How Third Parties Can Hurt Your Marketing Analytics


(Image: visualpower/Shutterstock)

(Image: visualpower/Shutterstock)

Ever have that feeling that too many people are at a party or an event? Well, imagine it digitally and you'll probably describe the current marketing technology or martech environment to a T.

In supporting marketing campaigns, analytics solutions have faced an invasion of third-party efforts that can add confusion in measuring meaningful results. Much of these efforts are benign in nature, but analysts must not leave them unchecked. If they do, branding and consumer privacy catastrophes are a risk.

We've been seeing hints of these challenges as analytics usage has advanced over time. One example is marketing campaigns that run outside a media plan's original scope. Agencies that create digital advertising campaigns for retailers are increasingly discovering their ads appearing on additional sites inserted with third-party tags.

The end result creates confusion on which party is capturing which data with which tag. Referral reports reveal these additional sites that are augmenting ad reach. But a downside from the augmentation is potential erosion of data control. If these sites are outside of the network originally planned for a media buy, then the data becomes skewed by these unexpected sources.

Another example is tag bloat, something that happens when the various tools on a given HTML page add additional tags. These additional tags can potentially increase the page load, which slows response for a website. That means page rendering in browsers or response to links is slower, creating a poor user experience. Data bias can occur as more tags are added over time.

A third example is based on how the user relationship of tag is viewed legally. The technology used and the relationship to that technology introduces regulatory risks. For example, take the cookie, the venerable text file used in browsers. Whether a cookie is added via first-party or third-party (set by a web service you may not be aware of) determines the legal risk. First-party cookies are generally low risk. Third-party cookies sometimes breach compliance through identifying what you click on and tell advertising websites to display that type of product or service wherever you go afterwards. Website owners based in the EU territories must disclose or seek permission to use this type of cookie. The key to compliance is disclosure or permission.

In all, marketers must consider mapping their analytics as an ongoing endeavor, not just a one-time event. There are now many unfamiliar ways to insert more data over time than may have been originally intended. Analysts must be more vigilant on data sources, acknowledging that vetting data quality impacts other kinds of breaches more severe than having poor data for a model.

Take the campaigns that run outside the media plan. Brands can become exposed to unintended audiences, a precarious position given consumer activism occurring in the current political climate. The consumer activism associated with opposition to or approval of the Trump administration is a clear example. Consumers, more empowered and vocal on social media about their experiences, are selecting and deselecting brands based on their feelings on Trump and whether a particular brand shares their sentiment. The Twitter profile, Sleeping Giants, has asked Twitter users who see brands advertising on the pro-Trump Breitbart site to highlight that ad placement and contact those brands to ask that they suspend their ad spend. Kellogg has been the highest profile brand that has discontinued its appearance on Breitbart. Sleeping Giants has said that more than 490 brands have been notified of their programmatic ads appearing on the controversial media site.

The introduction of new media will make vetting the relationship with outside parties more complex. App developers, for example, define first or third party differently, leading to confusion on which developer permissions are allowed. Mobile apps rely on third-party integrations for user details and to enable convenient sign in.

Data layers and tag managers have already demonstrated business value as programmatic advertising has grown. They have rightly positioned analytics teams at the center of the third-party storm. Analytics practitioners must be the bouncers for third-party sources, keeping their organizations honest about who is invited and who should stay away.

And as we have seen from the current consumer boycott activity over which brands are associated with Trump, everyone is deciding which party they want to attend.

Pierre DeBois, Founder, Zimana

Pierre DeBois is the founder of Zimana, a small business analytics consultancy that reviews data from Web analytics and social media dashboard solutions, then provides recommendations and Web development action that improves marketing strategy and business profitability. He has conducted analysis for various small businesses and has also provided his business and engineering acumen at various corporations such as Ford Motor Co. He writes analytics articles for AllBusiness.com and Pitney Bowes Smart Essentials and contributes business book reviews for Small Business Trends. Pierre looks forward to providing All Analytics readers tips and insights tailored to small businesses as well as new insights from Web analytics practitioners around the world.

