Most email marketing guides treat the topic as a content and design problem: which subject line, which template, which send time. For a business running a CRM underneath its marketing, the harder problem sits one layer down, in how the sending platform, the contact record, and the sales team’s own outreach stay in agreement about who a person is, what stage they are at, and whether they should be receiving anything from you at all. Get that layer wrong and no template will save the campaign. Get it right, and the rest of this guide becomes far easier to execute.
Why Email Marketing Setup Belongs to RevOps, Not Just Marketing
In most organisations that have grown past a handful of staff, the email marketing tool and the CRM are owned by different teams, and each treats itself as the source of truth. Marketing exports a segment, sales works a list in the CRM, and neither system knows what the other just did. The common failure modes are predictable once you have seen them a few times: a contact who unsubscribed inside the email platform keeps receiving one-to-one sales sequences because the unsubscribe never wrote back to the CRM; a lead who was marked closed-lost six months ago keeps getting nurture emails because the email tool’s segment logic runs on a stale export rather than a live CRM field; a prospect converts from a paid campaign but the email platform and the CRM disagree about which channel gets credit.
None of these are content problems. They are record-of-truth problems, and they belong to whoever owns the operating model connecting the two systems, which in most organisations is RevOps or sales operations rather than the marketing team alone. Treating email setup as an architecture decision first, and a copywriting decision second, is what separates a list that compounds in value from one that quietly degrades every quarter.
Building the Core Architecture: List, Sender Identity and CRM Sync
Before choosing templates or subject lines, three foundational pieces need to be in place: the domain sends mail with proper authentication, the list only contains people who should be on it, and engagement data flows back into the CRM record rather than staying trapped in the email platform’s own reporting.
Setting Up Domain and Sender Authentication
Three DNS records do the actual work of proving a message is legitimately from you. SPF lists which mail servers are authorised to send on behalf of your domain. DKIM attaches a cryptographic signature to each outgoing message so a receiving server can confirm the content was not altered in transit. DMARC sits on top of both, telling receiving mail servers what to do when an SPF or DKIM check fails, and sending you reports when it happens. None of the three is optional if you want consistent inbox placement, and all three need to point at the same sending infrastructure, which matters because many teams add a new email platform without updating DMARC alignment, leaving the new sends failing authentication silently.
If your CRM has its own email-sending capability, check whether it uses your domain directly or a shared sending domain behind the scenes. HubSpot, for example, documents its own domain authentication requirements for connected sending domains in its developer resources, and the setup steps differ depending on whether you send transactional mail, marketing mail, or both through the same connection (developers.hubspot.com).
Keeping List Hygiene and Suppression Logic Under Control
A list grows dirty in predictable ways: role-based addresses that were never a real person, contacts who bounced hard on send three and were never removed, and people who unsubscribed in one system but remain reachable through another. Each of these damages sender reputation, because inbox providers weight complaint rate and bounce rate heavily when deciding whether your next message lands in the inbox or the spam folder.
Suppression needs to live as a CRM-level rule, not an email-platform setting, so that every tool sending on your behalf (marketing sequences, sales cadences, transactional receipts) reads from the same list. In the UK, direct marketing by email is also governed by the Privacy and Electronic Communications Regulations, which set consent requirements separate from GDPR; the ICO publishes guidance for organisations on what counts as valid consent and what exemptions apply to existing customers (ico.org.uk). Building suppression logic around documented consent status, rather than assuming everyone on a CRM list has opted in, avoids both deliverability damage and regulatory exposure.
Syncing Email Engagement Back Into the CRM
Opens, clicks and bounces are only useful if they change what happens next in the CRM. A contact who clicks through from three nurture emails in a week should trigger a different sales motion than one who has not opened anything in two months, but that only happens if the engagement event writes back to a CRM property or triggers a workflow. Native integrations (HubSpot’s own marketing email, Salesforce Marketing Cloud connected to Salesforce CRM) handle this natively. Bolt-on platforms such as Klaviyo, Mailchimp or ActiveCampaign need either a native app integration or a middleware layer, and workflow automation tools like n8n are commonly used to build that bridge when no off-the-shelf connector covers the exact fields you need to sync (docs.n8n.io).
