Automate Your Analytics Implementation QA

April 8, 2019 Matthew Maddox

We live in a time where everything we need is at our fingertips or the tip of our tongue. 

You want to know the weather? Ask Alexa. You want to know how to fix your garbage disposal (as was my case 30 minutes ago)? Type it into Google. You want to know how many days there are until March 12, 2032? Ask Siri. 

With the advanced technology available today, the possibilities are endless. So why should that be any different for your analytics implementations? 

It’s not. Not anymore, that is. 

With an automated QA platform like ObservePoint, you don’t have to continually worry about whether your analytics are being implemented properly within your site. You can have confidence that you’re receiving accurate data from your analytics. 

But before we dig too deep into how automated solutions can drastically improve your website’s analytics, let’s take a step back and look at the current problems with testing analytics. 

Manually Testing an Implementation is the Worst

That’s right, manually testing marketing technologies on your web pages is the worst! It’s draining, tedious, and time-consuming. And with all this comes human error, and by the time you’ve manually tested all the implementations on your website, there have been 15 new updates to it and you’re already miles behind and unlikely to ever catch up—making the worst thing even worse. 

As an analytics professional, you place a lot of stock in the ability of your implementations to collect accurate data. That data creates insights for decision-making, helps you to improve user experiences, and is a major contributor to growing top-line revenue. 

So, how do you achieve a strong analytics implementation without draining your team’s resources? By testing your marketing analytics—automatically. 

Unfortunately, many companies still rely on archaic manual methods for testing their implementations. Let’s talk about two of these methods of testing and then get on with introducing you to the new way of testing: automatically. If you want to skip the history lesson, you can jump ahead to “Automate Analytics Testing.” 

Manual Debugging with Developer Tools 

The most basic form of analytics debugging can be done using your browser’s developer tools. 

As your browser makes requests to various servers (be they your own or a third-party’s), the browser can show those requests for you in the network tab of the developer tools. The network log holds a wealth of information that you can check to make sure data collection is happening correctly. Specifically, you can check: 

  • What data is being collected 
  • Where the data is being sent to 
  • Whether the data was successfully received by a third-party user 

While the network log is relatively robust, it has serious limitations as an analytics debugger. For example: 

1. The network log is full of noise. The requests in the network log aren’t restricted to just analytics data collection. Literally, any call to a server from your client (the computer) will show up in the network log. That means requests for HTML files, JavaScript libraries, CSS files, anything. When your job is to debug tags, you don’t want to bother sifting through all these other requests. 

2. Debugging with dev tools is not scalable. If you rely on your browser’s dev tools to debug, you can only debug one page at a time. Manually debugging request after request, page after page, day after day, is not only tedious, but also not feasible. 

3. Network requests are hard to read. Network requests were written to be read by computers, not humans. As a result, reading through them can be difficult and presents opportunities for human error.

A lot of analysts (very skilled analysts) use this method to debug websites. Spot-checking pages via the developer tools has its place, which is why ObservePoint has published a tag debugging assistant called the TagDebugger. But manual tag debugging is not a good solution for the long run. 

Manual Debugging with a Packet Analyzer 

An alternative to using your browser’s developer tools is to use a proxy server, known as a packet analyzer or packet sniffer, to capture the requests coming off your connected device. Essentially these tools intercept requests and log the information like your browser’s dev tools, with some additional features above and beyond your browser’s tools. 

Packet analyzers like Charles and Fiddler are inadequate for some of the same reasons we mentioned above: they capture too much noise, are not very scalable, and their requests are not human-friendly. 

There is a lot to test when it comes to your marketing technologies. How do you find the time to do it as often as it’s needed—as often as pages and mobile apps are updated? How can you possibly stay on top of testing everything manually when everything is constantly changing? And why are you even still doing this tedious process over, and over, and over…and over again when what you really want to be doing is analyzing the data you’re collecting, not just constantly testing and cleansing it? 

With automated QA technology, you can spend more time analyzing, and less time debugging. 

Automate Analytics Testing

By automating analytics testing with a platform tool like ObservePoint, you will ensure your implementations are deployed correctly, collecting accurate data and increasing your ROI.

ObservePoint’s automated auditing solutions help you establish a repeatable framework that aligns teams and technology: the Tag Governance Framework. The Tag Governance Framework encompasses the essential components of an effective MarTech governance program, giving you the ability throughout your process to set up rules, audits, reports, and notifications specific to your organization’s needs. 

ObservePoint offers different features that allow you to accomplish the steps of this framework and create a more effective system for your marketing technologies. 

No more spending hours upon hours on manually testing and correcting implementations only to turn around and start again any time a change is made to your website or mobile app. Following ObservePoint’s automated QA solutions and Tag Governance Framework for your analytics testing gives you the reigns to take control of your analytics, faster than ever before. 

But, don’t take our word for it. The following stories are from four other companies that have experienced the impact of ObservePoint’s automated platform within their analytics testing. 

How Four Data-Driven Companies Use Automation

Exture Reduces Recruit Lifestyle’s QA Time by More Than 75% 

Founded in 2012, Exture has become a highly respected consulting services firm in Tokyo, Japan, helping users of Adobe Analytics and Google Analytics to do a more effective analysis with their web analytics and other big data platforms. 

Together, Exture and ObservePoint worked with Exture’s client, Recruit Lifestyle, to establish superior practices of data governance within their organization. 

