Web Analytics & Customer Analytics Face Off: Tracking Users Across Multiple Devices

Web Analytics Customer Analytics Face OffWeb analytics and customer analytics go head to head on the topic of tracking users across multiple devices. We conclude with a final showdown comparing how the differences between them will be reflected in your reports.

Let the match begin.

In This Corner: Web Analytics

Web analytics relies on cookies to track your visitors. Whenever someone lands on your website, a cookie is stored on their browser to track their activity. However, since cookies are stored on browsers, if someone visits your site from a different browser (e.g. Firefox and then Chrome) or a different device (e.g. laptop and then tablet), there is no way using web analytics for you to know that they were the same visitor. But that’s a rare situation, right?

In This Corner: Customer Analytics

Let’s start with a question: How many devices do you use? According to a 2012 report by Cisco, there will be 1.4 mobile devices per capita by 2016. And that’s just mobile.

As we covered above, web analytics relies on cookies and therefore recognizes each cookie on each browser as a new person. Many of your customers use a work computer, home computer, mobile device, and possibly a tablet on a near daily basis. That means they’re probably also using your website and/or application from a work computer, home computer, mobile device, and possibly a tablet. See the problem?

Not having this reflected in your data creates a lot of noise. That’s why customer analytics uses a unique identifier, like an email address, to identify the same user across multiple devices. Here’s how it works:

      1. Anonymous visitor lands on your website from work computer. They now have a cookie on that device.
      2. Anonymous visitor then signs up on their laptop. They are now identified as Jim Halpert.
      3. Jim Halpert downloads your app on his iPad. His activity on your app is aggregated into Jim’s profile.
      4. Jim visits your website on his work computer and logs in to access his account. All of his previously anonymous activity (from step 1) is now aggregated into Jim’s profile.
      5. Jim visits your website from his laptop, and does not log in. The systems has already associated the cookie with his account and so he is still recognized as Jim Halpert

In a Woopra customer profile it would look something like this:

Track User Across Devices

Notice how the device type, browser, and location are reflected in each visit. We’ve successfully tracked Jim across his work computer, laptop, and iPad, and aggregated all of that activity into a single profile for Jim Halpert. One customer, one lifecycle, one data set.

The Verdict

Now the real question is how is this difference in tracking methods reflected in your analytics reports?

Example 1: Funnel Report

Customers don’t always complete a funnel in the same visit. For example, they may add items to their cart during their first visit on their work computer and then checkout during their second visit on their home laptop.

Web analytics will fail to report this conversion in your funnel because it sees each visit as a different person. Customer analytics, however, will recognize that this is the same customer, and you will see the conversion accurately reflected in your funnel.

Example 2: Retention Report

Retention reports tell you how many people complete an initial goal on your website and for how long they continue to complete a subsequent goal. For example, you may want to see how many people make a purchase (initial goal) and then for how long they continue to make purchases (subsequent goals).

Just as with funnels, if a customer completes the initial goal on one device and the subsequent goal on a different device, web analytics will fail to report this. Web analytics sees each device as a different person and so you will not have an accurate idea of how long people continue to come back and make purchases.

Example 3: Reporting Number of People

Say you want to see the number of people who came from a particular campaign. Some people will likely click on your campaign more than once and from more than one device. Web analytics will tell you the number of pageviews you received from each campaign, while customer analytics will tell you the number of actual people who came from that campaign and also the number of visits.

Take a look at the screenshot below to see a customer analytics report in action.

Campaign Tracking

Your Experiences

What issues have you run into with traditional web analytics? Let us know in the comments and we may tackle your problem in an upcoming post.

What Should I Track? Part 2: Tracking SaaS (B2B)

As a SaaS company, your entire business lives online. Tracking user activity and understanding customer behavior is paramount. But how do you obtain this visibility? The first step is identifying the exact activity and customer information you need to track.

SaaS companies will have more variation in what should be tracked than other companies like e-commerce or lead generating websites. That is because each SaaS company provides a completely different type of product with very unique actions. Yet, by following some guidelines that apply to nearly all SaaS companies, you can make the process much more efficient.

