Email & CRM marketing: foundations & core concepts

Every other demand channel is rented. You bid for the impressionimpressionThe total number of times an ad or piece of content is displayed, regardless of clicks. Each display counts as one impression, even to the same person.View full definition →, the platform decides who sees it, and the price moves without asking you. Email and CRMCRMCustomer Relationship Management: software and strategy to manage and analyse customer interactions throughout their lifecycle.View full definition → messaging is the one place where you hold the audience yourself: the addresses, the order history, the consent record. Litmus has put the average return at around $36 for every $1 spent, and no paid channel gets close. The catch is that "owned" stops at the send button. Whether the message reaches an inbox is decided by Google, Microsoft, Yahoo and Apple, who publish their rules and change them on their own schedule. This lesson sets out what you actually own, what sits inside the record behind each message, what consent means both legally and technically, and which rules are not yours to negotiate.

What email and CRM marketing actually means

Owned lifecycle messaging is sending a message to a named person who has given you permission to contact them, timed by something that person did, for a commercial outcome you can measure. Each part of that sentence carries weight. A named person, so you are not buying reachreachThe number of unique people exposed to your message in a given period. Unlike impressions, reach counts each person once, no matter how often they see it.View full definition → from an intermediary. Permission, so the send is lawful and the provider does not treat it as unsolicited bulk mail. Timed by something that person did, which is what separates lifecycle messaging from a newsletter that goes out because it is Thursday.

CRM marketing is the half that makes the timing possible. A CRM (customer relationship management system) holds one record per person, and that record is the real unit of work in this module. A usable record carries four things: an identity (email address, and usually a phone number and a customer ID that ties back to orders), a consent state (what they agreed to, when, and through which form), an event history (pages viewed, carts started, orders placed, tickets opened, messages received and clicked), and a lifecycle stage that says where this person sits in the relationship. Messaging is what happens when a change in the third field, checked against the second, moves the fourth.

The vocabulary you need to follow the rest of the module:

• List: the set of records who have given you permission to send marketing messages. Not everyone in your CRM is on it.
• Segment: a subset of the list defined by shared attributes or behaviour, recalculated as records change.
• Campaign: a message you decide to send to a segment at a time you choose.
• Flow (also called a sequence or journey): a series of messages that starts when a condition is met, and runs per person rather than per calendar date.
• Trigger: the event that starts a flow. A cart abandoned, a trial started, a subscription 25 days old.
• Suppression list: records excluded from a send regardless of segment membership. Unsubscribes, hard bounces, complainers, recent purchasers.

Sub-concept 1: the record is the asset, not the list size

A list of 500,000 addresses with no event history behind them supports one behaviour: batch sending. A list of 50,000 records where each one carries order history and a browse trail supports triggered messaging, and triggered messaging is where most of the revenue in this channel sits. The difference is not list quality in the abstract. It is whether the fields exist.

This is why the CRM and the sending tool have to share one record rather than two copies. When they are separate systems syncing nightly, every trigger fires on yesterday's version of the person. Someone who bought on Tuesday still receives Wednesday's "still thinking about it?" message, because the sending tool never saw the order.

Sub-concept 2: consent is the licence to send, and it is a field you must be able to produce

In the EU and the UK, marketing email to individuals runs on prior consent under GDPRGDPREU regulation governing how organizations collect, store and use personal data, with fines tied to global revenue for breaches.View full definition → and the ePrivacy rules (PECR in the UK): freely given, specific, informed, and recorded, with a narrow soft opt-in for existing customers who bought a similar product and were offered a refusal at the point of collection. Every message needs a working unsubscribe. In the United States, CAN-SPAM works the other way round: no prior consent is required, but the unsubscribe has to be honoured within ten business days, the header and subject line must not mislead, and a valid physical postal address must appear. Canada's CASL sits between the two, on express or implied consent with record-keeping obligations.