How Voice Assistants Impact SEO, Machine Learning

Amazon is making an impact in voice search with its Amazon Echo and Alexa ecosystem of products. The rise of Alexa shows how machine learning strategy ultimately becomes an effective sales and business strategy.

Understanding Social Media with Sentiment Analysis

The rise of social media has led to a drive to understand all the new data it generates. Here's a look at how sentiment analysis can be applied.


Tag bloat is real.
  • 2/22/2017 6:22:13 PM
NO RATINGS

> Another example is tag bloat, something that happens when the various tools on a given HTML page add additional tags. These additional tags can potentially increase the page load, which slows response for a website. That means page rendering in browsers or response to links is slower, creating a poor user experience. Data bias can occur as more tags are added over time.

Yep.  Speaking for myself, I click away all the time from websites when it takes too long and I see too many widgets getting loaded.

Re: thanks
  • 2/21/2017 5:09:42 PM
NO RATINGS

Pierre as always excellent information that helps us all not fall victim to analytics that could be false. You have given us all great questions to ask. Are there any more questions you feel are especially important to consider when viewing your analytics given this information?

Re: thanks
  • 2/21/2017 4:15:37 PM
NO RATINGS

You are welcome. Glad you liked the post. 

Re: Bloated data
  • 2/21/2017 11:27:35 AM
NO RATINGS

I think it also merits stating explicitly that datasets are often scrubbed, massaged and otherwise made to look better than they really are for a consumer or a client. This isn't a topic I've seen addressed much til Pierre's blog here, but we need to become smarter consumers/recipients of the data being offered us, and by association, its conclusions.

Re: Bloated data
  • 2/21/2017 11:25:04 AM
NO RATINGS

Money changes everything, right? And I've got less of a problem with that if it's also accompanied by the transparency and disclosure of Nascar patches.

 

Re: Bloated data
  • 2/21/2017 8:19:48 AM
NO RATINGS

I know there have been jokes about making politicians wear sponsorship badges like auto racing team.  Maybe this isn't such a silly idea anymore.  If we have more corporate influence now we should be looking at why it is growing and how it influences individual politicians.  Again the transparency issue isn't likely to address this so it would be interesting to see if we could get regulations put in place to keep elected officials working for the individuals.       

Re: Bloated data
  • 2/20/2017 4:05:34 PM
NO RATINGS

The association is indeed amazing in terms of scale.  The business community has had this for a while, but it is also a by-product of the personalization trend. By using personalization strategues, marketers have established brands as having the same values as their customers.  When Trump ran, and introduced extremes - immigration, racial commentary, etc. - businesses that supported Trump had to decide their association. We probably would not have seen so much debate about affinity if Trump had loss.  His business history introduced a spotlight, one that certainly has implications for analytics.

Re: Bloated data
  • 2/20/2017 8:43:29 AM
NO RATINGS

@ T Sweeney I like that observation it is SEO for analytics.  It wouldn't be surprising to see this morph into an SEO like industry of driving marketing for analytics metrics.  "Use these three KPIs to double your sales to Trump supporters."  I'm still a little taken back that we're doing so much association of a company with a political party or a single politician.  At the very least this feels like a conflict of interest.

Re: thanks
  • 2/20/2017 6:23:10 AM
NO RATINGS

Thank's Pierre Debois for this news !

Bernard

Bloated data
  • 2/18/2017 1:58:27 PM
NO RATINGS

Let's call this phenomenon what it really is: Analytics spam. Or maybe SEO for analytics, since money's involved to inflate numbers and results and make something look more successful than it really.

But like spam, the quality is abysmally low. Ask anyone who's ever used one of Twitter's paid followers programs.

Page 1 / 2   >   >>
INFORMATION RESOURCES
ANALYTICS IN ACTION
CARTERTOONS
VIEW ALL +
QUICK POLL
VIEW ALL +