How we replaced a manual, campaign-by-campaign marketing operation with 190 automated flows that run 24/7 — generating attributed revenue while the team sleeps.
The marketing team at a major UK fashion e-commerce retailer was running on manual campaigns. Someone would build an email, pick a segment from a list, write the copy, select the products, and hit send. Every campaign was a one-off. Every email was a project. And every week, the same team that should have been thinking about strategy was spending 80% of their time on execution.
The problem with manual marketing is not that it doesn't work. It does. The problem is that it doesn't scale, it's inconsistent, and it's inherently reactive. You send a cart abandonment email when someone on the team remembers to set one up. You run a win-back campaign when the database starts looking stale. You react to problems instead of preventing them. We replaced that entire model with 190 automated flows that trigger based on customer behaviour, lifecycle stage, and real-time data signals. Marketing went from a calendar-driven operation to an always-on revenue engine.
Manual campaigns have three structural problems that no amount of hustle can fix.
First, they're labour-intensive. Every campaign requires design, copywriting, segmentation, QA, scheduling, and post-send analysis. A team of four can produce maybe 8-10 campaigns per week at quality. That sounds like a lot until you realise that a properly automated marketing programme runs hundreds of touchpoints per day across different customer segments, lifecycle stages, and channels.
Second, they're inconsistent. Manual execution means variable quality. The campaign someone builds on Monday morning after a good weekend is different from the one built on Friday afternoon under deadline pressure. Subject lines, product selection, send timing, segmentation criteria — all vary based on who's building it and when. There's no compounding because there's no consistency.
Third, they're reactive. A customer abandons their cart at 11pm on a Saturday. The marketing team sees it in Monday's report. They send a cart abandonment email on Tuesday. By then, the customer has either bought from a competitor or forgotten about it entirely. The window of intent — typically 1-4 hours for cart abandonment — has long closed. Manual execution cannot match the timing that automated flows deliver.
The 190 flows sit on top of a customer data platform that unifies behavioural data (browse, search, cart, purchase), transactional data (order history, returns, lifetime value), and engagement data (email opens, clicks, site visits) into a single customer profile. Every flow triggers based on real-time events or computed attributes from this unified profile.
The architecture follows a simple pattern: event triggers flow, flow checks conditions, conditions determine content, content is personalised and sent. A cart abandonment event triggers the abandonment flow. The flow checks: is this customer a first-time buyer or a repeat purchaser? What's the cart value? Are the items on sale or full price? Based on these conditions, the customer enters one of several variants with different messaging, timing, and incentive levels.
Flows are organised into lifecycle stages, each with its own set of triggers and objectives. Acquisition flows convert anonymous browsers into known customers. Onboarding flows turn first-time buyers into second-time buyers — the single most important conversion in e-commerce. Retention flows keep active customers engaged. Reactivation flows win back lapsed customers, informed by the same discount suppression logic that prevents unnecessary discounting across the broader CRM programme. And suppression rules ensure no customer receives more than a defined number of messages per week, preventing fatigue.
Curious what your margin opportunity looks like?
Free Tool
How much margin are you leaving on the table?
Answer 6 questions. Get a personalised margin estimate in under 2 minutes.
Take the Free Margin Audit190 flows sounds like a lot. It is. Here is how they break down by category.
Welcome and onboarding (12 flows): New subscriber sequences, first-purchase thank you, second-purchase nudge, category education based on first purchase, size guide follow-ups, and loyalty programme enrollment. The welcome series alone has 6 variants based on acquisition source and initial browse behaviour.
Browse and cart abandonment (24 flows): Browse abandonment by category, cart abandonment by value tier, cart abandonment by customer segment, wishlist reminders, back-in-stock notifications, and price-drop alerts on viewed products. Each has its own timing cadence and escalation logic — the first reminder is informational, the second adds social proof, the third (for non-VIP segments only) may include a modest incentive.
Post-purchase (38 flows): Order confirmation, shipping notification, delivery follow-up, review request, cross-sell based on purchase, replenishment reminders for consumable categories, care instructions for premium items, and outfit completion suggestions. The post-purchase window is where loyalty is built, and most retailers waste it on a single transactional email.
Retention and VIP (41 flows): VIP tier notifications, early access to new collections, birthday and anniversary triggers, milestone celebrations (10th order, 1-year anniversary), personalised edits based on style profile, and seasonal wardrobe recommendations. These flows maintain engagement with the most valuable customers without requiring any manual intervention.
Win-back and reactivation (29 flows): Lapse warning at 30/60/90 days, win-back sequences with escalating incentives calibrated to customer lifetime value, "we miss you" campaigns with personalised product recommendations, and final-attempt offers before suppression. Win-back flows are segmented by previous purchase behaviour — a lapsed VIP gets a different approach than a one-time bargain buyer.
Operational and transactional (46 flows): Order updates, return confirmations, refund notifications, loyalty point updates, account changes, payment failure recovery, and delivery preference updates. These aren't glamorous, but they're high-open-rate touchpoints that build trust and reduce support contacts.
Within four months of full deployment, automated flows were generating 37% of total email-attributed revenue. Not 37% of email volume — 37% of revenue. Because automated flows hit customers at the right moment with the right message, their revenue-per-send is 4-6x higher than manual campaign blasts.
Cart abandonment flows alone recovered an estimated £1.2M in annualised revenue that would have otherwise been lost. Browse abandonment flows — often overlooked because they target earlier-stage intent — contributed another £340K. Post-purchase cross-sell flows increased repeat purchase rate within 30 days by 18%.
But the most important result is what happened to the marketing team. They stopped building emails and started building strategy. With execution automated, the team redirected their time to creative testing, segmentation refinement, and channel expansion. Campaign quality improved because every campaign was now a strategic choice, not a production obligation.
The team went from producing 8 manual campaigns per week to overseeing 190 automated flows plus 3-4 high-quality strategic campaigns. Total customer touchpoints increased by over 400% while the team's execution workload decreased. That is what automation is supposed to do — not replace the team, but free them to do work that actually requires human creativity and judgment.
Every e-commerce business has a few automated flows. A welcome email. A cart abandonment sequence. Maybe a birthday message. Then progress stalls. The reason is almost always the same: the data infrastructure can't support it.
Building 190 flows requires a unified customer data layer that makes real-time behavioural data available to the automation platform. Without it, you can trigger on basic events (cart add, purchase) but you can't trigger on computed attributes (customer entering a lapse window, lifetime value crossing a threshold, browse pattern indicating category interest). The data layer is the bottleneck, not the automation tool.
This is why we built the CDP first and the flows second. The broader transformation programme invested heavily in data infrastructure — unifying disparate data sources into a single warehouse, building real-time event pipelines, and computing customer attributes that the automation platform could act on. The 190 flows are the visible output. The data infrastructure underneath is what makes them possible.
If your marketing automation is stuck at a handful of basic flows, the problem is almost certainly not your automation platform. It is your data. Fix the data layer, and the automation scales naturally.
Book a free margin audit. We'll map your current automation coverage and show you where revenue is falling through the gaps.
We go into businesses and make them permanently more profitable. Every initiative is EBITDA-tracked.