The diagram below shows the loop this creates once it is built correctly: a CRM contact record feeds a segment, the segment gets sent to, engagement events come back through a webhook, and the CRM lifecycle stage updates as a result, which in turn changes who is in the next segment.
Choosing a Platform: What Actually Differs Between the Tools
The meaningful split is not between well known and lesser known tools, it is between platforms that sit natively on top of your CRM’s contact record and platforms that maintain their own separate list. HubSpot’s marketing email and Salesforce Marketing Cloud both write and read from the same record the sales team uses, which removes an entire category of sync failure but can mean less mature email design and deliverability tooling than a dedicated email service provider.
Dedicated platforms such as Klaviyo, Mailchimp or ActiveCampaign generally offer stronger e-commerce behavioural triggers, more granular send-time optimisation, and better template design tools, at the cost of needing a reliable, tested integration back to the CRM. Apollo.io and similar sales engagement tools sit in a third category again, built for sequenced one-to-one outreach rather than bulk marketing sends, and they need to respect the same suppression list as the marketing platform or a prospect can end up contacted by both simultaneously with contradictory messages.
When evaluating a platform, test the integration before testing the template editor. Ask what happens to a CRM record within minutes of an unsubscribe in the email tool, what happens to an email-platform segment when a CRM lifecycle stage changes, and whether bounce and complaint data is visible anywhere sales can see it before they call the same contact. A platform with a beautiful drag and drop editor and a weak or delayed sync will cost you more in list damage than it saves in production time.
Matching Email Types to the Stage They Actually Serve
Each email type below solves a distinct problem and fails when applied to the wrong stage of the relationship. The trigger condition matters more than the copy.
Sales and Outbound Emails
Outbound sales sequences work best against a tightly defined ideal customer profile and a small number of distinct pain points addressed across the sequence rather than one message repeating the same pitch with different wording. Many practitioners structure a sequence across roughly three to four weeks with a handful of touches, spacing them out enough that each message can stand on its own rather than reading as a repeat of the last. Keep messages short enough to read cleanly on a phone screen, since a large share of business email is opened on mobile first, and always include a working unsubscribe or opt-out path, since its absence is itself a signal spam filters weight against you, separate from any regulatory requirement to include one.
Nurture Sequences
Nurture sends exist to keep a prospect informed and engaged between the point of initial interest and the point of sales readiness. The trigger should be a CRM lifecycle stage change (for example, moving into a “marketing qualified” stage) rather than a manual list export, because that is what keeps the sequence from firing on people who have since moved to a different stage, including people who became customers or went into a lost stage in the interim. Content should map directly to what the person has already shown interest in based on the pages or resources tied to their CRM activity timeline, rather than a generic monthly newsletter sent to the entire nurture segment regardless of interest.
Re-engagement and Win-back Emails
Re-engagement sends exist for one operational reason beyond reviving interest: protecting sender reputation. Inbox providers weight recent engagement heavily when deciding inbox placement, so a large block of contacts who never open anything drags down deliverability for your entire list, including the sales team’s cold outreach if it shares infrastructure. Define disengagement with a fixed, defensible window, run one or two re-engagement attempts, and suppress anyone who still does not respond rather than continuing to mail them indefinitely on the hope they return.
Cart Abandonment and Behavioural Triggers
Behavioural triggers such as cart abandonment depend entirely on event data reaching the email platform in near real time, typically via a webhook from the e-commerce platform carrying the specific items, quantities and cart value, not just a flag that a cart exists. Timing matters more than content here: a message sent within the first hour, while purchase intent is still warm, outperforms the same message sent the next day, and Klaviyo in particular is built around this kind of event-triggered send for Shopify and similar platforms.
Testing Templates With a Method Instead of a Guess
Template testing fails most often not because teams skip it, but because they change more than one variable at once and then attribute the result to the wrong cause. A proper test changes a single element, such as subject line, sender name, or call-to-action wording, while holding everything else constant, and runs long enough to gather a meaningful sample before declaring a winner. Stopping a test the moment one variant pulls ahead, rather than waiting for the planned sample size, is one of the most common ways teams draw the wrong conclusion from an A/B test, because early leads frequently reverse as more of the list is exposed.
