The 3 Least Useful Metrics in Analytics and What to Check Instead

“What should I do with this analytics report?”

People frequently ask me how they should use this or that analytics report. Usually, I have no clue and there are many reports that I never review. Google Analytics alone has 100-plus reports, some of which I’ve never seen.

But I’m probably not missing out.

I never start with a report and then try to find an insight. I start with a question and use the report that has the answer.

Think of it this way: In the bottom of one of my kitchen drawers, I have these scissors for cutting pizza:

Imagine walking into your kitchen, grabbing these and then looking for a way to use them. We would never do that, but instead we reach for the tools when we need them.

I’ve compiled a list of the tools I never reach for because, to me, they are the least useful. (Call them the pizza scissors of Google Analytics.)

New vs. Returning

There’s bad news in this report for everyone. You can see it and say: “Why are so few visitors coming back?” Or you can say: “Why aren’t I attracting more new visitors?” You could lament the numbers either way.

• If returning is low, you’re not triggering loyalty or creating a memorable brand.
• If new visitors is low, you’re not growing an audience and expanding brand awareness.

Because most people don’t have access to any account but their own, they have no point of reference. After I looked at a dozen or so accounts, it seemed that 8-12% of returning visitors is typical for sites with high-ranking content, while 12-20% is typical for sites without search-optimized content. This is because sites with lots of high-ranking content tend to attract lots of new visitors, changing the ratio and pushing down the returning visitors percentage.

There are also accuracy issues in this report, as it suffers from the “one visitor on two devices” problem. If I visit you from my laptop, then return from my phone, I’m two new visitors rather than one returning visitor.

I could live a long, happy life without analyzing this report. It’s at the top of my probably-will-never-need-it list.

A useful alternative: All Channels Report

Here you can view the breakdown of where people are coming from and the conversion rate for each of the default channel groupings.

Bounce Rate

Controversial, I know, but I pay very little attention to bounce rates for two reasons:

1. I don’t do any paid marketing, so I’m never optimizing pay-per-click landing pages. Bounces don’t cost me any money.
2. I have a lot of articles that rank high for information-intent key phrases. These visitors bounce at super high rates, which is fine with me. They came for an answer, found it and left.

I realize that bounce rates are important to some people and for some pages. But when people look at aggregate numbers for an entire website and panic, we need to explain that not all bounces are the same.

Here are bounce rate benchmarks based on 500+ Analytics accounts. You can see that average across all the websites and all traffic sources is 61%.

So yes, if the traffic source is paid search, you should be panicking if your bounce rate is 80%. But if the traffic source is social media, those visitors don’t have strong intent. A bounce rate of 70% isn’t really off the charts.

A useful alternative: Time on Page

Check the average time on page for a given page. Although it has some of the same accuracy issues (Analytics doesn’t know time on page for one-page visits), it’s a better indication of whether the content is connecting with your audience. You’ll quickly notice which pages and topics get the visitor to linger.

Behavior Flow/Users Flow

Note: These are similar, so I’ll combine them. Behavior Flow shows top page paths (with or without events included) and the Users Flow shows visitors’ paths (no events can be included).

These might be the prettiest reports in Google Analytics, but they’re difficult to read and very hard to pull insights from.

It’s an impressive feat of engineering. On one screen, you can see how people flow from page to page from any traffic source, landing page, event, etc. It also shows drop-offs from each page in the flow. It’s an attempt to show everything in Analytics on one screen—but it’s just too much.

These reports are also buggy. If you click on the green box for any page, you get a menu. The “Highlight traffic through here” works, but the “Explore traffic through here” almost never works. It’s supposed to remove everything but traffic that flows through that page.

These reports show a very high-level look at the top highways and traffic paths. If you didn’t know this already, that could be useful, but I’ve never done any detailed analysis from this.

A useful alternative: Navigation Summaries

Go to the specific page in the All Pages report and choose Navigation Summary. Here you can see, at the page level, where visitors came from and where they went. It’s an incredibly useful report, answering some of the most important questions about user experience:

• What is the most-clicked item in your navigation?
• Which items in your navigation never get clicked?
• Are there little things that get clicked a lot?
• What percentage of visitors click on the big call to action?

If the biggest, most prominent navigation item is clicked on by less than 1% of visitors to your homepage, you have some options: Relabel it, move it or remove it.

It’s About Analysis, Not Reporting

This is the fundamental difference between good and great marketers. Some believe that reports are inherently useful, but they’re not. A report is only valuable if it has a specific utility in that moment and if it’s applied for a purpose.

Reports don’t affect marketing outcomes, only actions do. Put down the pizza scissors, ask a question, find an answer and take action.