Archive of ‘Google’ category
I introduced the Cohort Analysis in an earlier post, noting that it is a Google Analytics feature that lets you “look at a group of users based on a shared acquisition date”. I noted that my preference has been to create segments, but I wanted to explore this beta feature in Google Analytics a bit more in this post and offer some use cases for it. You will still want to spend time getting familiar with segments because they make this feature more useful.
A Cohort Analysis is worth exploring if user retention is a key metric on your site since this feature measures behavior over time. If you see a significant increase in traffic, this helps you determine if a cohort is responsible for the majority of that change. You can also determine if a specific group completed an action on your site after their initial visit.
If you are interested in doing a Cohort Analysis, you can decide the size of your cohort by day, week or month; the metric (the number you want to measure), and the date range (how far back you want to analyze). With a size of day, the cells in your resulting table will each display a single date. With a metric, you can only select one number at a time and these are displayed as columns in your table. And the date range determines how much data will appear and corresponds to the number of rows that are in your table.
Although I primarily look at the tables, you can also view your data in a chart, which is a cumulative metric for your particular cohort. In the below example, you can see a significant drop in pageviews for each user after the initial visit, which could imply there is not enough compelling content that brings people back to the site.
Identifying micro and macro trends
Are you wondering how a specific campaign impacted your sales? Do you also want to know if a visitor from a campaign took action immediately or came back a month later? Since your cohort groups visitors by similar dimensions (such as traffic source), you can view them by a particular source, such as Paid Traffic by selecting that segment.
For this Paid Traffic group, session duration increases in some weeks after they first came on the site. The darker color in each cell indicates the value of the metric when compared to other values in the table.
Behavior by Device
You do not need to be told that a mobile-friendly site is non-negotiable for today’s businesses. But have you wondered how behavior varies on a mobile device? Here we can see that more purchases are made in following months for desktop users than mobile users. It’s worth exploring why that could be.
Analyzing by Channel
With segments you can select multiple channels at a time. Here we can see goal completions for this account’s paid traffic versus organic traffic. Both channels are bringing people back, but organic channels perform better when it comes to the goal completion metric.
One of the more useful features of a Cohort Analysis is that it helps you understand how long it takes for a visitor to make a decision. These views can also help you determine if a remarketing strategy is good for your business. If you notice that traffic declines after a certain day, you may want to start remarketing at that time.
If you discover that a Cohort Analysis is useful for your business, you can save a view as a shortcut, email it, or add it to a dashboard like you would with other reports in Google Analytics. To reiterate, this helps you analyze group behavior on your site based on common attribution. In my opinion, it is still limited since only Cohort Type is Acquisition Date. If Google makes other options available, it will increase the usefulness of this tool. In the meantime, use segments to drill into your data for a Cohort Analysis.
In Google Analytics, you will find a section specifically for AdWords under the Acquisition section. In this section, you will find eleven views that provide information about your visitors coming from AdWords which are described below. (more…)
Since I spend a lot of time in Google Analytics accounts, I can easily get bogged down by the all the data available. Occasionally, I need a reminder to step back and look at the big picture and the overall marketing strategy for each client. What exactly are clients hoping to accomplish and what actions on their web sites will lead to meaningful outcomes? Rather than looking at numbers that are subjectively judged as “good” or “bad”, they need to be viewed in the context of the client’s big picture.
Google Analytics Dashboards are a great tool for my clients who do not like logging into their accounts. They often become overwhelmed with the amount of data in analytics and prefer to see data at a glance.
I have a previous post on how to use the Google Analytics Solutions Gallery for dashboards. The gallery permits you to use dashboards created by other Google Analytics users. But even that can be overwhelming with hundreds of dashboards to choose from.
Below I’ve listed five of my favorites for measuring performance with Google Analytics. Please note that I did start with some dashboards I imported from the gallery and modified them to be a better fit for my reporting needs. Each heading below links to a dashboard you can automatically add to your account as long as you are logged into Google Analytics when you click on the link.