Stage 1: Marketing Pages & Signups

First and foremost, every SaaS company should track signups. This is one of the most important actions visitors take on your website and you will want to analyze and optimize it over time. For example, if you’re tracking signups, you can create a report that looks like this.

SaaS Signups by Company Size

In this report, we see the number of signups that eventually upgraded (see segment at top left “Upgraded”) organized by company size. The reason we were able to sort by company size is because we are sending the property “company size” with the event “signup”. Other properties you can send with the signup event include:

  • Email address
  • Name
  • Username
  • Plan/account level
  • Company name

The most important property to send with the signup event is the customer’s email address. This is what allows you to identify the visitor and track them across multiple devices (e.g. laptop, iPhone, work computer, etc).

Customer Analytics Profile

Customer profiles, like the one above, detail every action taken by and the engagement level of each user. Providing the customer’s email address via the signup property allows the system to associate information with this user, giving them a name, username, avatar, etc. This insight is extremely helpful for sales, support, and customer success teams.

In addition to signups, you will want to track any important activity that happens before sign up. This information helps your marketing folks understand what’s working and what’s not.

For example, if you have a promo video on your homepage, you will want to track whenever someone plays the video. You will also want to track whenever someone downloads a white paper, submits a demo request form, shares a blog post on Facebook, etc. With all that time and effort your marketing team puts into producing and maintaining the content on your website, they will want to measure and optimize its effectiveness.

Stage 2: Onboarding & Free Trial

After sign up comes user onboaridng and the extremely critical free trial stage. Each SaaS company has its own onboarding process and key milestones, which all need to be tracked in order to monitor and analyze product adoption and usage. This is where you should give a lot of thought to answering the following question:

What are the key actions a user must take to get value out of my product?

It also helps if you pretend to be a user and go through the sign up, implementation, and onboarding process to document every critical action.

For example, a task management SaaS product that helps teams collaborate efficiently would have the following onboarding milestones for a new account:

  1. Setup profile
  2. Invite team members
  3. Create first task
  4. Assign first task

Once we are tracking each of these onboarding steps, we can then analyze, slice, and dice all the data. We can see this activity in each customer’s profile and also create all kinds of reports that give us more insight into our customers’ behavior during onboarding.

For example, a funnel displaying the task management company’s onboarding milestones helps us find our bottlenecks and optimize our conversions.

Task Management Funnel

Stage 3: Usage & Retention

You will want to track the key actions customers commit while using your product. This will vary quite a bit from one SaaS company to another, so again you have to ask yourself a question:

What are the key actions a user takes to continue to get value from my product?

For example, Zendesk will want to track when users assign and reply to tickets, while Salesforce will want to track when users add and edit lead records.

Once we’re tracking these events, we can use them throughout our system so that our reports are directly relevant to our business. For example, we can set up a retention report to see for how long customers continue to use our product and therefore how valuable it is to them.

An online meeting company tracks when users sign up and also when they host online meetings. They created the below retention report which measure how long users continue to come back and host online meetings after they first sign up.

Online Meeting Retention

In addition to your “core” actions, you should also track the actions you would expect to see advanced users doing, such as using your more complex features. This will enable you to analyze the behavior of power users and give you clues as to how you can help move your less advanced users along this path.

Other Events to Track

Some other events nearly every SaaS company will want to track include:

  • Payment – Whenever a user makes a payment. You will want to also send properties such as payment amount and service the payment was for.
  • Sent email – Whenever you send an email to a user, such as one that is part of an automated campaign. Properties you may want to send with this event include email subject, email sender, and campaign ID.
  • Support request – Whenever a user submits a ticket. You can also send properties like ticket ID or subject with this event.

Do you need help tracking custom events? Read the documentation or send us an email to support@woopra.com

Disclaimer: Some of the examples above reflect stories from Woopra clients and have been included with the consent of those clients. Others are fictional examples used purely for illustrative purposes.

