A well-segmented email loyalty program can lose 30% of its deliverability in 3 months without a single content change. Lifecycle managers keep running into this, and they often can’t explain the drop to leadership. Relationship marketing covers the actions that maintain a long-term relationship with an already acquired customer, as opposed to transactional marketing, which targets a one-off sale. On the email channel, this strategy collapses as soon as part of the list contains dead addresses: every hard bounce damages sender reputation before the personalized message even has a chance to be opened.
Relationship marketing, transactional marketing: same address, two goals
The term relationship marketing goes back to 1983, when researcher Leonard Berry first theorized it in the services sector, well before email automation made it operational at scale. Transactional marketing responds to an immediate action: order confirmation, password reset, invoice sent after payment, shipping notification. Relationship marketing aims for a longer horizon. It builds brand preference through messages that expect no immediate purchase in return: birthday email, content newsletter, review request after a purchase, re-engagement message after a period of inactivity. Automation blurs the line. A flow triggered by an event (signup, abandoned cart, first purchase anniversary) stays transactional in its trigger but relational in its intent, as long as it aims to strengthen the relationship rather than force an immediate conversion. The CRM sits between the two logics: it feeds automated sequences with behavioral data (last purchase, open frequency, engagement score, time on the list) that distinguish a well-targeted relationship message from a generic send. A relationship email built on brand storytelling reinforces this long-term preference, but only if the message reaches an inbox still willing to read it.
The customer lifecycle, backbone of the sequence
A customer never receives the same email at the same time. The email customer lifecycle breaks the relationship into stages: prospect, new customer, active customer, customer at risk of disengagement (often defined as 90 days without an open), lost customer. Each stage calls for a different type of message. The resulting segmentation often relies on RFM analysis, which ranks customers by the recency and frequency of their purchases, as well as their average value: a customer who orders every 30 days has a different profile from a one-time buyer. The welcome email remains the first automated touchpoint of the relationship sequence. How you introduce yourself in that first message sets the tone for the entire relationship to come, even before the first loyalty program or first newsletter.
Few teams map out these stages alone. Growth or CRM teams often run a group workshop with sticky notes on a whiteboard to define the segments and triggers for each flow. In a CRM like HubSpot, these flows then live under Automation > Workflows, before being connected to the sending tool.
What loyalty delivers, when the message actually arrives
The math that justifies investing in relationship marketing is well known. According to the 2025 edition of Brand Keys’ Customer Loyalty Engagement Index, acquiring a new customer now costs 15 to 22 times more than retaining an existing one, a gap that has widened 20% since 1997 (Sens du client, 2025). Yet the average global inbox placement rate caps out at 83 to 84%, according to the Email Deliverability Benchmark Report published by Validity in March 2025, far from the 100% this math implicitly assumes. The ratio keeps improving, but it assumes a rarely stated condition: the message has to arrive. A perfectly targeted loyalty email that lands in spam delivers nothing, whatever the theoretical ratio. The levers behind building customer loyalty through email only matter if the list actually reaches the inbox. That’s what weighs on customer lifetime value (LTV). The same Brand Keys study puts the cumulative profit gain per customer at 86% for just 5 additional loyalty points, a gain that assumes the relationship sequence actually reaches the inbox for the entire length of the relationship.
“One in six legitimate marketing emails never reaches the inbox.” (Validity, Email Deliverability Benchmark Report, 2025)
| Scenario | Relative cost per useful contact | Share of the list actually reached |
|---|---|---|
| Acquiring a new customer (advertising, prospecting) | Baseline, 15x to 22x the cost of retention (Brand Keys, 2025) | Varies by channel, outside the scope of email deliverability |
| Relationship email on a never-verified list | Looks low, diluted by unhandled hard bounces | 83 to 84% global average, one in six emails disappears before the inbox (Validity, 2025) |
| Relationship email on a list verified before sending | Added cleaning cost, no budget wasted on dead addresses | 10 to 12 extra points of inbox placement under a 1.5% bounce rate (Validity, 2025) |
The column on the right shows why this matters. A 1-to-20 acquisition-versus-retention ratio means nothing if a fifth of relationship messages disappear before reaching an inbox.
The prerequisite most guides skip
None of the general guides on relationship marketing mention the technical mechanics that precede an email being opened, mechanics inherited from the SMTP protocol standardized as early as 1982 in RFC 821. It comes down to 4 steps.
- An address collected at a trade show, through a form, or via a CRM import becomes invalid over time: job change, expired domain, a typo that was never corrected.
- When a relationship email is sent (loyalty program, anniversary reminder), the recipient’s server returns a hard bounce, often with an SMTP error code such as 550 5.1.1, what the standard calls an Enhanced Status Code, a sign of a permanent rejection logged in an NDR (Non-Delivery Report). A soft bounce, on the other hand, returns a temporary code such as 450 4.2.1, not to be confused with the definitive 550 5.1.1 rejection.
- The mailbox provider (Gmail, Outlook, Yahoo) adds up these rejections and lowers the sender reputation of the IP or domain: the alert threshold sits around a 2% bounce rate on a single send. That’s far above the average measured across all industries, 0.21% hard bounce and 0.70% soft bounce according to Mailerio, in its 2025 benchmark.
- Subsequent messages, including those sent to active, engaged contacts, get filtered into the “Spam” tab or silently absorbed, after a series of deferrals that often precedes silent blocking.
Filtering bounces after the fact fixes nothing. The moment the recipient’s server responds with a 550, the rejection is already logged against sender reputation. SPF, DKIM, and DMARC authentication has been required by Google since February 2024 for senders of more than 5,000 emails a day: a domain whose reputation is already damaged by accumulated hard bounces risks outright rejection, more severe than simply being marked as spam. Upfront verification, before sending, is what keeps this negative signal from happening in the first place.
The metrics that drive a relationship sequence
4 metrics are enough to tell whether a relationship sequence is sinking before the sales numbers show it. The hard bounce rate on each send, the Postmaster Tools view on the Gmail side (or its Microsoft equivalent, the SNDS dashboard, known by its full name Smart Network Data Services), the complaint rate, whose watch threshold sits around 0.1% at Gmail, and finally the CTOR (click-to-open rate) once the list has been cleaned. The last one confirms that the audience actually reached engages more, at the same send volume or even a lower one. The Google Postmaster Tools dashboard splits into two clearly separate views, “IP reputation” and “Domain reputation,” which never move together; this is what deliverability practitioners call sender reputation. The complaint rate, meanwhile, comes through the “Feedback loop” report provided to high-volume senders.
List cleaning doesn’t solve everything, and that’s a real limitation. Past a certain history of spam traps or complaints accumulated on an IP or domain, reputation takes several weeks to recover even after a complete list purge. After a major purge, a gradual IP warmup, reintroducing volume in stages over 28 to 42 days rather than all at once, limits the risk of reproducing the initial collapse, according to the usual recommendations from routers like Mailjet or SendGrid. Upfront verification keeps this debt from getting worse; it doesn’t erase it retroactively.
A poorly fed welcome sequence
A lifecycle manager who launches an automated welcome flow (Klaviyo flows or equivalent) on a never-verified list often discovers the problem after the fact. The culprit: 8 to 10% dead addresses accumulated since a trade show lead import, and an anniversary reminder campaign that ends up in the “Promotions” tab instead of “Primary.” List verification plugs in ahead of this existing automation, a one-off pass before sending, without replacing the CRM or the sequencing tool already in place in the stack.
Running a sample of the list through a check before the next loyalty reminder remains the only way to know whether the message has a real chance of being read.
A personalized message never makes up for an address that never existed.