Organic Search Performance
How often are people finding you through organic search? With this dashboard, you can see the behavior of people coming to your site through organic search results. And yes, it probably will primarily be Google for your organic source, but you can see other sources of organic searches such as Bing, Yahoo, or AOL. You can also measure how this source impacts the number goals completed and pages viewed.
Paid Search Performance (AdWords)
Advertisers will definitely want to see how their paid campaigns perform when compared to other channels, such as organic or email. It’s also valuable to drill into paid campaigns to compare them to each other. Are visitors performing a desired action in one campaign more so than the others? Although this dashboard is for AdWords, it can be modified to also view other paid campaigns, such as with Facebook or Bing.
Social Media Performance
When it comes to online marketing, measuring the performance of an ad on the Display Network is similar to measuring the performance of any other traffic source, such as Search Network ads. But what needs to be considered when viewing these metrics is that display ads are often better for branding rather than a direct conversion. When you analyze your data, it is important to keep this in mind when deciding whether your display ads are worth the investment.
In general, ads on the Display Network have a slightly higher bounce rate and fewer pages per visit than other traffic sources. Users who respond to a display ad often were not looking for your product or service at the time so their response is more passive than a visitor who is actively searching for something. As a result, your visitors are somewhat interested in what you have to offer but they are not actively seeking it. As a side note with these ads, you want to write compelling headlines, but not click-bait headlines. With a click-bait headline, visitors click only out of curiosity, not necessarily a genuine interest in what you have to offer so it is probably not worth paying for those clicks.
With the bounce rate, you may see a bounce rate for your display campaigns of 80% whereas the bounce rate for your other sources of traffic is only 50%. Rather than focusing on that single number of 80% for display ads, drill into this and look at the specific placements, ad copy, and and targeting methods. Google Analytics provides this granular data about behavior by each of these dimensions. As you drill into your data, you might find there are several placements that are responsible for the highest bounce rate which you can then exclude in your Google AdWords campaign.
Also, as you review your Google Analytics data, give a little more credit to your engagement metrics, such as pages per session, for people who come from display ads. When people arrive on your site they’re passively searching somewhere else and they spend time surfing your site, it’s a good indicator of the effectiveness of your ads on the Display Network.
The value of this initial introduction, or branding, is seen when someone does make a purchasing decision. Because they saw your ad earlier, they remember what it is that you offer and are willing to visit your site again when it comes time to make a purchase. They may bookmark your site when they reach it through a display ad to revisit at a later date when it is time to buy. Or it could be they may not even remember you when they are ready to buy, but in doing an internet search at a later date, your brand does comes back to mind when they see your results again. Keep these factors in mind when you review the performance of your display ad campaigns.
Of course, everyone who runs paid ads for a site wants to see conversions in AdWords or goal completions in Google Analytics. It makes sense. When you are using cold hard cash in your marketing, you want to see results. But when you look at completed goals with display ads, look beyond just the straight conversion metrics. Make sure you set engagement related goals which include pages per session, mentioned above, or time on site. This is also when it makes sense to have a goal value set for valuable behavior that is not revenue-generating.
Remember that your metrics for display campaigns are going to be different than your metrics on your search campaigns with fewer clear-cut conversions. This is a campaign type that is still worth using because it can attract people who have an interest in what you offer and it could get you to the top of their mind when it is time for them to make a decision about their purchase. So go ahead and try the Display Network but keep these differences in mind when reviewing your results in Google Analytics.
*This blog post was originally published by RSO Consulting.
If you’ve ever looked at your website analytics platform and wondered where your conversions are coming from or who is clicking on your links, then you’re probably going to be a big fan of Google Tag Manager. Not only does it save you time, but it also makes it really easy to gather exactly the types of data you want.
Once your tags are in place, they send information to your website analytics platforms like Google Analytics or Facebook. Let’s say you want to know the number of users who are clicking through to your purchase page, and you want to know which page the clicks are coming from. You can place conversion tracking and analytics tags on your pages, which then sends this very data to your analytics platforms for assessment.
Or let’s say you only want to know when a visitor clicks on a certain button. You can set up your tags to “fire” only when a specific action is taken. Google Tag Manager gives you lots of flexibility to set up your own parameters, so you can get refined data about your website visitors.