Success Story: GetGoing, Travel Startup That Helps You Save Big (Video)

GetGoing is a San Francisco-based (in fact, just down the block from us!) travel startup that allows you to choose two travel destinations and let their system pick one for you, with up to 40% in savings on the ticket price. With their eight patents and partnerships with major airlines, GetGoing is able to offer customers some serious savings that you won’t find elsewhere.

Eli Rosenberg, the Director of Product at GetGoing, gave some of his time to tell us about how their team uses Woopra. Check out some of Eli’s key points:

  • On Choosing Woopra – On the recommendation of another CTO, GetGoing was introduced to Woopra, which they ultimately chose for their analytics based on Woopra’s ability to give them a customer-centric view
  • Quantifying Efforts – GetGoing uses Woopra to understand which features are most valued by their customers, where their bottlenecks are, and which campaigns deliver the best ROI
  • Making Informed Decisions - With Woopra, GetGoing now has hard numbers to direct their decisions, such as which features to keep and which ones to eliminate
  • Analytics Like No Other – Eli explains how Woopra’s ability to track both an event (e.g. search for flight) and its properties (e.g. destinations selected) allows GetGoing to get insight that they haven’t seen available anywhere else
  • Measuring New Features – As a product manager, Eli is able to watch how customers interact with new features in real-time, giving him a deep understanding of how users respond to updates
  • Improving Customer Support – Whenever GetGoing receives a call or email from a customer, they go first to Woopra to see the customer’s exact activity, which helps them troubleshoot efficiently

Of course Eli says it much better himself, so take a look at the video for the full story!

Interested in sharing your success story with us? Shoot us an email to blog@woopra.com

Woopra’s New Home! And We’re Hiring!

The Woopra office has picked up and moved one block away to make room for our new team members!

Woopra New OfficeThat’s right, we’re currently hiring in San Francisco and scouring the area for the best of the best. If that’s you, check out our careers page to learn more about these positions:

  • System Engineer
  • Front-End Engineer – Javascript
  • Software Engineer – Java/Servlets
  • User Interface Designer
  • Sales Specialist

We’re looking for proactive self-starters who thrive in (proven!) startup environments.

Wonder where you’ll be working?

Our new office is a sun-filled open space with a killer city view, right in the heart of SOMA – San Francisco’s hottest tech scene.

Oh yeah, and we’re right around the corner from the CalTrain!

Woopra Map

So you want to apply?

Drop us a line at careers@woopra.com and tell us why you’d be a great fit. Don’t forget to include examples of your work so we can see you truly shine.

Learn more about our job openings

Customer Analytics for Growth Hacking

If you’ve been around the tech industry the last couple years, you’ve heard a lot about growth hacking, or, the practice of finding innovative ways to accelerate the number of users for your product.

While the practice is often talked up to sound almost mystic, no growth hacker can bring you tens of thousands of customers with the wave of a magic wand. In reality, growth hacking requires quite a bit of analysis to deeply understand how customers use your product. As they say, you can’t improve what you don’t measure.

Your company may or may not have a designated growth hacking team, but the practice of understanding how customers use your product in order to drive growth is a crucial component for every business. Below, we outline a few of the growth hacker’s secret weapons.

Funnel Analysis

Spot & Optimize Key Problem Areas

A central part of a growth hacker’s job is discovering where and why you lose customers. Simply finding and optimizing one key area where you have a large drop off can have a ripple effect, leading to a major improvement in your overall conversion rate. But in order to do so, you have to first find the main problem areas so you know where to focus your efforts.

The best place to start is with your funnel analysis. This report is built to show you exactly where you have the largest drop offs in your conversion funnel.

The example funnel below shows the conversion process for Facebook users. The conversion steps are:

  1. Complete sign up form
  2. Confirm email address
  3. Add 10 friends
  4. Complete 60% of profile

Facebook Sign Up Funnel

We can see from this funnel that there is a major drop off at the “add 10 friends” step. This is of course a critical requirement for a user to get value out of Facebook’s product as Facebook is essentially useless if you aren’t connected to any friends.

Clearly, in this example, Facebook’s growth hacking efforts should be focused primarily on getting new users to add friends.