Beyond subject lines, test the template’s actual rendering across clients, since a design that looks correct in the platform’s preview pane can break in Outlook’s rendering engine or collapse awkwardly on a narrow mobile viewport. Keep a healthy ratio of text to images, since image-heavy emails with little actual text can read as suspicious to spam filters and also fail entirely for recipients who have images blocked by default. Always send a plain-text fallback alongside the HTML version, both because some recipients’ clients render plain text by preference and because a plain-text alternative part is itself a signal of a well-formed, legitimate message.
Protecting Deliverability: The Infrastructure Work Nobody Budgets For
A new sending domain or subdomain has no history with inbox providers, and sending full volume from day one is one of the most reliable ways to land in spam. Warming a domain means starting with small, highly engaged segments and increasing volume gradually over several weeks, allowing the sending reputation to build before the full list is exposed to it.
Separating sending purposes onto distinct subdomains is worth the initial setup effort: transactional mail, marketing nurture sends, and sales outreach each carry different volume and engagement patterns, and if they share a root domain, a reputation problem in one (a nurture campaign that trips spam complaints, for instance) can drag down deliverability for the others. Bounce handling needs the same rigour as suppression: a hard bounce (invalid address) should remove a contact from future sends immediately, while a soft bounce (temporary delivery failure) should only trigger removal after repeated failures, since treating every soft bounce as permanent strips genuinely reachable contacts from the list unnecessarily.
Measuring Results Beyond Open Rate
Open rate has become an unreliable primary metric because Apple’s Mail Privacy Protection pre-fetches images for a large share of iCloud and Apple Mail recipients, registering an open regardless of whether a human actually read the message. Click-through rate remains a stronger signal since it requires an actual action, and for sales sequences, reply rate matters more than either.
The metric that ties email performance back to revenue is pipeline influenced or created, tracked through the CRM rather than the email platform’s own dashboard. That requires the original source and any subsequent touches to be logged on the CRM opportunity or deal record, typically through UTM parameters captured on click-through combined with the CRM’s own attribution fields, so that a deal closing months after a nurture sequence can still be traced back to it. Reporting solely from inside the email tool, disconnected from what the CRM shows for pipeline and closed revenue, is the most common reason RevOps and marketing teams disagree about whether a campaign actually worked.
Related Reading
Frequently Asked Questions
Do we need a dedicated email platform if our CRM already sends email?
Native CRM email tools such as HubSpot’s marketing email cover most nurture and lifecycle use cases without adding a second sync point. A dedicated platform earns its place when you need e-commerce behavioural triggers, advanced deliverability tooling, or send volume that outgrows your CRM’s native limits, in which case the integration between the two systems needs the same design attention as the emails themselves.
How long should a contact go without opening or clicking before we suppress them?
There is no universal number, but most inbox providers judge your reputation on recent engagement, so a fixed inactivity window such as ninety or one hundred and twenty days is easier to defend than an unstated one. Set the window in the CRM as a lifecycle rule rather than inside the email tool alone, so sales sequences respect the same suppression list as marketing sends.
Why have our open rates looked inflated for no obvious reason?
Apple’s Mail Privacy Protection pre-fetches images for a large share of iCloud and Apple Mail recipients, which registers as an open even when nobody has read the email. Click-through rate, reply rate, and CRM-attributed pipeline are more reliable signals once a meaningful share of your list uses Apple Mail.
Should sales outreach and marketing nurture emails send from the same domain?
Keep them apart. A high-volume nurture send that trips spam complaints can drag down deliverability for the sales team’s one-to-one outreach if they share a root domain, so a dedicated subdomain for each sending purpose limits how far a reputation problem spreads.
What is the practical difference between SPF, DKIM and DMARC?
SPF lists which servers are allowed to send on your domain’s behalf, DKIM attaches a cryptographic signature so receiving servers can confirm the message was not altered in transit, and DMARC tells receiving servers what to do when SPF or DKIM checks fail, and sends you reports when they do. All three need to be configured together for authentication to actually protect your domain.
For more on this, see more RevOps strategy posts, including Revolutionise Your Email & SMS Marketing Strategy: Why Businesses Should Leverage Omnisend, Accelerating Enterprise AI Sales with Privacy-Preserving Security, and Maximize Your Sales Efficiency: How a Consultant Can Transform Your Sales Process.
Leave a Reply