It’s also a great accompaniment to Google Analytics. Whereas Analytics is fantastic at delivering statistics about how many people visited your website, which pages were visited or which web browsers were used, Tag Manager lets you dial in on more specific behavioral data. You can even generate user IDs, so you can gather data about specific visitors.
All of the information you can glean from Tag Manager helps you make better marketing decisions, such as where to spend your ad dollars, what type of content resonates with your audiences, and how to improve your website for an even greater user experience.
When it comes to your website, your visitors are the ones who are in charge. With Google Analytics, you can use that data to learn what it is visitors want to accomplish. As you analyze this information, you can see if there’s any disconnect between what somebody is coming to the site for and with the site has to offer them. Here are three strategies to understand visitor intent with Google Analytics. (more…)
“AdWords is a money pit”. I hear that a lot from people who are cynical about their experience with AdWords. They spend a lot of money yet do not see any sales as a result. And I agree that it can be a money pit when it’s not monitored closely. Advertisers who set up an account, then barely look at it will find themselves maxing out their budget very quickly with little to show for those dollars spent. But when your Adwords Budget is monitored and optimized, AdWords has the potential for great results for advertisers – both for online and brick and mortar store. The major concepts to learn and manage so AdWords does not become a money pit is the daily budget, bid strategy, keyword costs, and conversions.
The highest level is your campaign budget. A simple account for a small company may have only one campaign because they do not need to account for variation in geographic locations or language. Others may have a separate campaign for each country or each product or service. Note that you are setting a budget by campaign. If your total spend for the month is $5,000 and you have five campaigns, you might set each campaign at $1,000. If it was one campaign, then you set the one budget of $5,000 for that single campaign. Start here when determining your maximum spend for the month.
Ad Group Budget
Next you can set a budget for each ad group within a campaign. In the below example for a florist, they bid slightly higher on their “Brand Name & City Name” ad group because it produces the most revenue. Less money is spent on RLSA (remarketing lists for search ads) because the return is lower. If you are just getting started, there is no need to make edits yet at the ad group level. You may want to first collect data on how your ads perform in each group.
You can also manage bids for keywords. Below you see a slighter higher bid for the keyword phrase marketing conference compared to a business conference so this advertiser is bidding based on the return each of these words provides. The expectation is that searches on the phrase marketing conference will result in more ticket sales than business conference. If you are just getting started with AdWords, bidding at the keyword level may be more granular than you need initially. Unless you are confident that some keywords will produce a higher ROI than other keywords, allow some time for your account to collect data, then come back and adjust bids at a later date depending on how the words perform.
Now that you have the basics of budgeting down, it’s time to track performance on your ads. While getting a click on an ad is generally a good thing, you do not know how good of a thing it is unless you measure what happens after the click. Previous articles on this blog talk about the importance of tracking conversions and how to set it up.
In Part II, we’ll review bid strategies which are a more advanced option for bidding.
Google Analytics not only provides a picture of the good things happening on your site which include sales and lead generation. It can also point out the things that are not working well that need to be changed. Below are five steps for optimizing your website based on data available in Google Analytics.
Identify pages with the highest bounce rates
In the Site Content section of Behavior data, you can view your landing pages. View which pages are the common entry points to your site. Are those pages designed to receive new visitors? In other words, will a new visitor know what your website is about if they land on one of these pages rather than the home page? If you see a high bounce rate, it could be that it is not a good introduction for people who are new to your business and there is some room for improvement. Pay special attention to landing pages with a large number of visitors and a significantly higher bounce rate than the other pages on your site.
Analyze goal performance
The easiest way to see how you are doing on your site is to look at your goals in Google Analytics. However, rather than looking at just the raw numbers or the absolute percentages, examine what’s been happening over time so you can see if there’s been an improvement in site performance. So yes, you will still want to look at conversions as a number on it’s own, but make sure you do watch for trends over time. And give extra attention to any spikes and conversions that may be related to a new marketing activity that you started or perhaps an update that was made to your website.
