Measuring marketing channels and common mistakes

Comparing channels fairly is harder than choosing one. Each platform reports in its own way and likes its own numbers. This page sets out how I would measure all four on the same footing, and the mistakes I would watch for. It follows on from the launch checklist, and the first choice is in which marketing channel a UAE business should do first.

One definition for every channel

Pick the action that matters to your business, and count it the same way everywhere. Google defines a conversion as an ad click or other interaction that leads directly to a behaviour valuable to you, such as a purchase, a newsletter sign-up, a phone call or a download, and says conversion tracking helps you see which ads, listings and keywords are most successful^[4]. The same idea works outside Google Ads. If a form submission is your action, count form submissions for SEO, ads, social and email, from your own records, so that no channel is measured on a friendlier number than another.

Tell the channels apart

Your analytics can only separate channels if the links carry labels. Google Analytics says that adding campaign parameters to a link lets you see which campaigns refer traffic, that you should always use utm_source, utm_medium and utm_campaign, and that missing parameters show as “(not set)”^[5]. It also warns that values are case-sensitive. Write the names down in one place and reuse them.

When to look

Judge each channel on a timescale that fits it. Google says some SEO changes might take effect in a few hours and others could take several months, and suggests waiting a few weeks to assess whether your work helped^[1]. When you hire for SEO, Google suggests asking what results to expect and in what timeframe, and how success is measured^[2]. Ask the same of a paid or social provider. Agree the date and the number in advance, so you are not choosing the yardstick after seeing the result.

Common mistakes

• Judging SEO in its first weeks, or paid ads on one day. The time Google describes for SEO is wide, and paid spend varies by day, so a short window proves little in either direction.
• Counting clicks instead of the action that matters. A click is not a customer. Count the action you named.
• Leaving search terms unchecked. By default a Google Ads keyword can match searches related to it, not only the exact words^[3], so read what people actually searched before you judge a result.
• Trusting a provider’s report without a way to check it. Ask which figure in your own records it should match.
• Starting every channel at once. You cannot tell which one worked, and each is too thin to learn from. This is my view, not a platform rule.

What to do with the result

Once you have a number for each channel on the same footing, put your own scores back into the channel planner. The measurement criterion is one of its five, and a channel you can now measure should score higher than it did when you were guessing. I describe my analytics set-up work on my own site; I sell it, so check what I say against Google’s own help pages.