Five Tips for Google Analytics

This blog was posted by Devon Smith on January 4, 2016 as part of the CultureHive Digital Marketing Academy. You can find out more about the project here.

I’ve been a CultureHive Mentor for two years and nearly every conversation I’ve had with folks in cohort 1.0 & 2.0 has touched on Google Analytics in some small way. Whether it’s getting over your reluctance to open up the tool because it seems scary/not useful/too time consuming, or diving into the data you have, or helping to figure out what new data you need to track your experiment, to revising your plans based on the data that’s coming in, data has played a role in our conversations.

I’ve found a few themes frequently come up, all of which might seem counter-intuitive, but should help guide you no matter what data tool you’re using.

Ask your data questions.  
Before you even log in to Google Analytics, generate a list of questions you want to know, or hypotheses you want to test. There is a never ending set of numbers, graphs, segments, filters, reports that you could try to look at in Google Analytics, the best way I’ve found to stay focused and not get lost down the rabbit hole is to keep myself restricted to specific questions. Give yourself a checklist of what you’re looking for, and you won’t get so distracted by the bells and whistles. Ideally, these questions are relevant to your organization, or some problem your organization is trying to solve right now. Sometimes you need to confirm or refute a commonly held belief in your organization (everyone starts on our homepage, nobody donates from their mobile phone, people are only on our site to buy tickets, etc). If you’ve gotten in the habit of reporting your website data regularly, that’s great! But just looking at the same data month-after-month rarely yields actionable results. Pick a new area of investigation each report. Brainstorming this list with other members on your team can be extraordinarily useful, to make sure you’re not just asking interesting questions, but relevant questions. Some questions you might ask your data:

  • What percent of my website visitors are coming from within/around my city? For everyone else, do the pages they’re looking at tell me anything about why they’re on my site?
  • Do most people enter my website via the homepage, or are there other key landing pages that introduce users to my organization?
  • What percent of visitors are on a mobile phone and what actions are they trying to take (buy tickets? donate? read content? etc)?
  • What words are visitors using to search my website, and on which pages are they most frequently using the search box? How could I make that content more prominent, or create new content to fulfill those search requests?
  • Are people downloading the PDF’s that I’ve posted on my website? Of all the people who visit the page that this PDF is featured on, how many actually download it?

Use less data.
It’s really hard to look at all users on your website all at the same time and find anything useful. Start stripping away all of the irrelevant visitors, depending on what your question is, and then dive into the details on the small number of visitors that remain. In Google Analytics, this often means creating Advanced Segments. Who might a segment be that you want to follow to learn more about them?

  • Visitors from social media who also bought a ticket
  • Visitors who look at a performance page, but don’t buy a ticket
  • Visitors who seem to be searching for a job
  • Visitors who read a certain category of blog posts
  • Visitors from government agencies or other funders

Add better data.
Google Analytics is popular because it’s free, it’s incredibly powerful, and it’s pretty easy to set up. But with just a little knowhow, you can add more and better data to your Google Analytics reports. Just a few of those new data points include:

  • How much money you spend on each marketing channel, so you can compare their effectiveness (e.g. you spent $1,000 on that Facebook ad campaign, and those visitors bought $2,000 worth of tickets). See Cost Data Import.
  • How far each visitor scrolls down each page of your site. If all of your important “buy now” “donate now” buttons are at the bottom of your page, what percent of visitors actually see those buttons? See Scroll Depth.
  • How many users are searching for something you provide, but don’t actually make it to your website. Connect your Google Webmaster Tools account to Google Analytics, and watch the data roll in.
  • How many visitors to my site are actually just my staff checking our website? See excluding internal traffic.
  • Which author on our blog gets the most pageviews? Can I inspire a little inter-office competition by making that data known to the rest of the staff? See Custom Dimensions.

Stop visiting Google Analytics (so much).
At first glance, Google Analytics is simply too messy/difficult to interpret for most people in your organization. But sometimes there’s a single report that would be really great to put in front of them on a regular basis. Here you have a few different options to make it easier for them, and thus easier for you!

  • Schedule a report to be emailed on a regular basis. Maybe you want to send your fundraising manager the geographic breakdown of everyone who visits your donation page on a monthly basis. Or you want to send your education manager the top 10 education resources downloaded from your site in the past week. See email reports.
  • Schedule an alert to be emailed to you if an event happens. Maybe you want to get a heads up if your local newspaper starts sending traffic to your website, or if your traffic increased by 20% relative to yesterday, or if the speed of your site suddenly slows down. See custom alerts.
  • Sometimes you want to create a bunch of versions of slightly different reports to lots of different people within your organization. Sometimes you want to create graphs (pie charts, line graphs, bar charts) that just aren’t possible in Google Analytics. Sometimes you just want to give your staff “safe” access to Google Analytics data without forcing them to log into a confusing tool. You can automatically link Google Analytics data to a Google Spreadsheet.

Add insights to your data.
Google Analytics won’t tell you what to do with your data, it just stores all of your data for you. A little like your refrigerator stores your food – you still need a human to pick a recipe, select all the right ingredients, put them together in an interesting way, and then deliver the final dish to you at the table. In any data project, a rough rule of thumb I use is to spend:

  • 10% of my time planning (figuring out – what questions do I need to ask my data, and what data will I need to add to Google Analytics to answer those questions?)
  • 40% of my time diving into the Google Analytics data, configuring reports, creating advanced segments, applying filters, and the such
  • 40% of my time analyzing the data I’ve collected to identify the trends, pick out the bits of data that actually turns out to be useful (don’t forget, sometimes the answers to your data questions won’t be interesting to anyone but you), trying to determine what I should do now that I know all of these data points, and packaging the findings in a way that will be clear and meaningful to other people
  • 10% of my time discussing the findings and recommendations with other people, helping them quickly pick out the most important points

It’s easy as the person who spends the most time with your organizations data to get consumed with all of the interesting things you’re finding, to keep reaching down the rabbit hole to find the next data point, to spin a lot of wheels trying to capture ALL the data instead of just the MOST IMPORTANT data that your team needs. Hopefully these tips help you spend less time, with better results (always be optimizing!).

If you want even more examples about how to implement some of these tips, check out this slidedeck from a recent talk I gave on the Hidden Secrets of Google Analytics.

Leave a Comment

Your email address will not be published. Required fields are marked *