Marketing Automation: Definition and Flows That Pay

TL;DR. Marketing automation is sending triggered by behaviour rather than by a date. The difference from a campaign is not technical, it is economic: a flow runs on existing traffic, so it produces revenue without adding pressure to the list. Seven flows cover the essentials, and the first of them : checkout abandonment, is the most profitable flow in all of Klaviyo.

The most common confusion is between the tool and the practice. Every platform can trigger an email four hours after an abandonment. What separates a programme that earns is the data feeding the trigger and the exclusions around it.

The definition, and what it implies

An automated message is triggered by an event : a contact action, a date, a status change, rather than by a human decision made for a given date.

The economic consequence is what matters. A campaign asks attention of a whole audience at a moment you choose. A flow addresses one person at the moment they showed intent. The first consumes patience; the second consumes very little.

Which is why the logical investment order is flows first, campaign frequency second.

Flows versus campaigns: the balance

Our reference framework targets roughly fifty-fifty between flows and campaigns in email revenue.

The most common imbalance is revenue carried almost entirely by campaigns. It looks fine as long as you keep sending, and it stops dead the day you stop. Flows are the passive engine of CRM: they keep producing through holidays, production gaps and periods when the team is busy elsewhere.

Across total revenue, the floor for a healthy account is 20% from email, the zone we target is 25 to 30%, and the best accounts reach 40%.

The seven flows, in order

1. Checkout abandonment. The most profitable of all. We run no account without it live: every day without it is lost revenue.

2. Welcome. The most-opened email in a customer's entire lifetime. A sequence, never a single email : see the Klaviyo welcome flow.

3. Cart abandonment, with a threshold and exclusions : see Klaviyo abandonment flows.

4. Browse abandonment. The visitor who viewed a product without adding it is your largest volume of intent. It is also the flow that saturates fastest : see Klaviyo browse abandonment.

5. Post-purchase, treated as a repeat-purchase channel rather than a confirmation: thanks, product education, cross-sell. A customer who just bought is the easiest to make buy again : see the Klaviyo post-purchase flow.

6. Replenishment, on products with a predictable consumption cycle.

7. Win-back and sunset, installed together on any mature account. Without them the list ages and deliverability degrades : see win-back and sunset.

Where to start when everything is missing

The order matters more than the tooling, and it is close to the same order in every account.

One flow before seven. Abandoned checkout carries more revenue than the other six combined in most catalogues, and it needs one event and one exclusion rule. A programme that ships it properly earns the credibility to build the rest.

Exclusions before scenarios. Before adding a sequence, make sure recent buyers stop receiving acquisition messages. The most common complaint in any audit is not a missing flow, it is a customer receiving a first-order discount two days after ordering at full price.

A frequency cap before segmentation. Segmentation decides who receives what; the cap decides how much anyone receives in total. Without it, each new flow quietly raises the pressure on the same engaged contacts, who are also the ones you can least afford to lose.

Then measurement. Not attributed revenue, which credits the last touch, but a holdout on the flows that matter. It is the only way to know whether a scenario produces revenue or merely records it.

The three design errors

Entry conditions that are too broad. A flow triggering on any visit produces volume and little revenue. The basket threshold and recent-buyer exclusion are not refinements: they are what make the flow profitable.

Overlap. A contact receiving cart abandonment and browse abandonment on the same day has received two messages on the same subject. Overlap rate : the share of contacts receiving more than one flow in 48 hours, is the saturation metric almost nobody tracks.

One message for different situations. A first-time buyer and a loyal customer abandoning the same cart do not have the same objection. A conditional branch costs little and changes the result.

Data: what separates a good programme from an average one

A replenishment flow needs consumption cycle by product. A post-purchase recommendation needs category affinity. A gift-buyer exclusion needs checkout to pass the information through.

None of this arrives by default. It is the part of the work brands most underestimate, and the one that caps performance when skipped : see Klaviyo custom properties and the Klaviyo events API.

What automation does not fix

Worth stating early, because it is where most budgets go wrong.

A weak offer. Automation sends the same proposition faster and to better-chosen people. If the proposition does not convert when a human sends it, sequencing it changes the volume of the disappointment, not its rate.

A silent site. Flows trigger on events. A tunnel that does not send the viewed category, the cart contents or the purchase does not become segmentable because you bought a platform. The work is upstream, in the tracking, and no amount of scenario design substitutes for it.

An unowned programme. Flows drift. Promo codes expire, catalogue references disappear, entry conditions stop matching a segment that has been redefined. A programme nobody reviews quarterly degrades quietly, and its revenue decline gets blamed on the market rather than on a welcome sequence pointing at a discontinued product.

The common thread is that automation multiplies what exists. It is an amplifier, which is exactly why it is worth installing on something that already works, and why it disappoints on something that does not.

Measuring incrementality

Native attribution systematically overstates. It credits the last clicked message with sales that would have happened without it.

The reliable measure is the holdout: a sample excluded from the flow, whose revenue is compared to the exposed group. The difference is the revenue actually created. The framework is in email marketing attribution.

The result often leads to cutting flows rather than adding them, which is usually good news for the list.

FAQ

What is marketing automation?

Sending messages triggered by behaviour or data rather than by a human decision on a chosen date.

How does it differ from email marketing?

Email marketing names the channel, automation the trigger logic. We target roughly fifty-fifty between the two.

How many flows do we need?

Seven cover the essentials, starting with checkout abandonment.

Do we need high volume?

No. Flows run on existing traffic, not extra sends.

How do we know a flow actually earns?

By holdout. Native attribution credits the last message with sales that would have happened without it.

Going further

Full architecture scoping is on our marketing automation agency page.

Review your email and CRM stack with Deliver