Learn more about how to build and optimize your onboarding funnel.

Find Your Best Performing Customer Segments

To get even more growth hacking insight, you can take a look at how different segments of customers move through your funnel. This will tell you which types of customers perform the best so that you can work on getting more of them to drive your growth.

Funnel Segments

In the second part of our Facebook example, we see that the three student segments greatly outperform the other segments when it comes to the “add 10 friends” step. Our conclusion makes sense as student users are most likely to quickly find their network on Facebook and add friends.

However, if we look more closely at the three student segments, we see that the elementary and high school students tend to struggle with the final step “complete 60% of profile”, while the university students convert very well here. What we can take away from this for growth hacking is that the elementary and high school student segments should be encouraged via UI/UX to add information to their profiles. We can also conclude that university students move through the funnel very well and are therefore a good segment to focus on to drive growth.

We also see that there is a very large number of sign ups among the “18-30 Non-Students”, however, they underperform at the “add 10 friends” step. For the purpose of growth hacking, we can conclude that adding features and UI/UX components that help users in this segment find their friends on Facebook will help convert more of them. The sheer number of users who sign up in this segment means that even a small improvement in the conversion will result in large gains.

Retention Analysis

Stay On Top of Customer Retention

No matter how many new customers you add, if you don’t retain existing customers, your growth will be slowed. You can fill a bucket faster than a sieve.

You need to understand if customers find your offerings valuable enough to continue using your service or purchasing from you. Retention reports help you get to the bottom of this by answering the question: for how long do my customers continue to come back?

You may be interested in knowing simply whether or not they continue to return, or you may want to know if they return and perform a specific action (such as make a repeat purchase or continue to upload files).

Let’s take a look at the retention report of an online meeting service. The report measures how long users continue to come back and host online meetings after they first sign up.

Online Meeting Retention

From the report, we can see that customers tend to use the product to host online meetings most 1, 3, and 4 weeks after sign up. By weeks 6 and 7, it starts to drop off, most likely because that is around the end of the free trial period.

Measure How Product Changes Affect Retention

A big part of growth hacking is understanding how product changes and new features influence product usage. By digging just a bit deeper into a retention report, you can get a good grasp of which changes created the most value for your customers.

In the second example from the online meeting company, we see from the report below that customers who signed up after December have shown better retention than those who signed up before that month. Knowing that the company launched a major product update at the end of December, the numbers indicate that customers responded to the change quite well.

Online Meeting Cohorts

Individual-Level Analysis

Watch & Learn in Real-Time

Often times watching how individual customers interact with your product is the fastest way to diagnose an issue or identify a useful feature. This level of granularity is a hallmark of customer analytics.

Customer Analytics Profile

For example, watching a customer as they navigate through your website in real-time can give you an instant sense of what features are “sticky”.

Do they quickly move through certain pages or features? That’s often an indication that the content there isn’t engaging to the user. On the other hand, if you see users spending a lot of time actively using a feature, that is generally an indication that the feature is valuable to them.

Answer Usability Questions

Watching users in real-time can be a more exploratory exercise, but you can also use it to answer specific questions. You can do this by looking for similarities in the behavior of individual users within a certain group.

For example, you can look at the past activity of several users who have been inactive for more than two weeks in order to uncover patterns in their behavior. Similarly, you can look at the the customer profiles for several users who recently cancelled accounts, who are power users, or who failed to convert from your free trial in order to find common behavior that will serve as clues for your team.

Going back again to the Facebook example above, we can look at the activities of several users in the “Elementary Students” and “High School Students” segments who completed the “add 10 friends” step, but failed to complete the final “complete 60% of profile” step. By simply monitoring users in this group, we can quickly begin to uncover clues as to why they fail to complete the last step. Once we know why, we can find the appropriate solution and ultimately improve conversions for this segment.

As one Woopra client recently expressed, “we could run a full usability analysis, or we could just spend 5 minutes on Woopra.”

Disclaimer: Some of the examples above reflect stories from Woopra clients and have been included with the consent of those clients. Others are fictional examples used purely for illustrative purposes.