Guess what – most Google Analytics setups are broken. People are reporting results using data that’s wrong. Organisations are changing things on their website based on the wrong information.

What can be done about it? Plenty.

Here’s the thing: Google Analytics is an amazing tool, but it won’t work ‘out of the box’. Yes, you can add its code to your website and start seeing data coming in straightaway (although it runs up to 24 hours behind). However, there’s plenty you need to do before you start making decisions based on your data.

You might think ‘I don’t really use Google Analytics, so I’ll get round to fixing it another time’. Of course, the best time to fix your data is when you launch your website. But the second-best time is after reading this article. Don’t put it off, because you can’t fix historical data, so you’ll miss out on any insight that you might need further down the line. Perhaps when you hire an agency for marketing, SEO, conversion optimisation, or UX.

Here’s how:

Google Analytics basics

1) Check you have GA code in the right place.
You’d be surprised at how many setups get this wrong. Put it just before the closing </head> in your header, and make sure it’s on all the pages you want to gather data for. To find your code, right-click on one of your website pages, select ‘View page source’ and then search (CTRL+F) for ‘UA-‘ (without the quotation marks).
search box
And make sure it only appears once – duplicate codes will distort your data. Not sure if you’ve got it on all your pages? Use this:

2) Audit who has access.
If you’ve taken over an agency account, it may have had plenty of users granted access over the years. Go to Account > User Management. Delete people who are no longer at your organisation.
shows analytics fields

3) Set the correct timezone
Google Analytics will record data based on the hour/time of the day. So if you want to find out whether your prospects convert better in the morning or afternoon, you need to be looking at the right time. Go to View > View settings:

shows time

Getting cleaner data

4) Keep a live backup.
Create a duplicate view of your data, just in case something goes wrong with your configurations. That way you always have something to refer back to. Use one view for doing all the things that will change and streamline your data. For example, adding filters, goals, events. Keep another view of your data, where nothing is touched. This is your raw data. If you want to be really cautious, set up a third view, just for tests. Click View > Create new view:
showing field to search for data

5) Filter out visits from people who aren’t part of your target audience
This is one of the most common reasons for screwed up data. You may have employees, web developers, SEOs – all clicking around your site, viewing pages, and messing up your reports. You can manually filter out hits from a user’s IP address, but an easier and quicker way is to get your team to add an ‘opt-out’ add-on to their browsers. Use Google’s official extension:

6) Exclude hits from all known bots
The curse of Google Analytics. The sad news is, you can never completely eliminate bots from your data. As fast as Google works to filter out the little critters, new ones appear. So do what you can. Head to View > View Settings, and tick the ‘Bot Filtering’ box:
tick box
It’s also worth knowing how to recognise a bot when it pops up in your data. At least then you can exclude them from your reports. Tell-tale signs include:
– new user
– 100% bounce rate
– Average time on page= 0 seconds

Getting better data

7) Avoiding duplicates
Sometimes Google Analytics records one page twice, such as with and without the trailing slash. For example www.chiefnation/blogs and You can fix this via your htaccess file, or by using a plugin (if you’re using WordPress). However, if you prefer to steer clear of website development/coding, here’s how to group those URLs together using Google Analytics filters. Go to View > Filters > Add Filter. Add the characters (known as regex) to the fields as below:

list of filters

8) Dealing with dynamic URLs
Imagine you use a marketing platform such as HubSpot. You want to email out an account-based marketing report to 10,000 of your subscribers. Within the newsletter you’ve included a link to your blog section ( When you send that email out, HubSpot will automatically add tracking code to the end of the URL (everything from the question mark onwards), unique to each recipient. Below is an edited example of what you’ll see:

long urls

If 500 recipients click on the link, Google Analytics will record 500 different URLs, even though they point to the same page. This makes your reporting look very messy, very quickly. Stop this happening by excluding URL parameters. Go to View > View Settings and then enter the relevant codes in the ‘Exclude URL Query Parameters’ box. For HubSpot you need to add the codes below:

exclude box

9) Check where your data is coming from
Any website that has your Google Analytics tracking code installed will send data to your account. So check your GA code isn’t accidentally pasted on any clone sites, such as your staging, dev or test sites. Do this by going to Audience > Technology > Network. Then click on Hostname as the Primary Dimension:


You’ll see all the websites which are using your tracking ID. Here are the top 10 in Chief Nation’s report:

list of names

Ok, so no test websites show up, great. But hang on, why has Fox News (#7) got Chief Nation’s tracking ID? The answer is… it hasn’t. This is an example of fake – ‘ghost’ – visits. They’re spammers who use the names of real companies and websites to appear legitimate. The URLs at #1, #2, #4 and #8 are all genuine. The rest need to be removed.

The best way to stop this from happening is by adding a filter so that Google Analytics only accepts data from the hostnames you know are related to your website. Go to Filters > Create New Filter > Custom. Then under Filter Field, choose Hostname and in the Filter Pattern field type in each one of the ‘legitimate’ hostnames.

Separate each hostname by | so the example above gives you|||

And finally

10) Avoid looking at averages
What happened when Bill Gates walked into a bar? Everyone became a millionaire… on average. When you review your Google Analytics data, always look beyond the average figures.

Take bounce rate as an example. You might want people to spend more time exploring your website, so you decide to reduce your website’s bounce rate. But hang on a minute. Most ‘contact us’ pages have high bounce rates. People view the page, see the contact details, and then leave the website.

Instead, look at specific pages where you want people to spend more time, such as product pages. Then drill down. For example, you might find that a page has a high bounce rate – but only from iPhone 6 users. That’s where you start fixing things. Going beyond the averages is the only way you’ll really know what’s going on across your website. That, and making sure you can trust your data.

The post Why your Google Analytics data is probably wrong – and how you can fix it appeared first on Chief Nation.