The practical version for your stack: consent is not a checkbox in a screenshot, it is five stored values. Source (which form, which page), timestamp, the exact wording shown, the scope agreed (email only, or email and SMS), and the current state. If your team cannot produce those five for any twenty contacts picked at random, you do not have consent, you have addresses.

Sub-concept 3: deliverability is set by the inbox providers, and they publish the rules

Deliverability is whether your message reaches the inbox rather than the junk folder or a rejection. You do not influence it by writing better subject lines. You influence it by meeting requirements the providers write down.

In February 2024, Google and Yahoo began enforcing a common set for senders above roughly 5,000 messages a day: SPF and DKIM authentication, a published DMARC record, one-click unsubscribe in the message headers honoured within two days, and a spam complaint rate held below 0.3%, with Google advising you stay under 0.1%. Microsoft applied comparable requirements to Outlook.com, Hotmail and Live from 5 May 2025 for high-volume senders, first routing non-compliant mail to junk and signalling outright rejection later. The pattern across all of them is the same: authenticate, make leaving easy, keep complaints low.

Two things follow. Reputation attaches to your sending domain and IP, not to your campaign, so one bad send damages the next twelve. And you can see the same data the providers see: Google Postmaster Tools reports your domain reputation and complaint rate, Microsoft's Smart Network Data Services does the equivalent for sending IPs. If nobody on your team has logged into either, you are running the channel blind.

Sub-concept 4: campaigns and flows are different products of the same list

A campaign is a decision you make: this segment, this offer, Thursday. A flow is a rule you write once, which then fires thousands of times without you. Campaigns give you volume and control over timing, and they are how you announce things. Flows give you relevance, because the message arrives while the behaviour is still fresh, and they carry far higher revenue per recipient at a fraction of the send volume.

Most programmes are built backwards: a heavy campaign calendar, a welcome flow nobody has touched in two years, and nothing in between. Which flows to build first, and in what order, is the method question the frameworks lesson answers. What matters here is the definitional split, because the two are measured differently and the same open rate means different things in each.

Where the definitions bite

Microsoft's May 2025 change is the clearest recent example of the boundary of "owned". Senders whose programmes had run for years on an unauthenticated domain, with no DMARC record published, found their mail to Outlook.com and Hotmail addresses filtered to junk. The list was theirs. The consent was real. The message still did not arrive, because a technical condition set by a third party had not been met, and a large share of most B2B lists sits behind Microsoft.

Klaviyo, which sells email and SMS software to well over 100,000 businesses (so read its benchmark reporting with that in mind), builds its whole product around the campaign and flow split described above, and its published benchmarks consistently show flow messages earning several times the revenue per recipient of campaign sends. That is the definitional point made commercially: the same list, two different mechanisms, very different economics.

CMO action items

• Open Google Postmaster Tools and Microsoft SNDS this week and read three numbers: domain reputation, complaint rate, authentication pass rate. If nobody has access, that is your first ticket.
• Ask for the consent record on twenty contacts chosen at random: source, timestamp, wording, scope, state. What comes back tells you whether you have a list or a spreadsheet of addresses.
• Split last quarter's email revenue into campaigns and flows, and put the send volume next to each. The ratio between the two shows you where the programme actually is before anyone proposes a redesign.

Common mistakes that kill results

Mistake 1: treating unsubscribes as the failure metric. An unsubscribe costs you one contact and protects your reputation. A spam complaint damages delivery for everyone else on the list, and the providers have set the ceiling at 0.3% with 0.1% as the target. Making it hard to leave produces complaints instead of unsubscribes, which is the trade you least want.

Mistake 2: confusing an address with permission. A conference badge scan, a purchased file and a form filled in for a webinar three years ago are not the same consent, and two of them are not consent at all. Sending to them costs you sender reputation before it costs you a fine.

Mistake 3: running the email tool and the CRM as two systems of record. When both hold a version of the person, the triggers fire on whichever version is stale, and the customer sees the discrepancy immediately. One record, one consent state, one event history. Everything else in this module assumes it.