General site performance
The Site Speed section under Behavior is more relevant to the website developers then to the marketers. However if you are involved in decisions about the design, it is helpful to know if an image-heavy page seems to be causing a slow load time. With short attention spans, you are losing potential customers who lose patience and visit another site. The good thing is that this is one of those problems that can be fairly easy to fix and it’s very clear-cut.
Check the Reverse Goal Path report
In a previous article, I wrote about how the User Flow through a website might not be that meaningful for many sites. If there is not a clear path through your website to reach a conversion, it is hard to know what you are measuring other than the end result. However, you may still want to check the Reverse Goal Path report to see the pages a visitor navigated before converting on your site. It can be helpful to know which pages are the common steps followed by people who eventually complete a desired action
Look for drop offs in your conversion funnel
Once you set up goals you can see how people travel through your site to complete them with a Funnel Visualization report (found in Goals section of Conversions). With this report, you can learn which pages have a high exit rate in the funnel and brainstorm ways to improve them.
Website optimization is all about using data to make your website more attractive to visitors and make visitors more inclined to complete a desired action on your website. Looking for ways to improve your site using your analytics and qualitative data such as customer surveys is an ongoing process.
And always remember that no matter how great you may think your website is, you are not the customer. If you can stay curious about what it is your website visitors want, it’s easier to get rid of your assumptions and to be more open to creative ways for improving performance. There can be room to change your overall copy, your images, your navigation, your offers, your landing pages etc. The bottom line is if you adopt a mindset where your website is not a one-time project, you will increase the ROI with the channels used to send you traffic.
A self-referral in Google Analytics means that your website is indicated as a referral source to your website which could have something to do with your installation. Like with spam referrals, it’s frustrating to site owners if it happens a lot because it throws off your data as a whole. You don’t know the true source of that great conversion traffic.
A lost referral is similar and occurs when there is a third party tool involved, such as PayPal for collecting payments off your website. Once you send visitors off your site to complete a task, like a purchase, the website where they complete the action is the one that receives credit for the conversion. I call it a “lost referral” because you don’t know if the originating traffic source was organic, paid, email or something else because Google sees the visitor coming back to your site from PayPal and assumes PayPal should get the credit for the conversion. Google Analytics also records that return visitor as a new session even though they only left your site for a moment to process payment.
You use the Referral Exclusion List as seen below when you do not want traffic from an external site to be a referral source or record a new session when the visitor returns from that site. Using the PayPal example, you would add http://www.paypal.com to your Referral Exclusion List.
But this isn’t a perfect solution, although it does at least prevent a new session from being created when the situation described above occurs. Simply adding a website to exclude list does not enable you to see what site visitors do on the third party site. You want to see the full journey the user took which is when you would set up cross-domain tracking AND add the site to your exclusion list.
To have a picture of the full user path, you need to implement cross-domain tracking if the back and forth happen across website properties where you can implement tracking code. Cross-domain tracking then would allow you to track your website visitors to the third party site and requires assistance from your web developer for this set-up.
When you don’t have access to the outside website property and cannot implement cross-domain tracking, with as a solution such as PayPal, you will instead edit the Website Payment Preferences on their website. In these preferences, you will turn ‘Auto Return’ to on and enter the URL of your ecommerce post-purchase page, such as http://www.website.com/ordercomplete.php. A the end of the URL for your post-purchase page, include ?utm_nooverride=1. Doing this gives credit to the original source of traffic rather than giving PayPal the credit for the purchase. This Website Payment Preferences screen was a bit tricky to find in PayPal’s help files, so go directly to this link to make this change rather trying to hunt for it.
You can start addressing referral issues by going to the Admin area of Google Analytics and adding relevant sites to the Referral Exclusion List. Then, if you have web properties that you can add tracking code to, you will want to use cross-domain tracking to properly record the shopper’s journey. With PayPal, follow the above steps. Every vendor is going to be a bit different so check their help files or contact customer support to ensure you have this set up right. Especially if you are using multiple digital marketing channels to sell products or services, you want to give credit to the right place so you know what to do more or less of.