live when Flaconi enters a new market, and campaign-creation effort expected to fall by up to 90% through automation. Read those together: automation did not just add revenue, it took hours off the team's calendar.
Lifecycle flows do the heavy lifting
The revenue-driving automations named in the Emarsys case study are ordinary ones: shopping-cart abandonment, back-in-stock alerts, and price-drop notifications. None of those require an enterprise budget. They are the exact flow types a store builds in an email platform today, which is why the concentration of revenue in them is the real headline.
The pattern to steal is triggering on behavior, not blasting a list. A cart-abandonment email fires because a specific shopper did a specific thing, so it reaches people already close to buying. That is what makes owned-channel automation cheap per dollar earned — you are not paying to reacquire attention you already had.
Real-time data and the mobile tilt
Flaconi's second case study, with Mapp, is about processing user interactions in real time to personalize the site and time outreach. The operator takeaway is not "buy a data-streams product." It is that fresh data makes automations fire at the right moment instead of a day late.
The mobile number reframes where to spend effort. With about 90% of sales on app and mobile web per Emarsys, Flaconi leaned into push notifications and mobile-first email design. If your own analytics show most orders coming from phones, your flows should be built and previewed for a phone screen first.
What an operating store can copy
Strip the case study down and three moves survive at any scale. First, build the three lifecycle flows Flaconi credits — abandoned cart, back-in-stock, price drop — before you build anything fancier. Second, trigger on real behavior so the message lands while intent is warm. Third, measure owned-channel revenue separately from paid, because it is the revenue you keep the most of.
That third point is where the vendor stories go quiet, and where the profit angle lives. A comparison of the platforms that run these flows is in our roundup of email marketing automation platforms, and the broader tool category is covered in our guide to AI marketing automation tools.
Worked example: the margin on a recovered order
Say your store does 340 orders a month at a $31 average order value — about $10,540 in revenue. Assume a print-on-demand product-plus-shipping cost of $14 an order, payment processing near $1.20 an order, and $2,800 a month in Meta spend.
Spread that ad spend across 340 orders and each paid order carries $2,800 ÷ 340 = about $8.24 in acquisition cost. Your per-order profit on a paid order is $31 − $14 − $1.20 − $8.24 = $7.56. Across the month that is roughly $2,570 in profit.
Now add one lifecycle flow. Say an abandoned-cart automation recovers 10 orders a month you would otherwise have lost. Those orders carry no fresh ad cost, so each one nets $31 − $14 − $1.20 = $15.80 — more than double the margin of a paid order.
Ten recovered orders is about $158 in near-pure profit, off an automation that runs itself. That is the mechanism behind Flaconi's 60%-of-CRM-revenue figure, scaled to your store: owned-channel automation is not just more revenue, it is your fattest-margin revenue. Doubling it does not double your ad bill.
Where the profit angle usually gets skipped
The catch the case studies never mention: a flow only makes money if the orders it drives are profitable after product cost, shipping, fees, and any discount code inside the email. A back-in-stock blast that leans on a 20% code can move volume and lose margin at the same time.
This is exactly the blind spot most automation tooling has. Your email platform reports opens, clicks, and attributed revenue — it does not know your per-order cost of goods, so it cannot tell you which flow actually cleared a profit. You end up optimizing for revenue that may be underwater.
There is a wider version of this problem across every tool you run, and it is the subject of our piece on AI business process automation. The short version: automating an action is easy; knowing whether the action made money is the hard part.
How this maps to an AI employee
The next layer of tooling is built to close that gap. Gartner predicts that by 2029, agentic AI will autonomously resolve 80% of common customer-service issues without human intervention, per its 2025 press release — a signal of where cross-tool automation is heading beyond single-platform flows.
PodVector AI's Victor is an AI employee built for print-on-demand and ecommerce operators. Victor is not a dashboard. It integrates with Shopify, Meta Ads, Google Ads, Printify, Printful, Gelato, and Klaviyo, so the same system that acts on your Klaviyo flows can also compute your true per-order profit — cost of goods, shipping, and fees included — and tell you which flows actually clear a margin.
Every write action Victor takes is approval-gated: it proposes the change and you approve before anything executes, the same human-in-the-loop pattern Shopify and Google build into their own tools. Recurring reports land in a folder in your own Google Drive, and its approval-gated support drafting lets you clear customer email without writing every reply from scratch. If you want to compare this cross-tool model against other options, see our breakdown of the best AI agents for business automation.
You can put Victor to work on your store and start with the profitability question the case studies never answer.
FAQs
What is the Flaconi GmbH marketing automation case study about?
It documents how Flaconi, Germany's largest online-only beauty retailer, uses lifecycle marketing automation to drive retention and revenue. Per SAP Emarsys, it runs 1,000-plus automated campaigns that generate about 60% of its CRM revenue, with retention up more than 80%. A companion Mapp study covers the real-time data behind the personalization.
Which automations drove the results?
The revenue-driving flows named by Emarsys are shopping-cart abandonment, back-in-stock alerts, and price-drop notifications, plus push notifications that grew to three times the year-to-date average. None of them are exotic. They are standard behavior-triggered flows any store can build in an email platform.
Can a small store copy this without Flaconi's budget?
Yes. The three core lifecycle flows are available in any modern email marketing tool, and they are what the case study credits — not the enterprise scale. Start with abandoned cart, back-in-stock, and price drop, then measure the owned-channel revenue separately from paid so you can see its higher margin.
Why is owned-channel automation higher margin than paid?
Because a triggered flow reaches a shopper you already reached once, so it carries almost no fresh acquisition cost. In the worked example above, a recovered cart order nets $15.80 versus $7.56 for a paid order, since the paid order still owes its $8.24 slice of ad spend. That gap is why Flaconi's 60%-of-CRM-revenue figure matters to your P&L.
How does an AI employee help with marketing automation specifically?
It connects the flow to the profit. Victor integrates with Klaviyo to act on your flows and, because it also computes true per-order profit across your live store and supplier data, it can flag when a flow's discount code is quietly eroding margin — a question your email platform's revenue report cannot answer on its own. Every action it takes is approval-gated, so you stay the decision-maker.