Recruit Lifestyle provides media services to some of the largest brands in Japan. Though Recruit Lifestyle had an understanding of the importance of proper data governance, with their team managing websites and implementions for over 25 media and services brands—and with more digital properties and companies being added all the time—the Recruit Lifestyle team was struggling to keep up with the extensive data QA processes and tasks. Who wouldn’t be? 

By not being able to stay on top of performing regular QA on their analytics implementations, organizations risked collecting, analyzing, and utilizing incorrect data in making their business decisions. 

According to Hideki Nakano, Web Analyst at Recruit Lifestyle:

“It is very difficult for us to check all the variables and all the pages manually. We have found ObservePoint to be the best way to solve these problems.”

Recruit Lifestyle’s decision to use ObservePoint was made based off of two reasons: 

1. ObservePoint’s powerful functionality with the ability to automatically identify tagging errors. This made creating rules based on the variable-finding audits so easy that Recruit Lifestyle now has over 800 customer rules that guide automated audits for 120,000 pages every month. 

2. The ability to export data from ObservePoint’s website audits and journeys that makes visualizing results from raw data, easy for the whole organization. 

What were the differences after implementing ObservePoint?

Before:

  • 4 full-time employees QAing per month
  • QAing only main path and pages
  • No website health monitoring

After:

  • Less than 1 full-time employee QAing per month
  • Audit all pages automatically
  • Monitoring slow-loading pages, JS, errors with broken links, etc. 

To see the full Exture and Recruit story, click here.

Pharma Marketing Agency Increases Efficiency 87.5% 

As a digital marketing agency, this Pharma Marketing company works with all their client accounts to implement and monitor online activity analytics and to discover insights that drive strategy. 

Serving many respected logos to date, Pharma has struggled with manually running QA processes during development, staging, and production for their clients, in addition to tracking the hundreds of client websites they maintain. 

With a high potential for failure at this daunting task of identifying any gaps in tag implementations—costing the agency’s clients crucial data, time, and money— they turned to ObservePoint’s tag governance solution to automate their intensive QA process. 

The Result? 87.5% increase in efficiency. 

The Director of Analytics explains, “It took two people in analytics a total of 16 hours to QA a site. But now that time is down to just the 5-10 minutes for an automated ObservePoint audit.” 

If anything breaks, ObservePoint’s automated system alerts the agency immediately, giving the Pharma Marketing Agency’s team peace of mind with their website and marketing technologies. 

To see the full Pharma Marketing Agency’s story, click here.

How HPE Cleaned Up Their Analytics Across Business Units 

With millions of pages across hundreds of sites and 4.9 billion annual pageviews, maintaining data quality can be extremely difficult. This was especially the case when HP split into Hewlett Packard Enterprise and HP Inc. 

While trying to clean up the siloed data across different business units, keeping clean data was hard. The digital analytics team at HPE had previously attempted to adopt several different universal analytics solutions, but with the clean slate opportunity presented by the split, they wanted a uniform tag governance strategy with a universal deployment of Adobe Analytics and one tag management solution across all HPE business units. 

The team performed an ObservePoint WebAssurance audit of the new HPE digital properties to see where the quality of their analytics implementations initially stood. The audit provided them with results such as: 

  • 6% URLs missing Adobe Analytics tags 
  • 7% redirecting on storage pages to 404 status codes 
  • 12% of “Page Not Found” on servers 
  • 10% of URLs with JavaScript errors 
  • And more 

These results were those of common errors found. However, these were nothing compared to the results they found with their paid search misallocation, and their search traffic and SEO being down by 50 percent. 

As Daryl Acumen, Senior Manager of Digital Analytics at HPE, put it, “The lesson here is that we’re smart people, but we’re not clairvoyant, we’re not perfect, and things aren’t always caught in our manual QA. You need to have some sort of governance plan to catch them before, during and after-the-fact—preferably automatically.” 

HPE is moving forward with a plan to improve its automated data quality by assigning ownership of its maintenance and formulating a clear tag governance plan using ObservePoint. 

To see the full HPE story, click here.

Start Saving Time and Money

With ObservePoint’s automated platform testing your analytics implementations, you will optimize the tools you are spending so much money on, giving you confidence in the data being generated. You can see real and lasting results with little work on your part. See how automation is possible with ObservePoint, and start saving time and money, today. We make it easy for you to get it right. Schedule a demo today.

About the Author

Matthew Maddox

Matt’s mission as the Director of Enablement at ObservePoint is to educate customers to use the marketing technologies they select for their sites most effectively. Matt delivered training at Omniture and Adobe for over eight years before joining ObservePoint. He was the dedicated trainer for several global companies, creating and delivering custom courses based on their corporate business requirements. With a wealth of experience solving analytics questions in many industry verticals, including e-commerce, media, finance, lead generation and automotive, Matt offers sound direction and analytics insight.

LinkedIn More Content by Matthew Maddox
Previous Article
8 Campaign Tracking Limitations That Frustrate You Daily
8 Campaign Tracking Limitations That Frustrate You Daily

If you’re using spreadsheets, SFDC Campaigns, or others to manage your campaign and touchpoint tracking, yo...

Next Article
Google Analytics Crash Course: 25 of the Best Tricks and Hacks
Google Analytics Crash Course: 25 of the Best Tricks and Hacks

Analytics solutions are constantly changing, making it hard to keep up. Learn the 25 most beneficial tricks...

Free Website Audit

START NOW