Five things decide whether marketing automation will work: whether your team can read the data, whether the data is right, whether campaigns are consistent, whether someone owns automation, and whether your tools match your team.
The Campaign Automation readiness score measures five dimensions. Each one captures a different reason automation succeeds or fails, and they are deliberately not weighted equally. Understanding them is useful whether or not you ever take the audit, because together they form a practical checklist for whether an organisation is ready to automate.
1. Democratization — 30%
The question: can the people running your campaigns act on data without queuing for an analyst?
In many organisations, answering a simple performance question means raising a request with one person who knows how to pull the data. That person becomes a bottleneck, and decisions wait.
What good looks like: channel owners can see live performance in one place, understand it, and act on it themselves. Ideally, they are alerted when something changes rather than having to go looking.
Why it carries the most weight: automation optimises, but humans steer. If the humans cannot read the instruments, they cannot tell whether the automation is doing the right thing — so every automation you add creates a new bottleneck rather than removing one.
2. Instrumentation — 20%
The question: is your measurement reliable?
What good looks like: conversion tracking has been audited recently, duplicates have been removed, conversion actions are documented, and there is a view of performance that goes beyond last-click.
Why it matters: automated bidding and automated rules act on whatever signal you give them. Gaps in tracking do not produce cautious automation. They produce confident automation pointed the wrong way. This is covered in the four layers of ad waste.
3. Standardization — 20%
The question: do your campaigns follow a consistent structure and naming convention?
What good looks like: a documented naming convention — objective, channel, audience, date, variant — that is actually enforced, and a standard process for launching campaigns.
Why it matters: automation acts on patterns. A rule that says “pause any prospecting campaign whose cost per acquisition exceeds target” only works if the system can reliably tell which campaigns are prospecting campaigns. Standardisation is not bureaucracy. It is what makes automation possible at scale.
4. Governance — 20%
The question: is there a clear owner of automation, with defined authority and change control?
What good looks like: one named person is accountable for automation rules, there is a review step before changes go live, and there is a log of every active rule and when it last fired.
Why it matters: when nobody owns automation, rules accumulate, overlap and eventually conflict — usually discovered through a spend anomaly rather than a review. The first governance action is not building a committee. It is naming a person.
5. Team and tooling — 10%
The question: do your tools and your team’s data literacy match each other?
What good looks like: the sophistication of the stack is matched by the team’s ability to use it.
Why it carries the least weight: tools are the easiest part to acquire and the least predictive on their own. What matters is the match. Advanced tools sitting on low data literacy reliably produce shelfware — you pay twice, once for the tool and again for the work it was supposed to remove.
The gate
There is one more rule that matters more than the weights. If democratization scores below roughly a fifth of its possible value, the total score is capped — no matter how strong the other four dimensions are.
The logic is the same as the weighting, taken to its conclusion. If the people responsible for campaigns cannot read the data, nothing downstream can be steered, and excellent tracking, tidy naming and a clear owner cannot make up for it. It is the clearest example in the model of your binding constraint setting your ceiling.
A note on the weights
These weights are a judgement built into the model, based on where automation programmes tend to succeed and fail. They are not a law of nature, and a particular organisation might reasonably weigh things differently. What they express is a view worth taking seriously: the human ability to read and act on data matters more than any individual tool.
Where to go next
The readiness ladder shows what each score range can support. Guardrails that make speed safe turns governance into practical rules. And the readiness audit scores all five in a few minutes.