23 April 2026 , Blackgate
Marketing automation only pays off when it is tied to measurable outcomes and clean data flowing to and from your CRM. Buying a platform first, then retrofitting strategy, is how teams end up with impressive logins and thin results. Below is a practical order of decisions for implementation - whether you are upgrading from ad hoc email or replacing a legacy stack. For platform-specific delivery, our marketing automation work usually sits alongside CRM and measurement work so handoffs stay honest.
Start with outcomes, not feature lists
Define what must change in the business: speed to lead, nurture coverage, win rate, retention, cost per qualified opportunity. Those goals decide which journeys to automate first and which metrics live on the dashboard. If you cannot name the success metric for phase one, pause tool selection until you can - otherwise every vendor demo will look equally compelling.
Map current steps: who owns lists, how leads enter the CRM, what sales actually does with a “marketing qualified” record. Automation magnifies whatever process you already have; it will not fix a broken definition of a lead or a sales team that ignores scored leads.
Choosing a platform you can live with for years
Vendor comparisons often turn into feature matrices nobody will use after year one. More useful questions: who will operate journeys in month six, what breaks when the CRM schema changes, and what the invoice looks like after you double the contact base.
Integration - Native or well-supported links to your CRM, ad platforms, and website matter more than a long feature grid. Broken or custom-only syncs become manual reconciliations and distrust between marketing and sales.
Adoption - If only one power user can build journeys, you have a bottleneck. Balance depth with who will operate the tool after launch - and budget for training, not only implementation hours.
Scale - Volume of contacts, emails, and multi-channel steps should match your growth curve. “We will grow into it” often means paying for unused capability or hitting limits during a campaign spike.
Cost - Understand seats, contacts, API calls, and overage rules. The cheapest headline plan is expensive if you pay three times for add-ons and contractor time to work around limits.
Governance - Role-based access, audit trails, and approval steps reduce the risk of a well-meaning clone of a journey that emails the wrong segment.
Data, CRM, and one source of truth
Automation is only as good as the fields and lifecycle stages you trust. Before you turn on complex scoring:
- Align lifecycle definitions with sales - what counts as MQL, SQL, and closed-lost for reporting.
- Run a dedupe and field clean-up pass; scoring junk doubles down on bad records.
- Document consent and lawful basis for messaging - especially for email and SMS - and keep proof in the CRM or your marketing tool as your team agreed.
For a focused read on duplicates, consent, and deliverability, see CRM and marketing automation hygiene. The point in implementation planning is still: do not automate across dirty data and expect improvement.
Implementation that survives go-live
Break the project into phases with clear owners: discovery, technical setup, data migration rules, UAT with real scenarios, training, hypercare. Sales and customer success should sign off on sample lead paths, not only marketing.
Reserve time for test sends, failure paths (bounced, unsubscribed, opted out), and edge cases - e.g. existing customers who fill a “new lead” form. Skipping this is how you email customers like prospects or double-notify the same person.
After launch, review weekly, then monthly: deliverability, engagement, conversion by source, and CRM match rates. Optimise with A/B tests on one variable at a time - subject lines, offers, timing - with a hypothesis tied to your original goals.
Security and compliance as design inputs
Pick tools that fit your retention, access, and subprocessors requirements - especially if you operate in regulated sectors or across regions. Marketing data is still personal data; treat integrations and API keys with the same care you would for any customer-facing system.
Frequently asked questions
What is marketing automation?
Software and workflows that execute and measure multi-step marketing across channels - often email, forms, and ads - with less manual repetition, using shared data with the CRM where possible.
Why plan before buying a tool?
So success measures and CRM reality drive the stack - not the other way around. Otherwise you configure expensive campaigns on top of unclear ownership and bad data.
How do we pick the right platform?
Match integrations, ease of use for your team, scale, total cost, and governance to your goals and technical environment - then shortlist and trial with real scenarios.
What derails implementation most often?
Weak data, no sales alignment on lead stages, under-training operators, and untested edge cases for consent and unsubscribes.
Is this the same as CRM hygiene?
Overlapping. Implementation is the programme to stand up automation end-to-end; hygiene is ongoing quality of records, consent, and deliverability - see CRM and marketing automation hygiene for that ongoing layer; both need attention.