Troubleshoot viewability reports

Find out how to troubleshoot and improve Active View viewability

Active View reporting in DoubleClick for Publishers (DFP) tracks the viewability of ads served via Google Publisher Tags (GPT). Below are tips to improve viewability performance for your inventory and a guide to comparing third-party viewability reports with DFP reports.  

Improve your Active View performance

Here are tips to optimize Active View metrics and troubleshoot low eligibility, measurability, or viewability in Active View reporting. Learn how Active View measures viewability.

Improve eligibility

  • Ensure that requests are not from GPT simple URL tags, which aren't enabled for Active View.
  • Use Chrome Developer Tools to read the Active View pingbacks and check the &oid= parameter. If the parameter is missing, the GPT implementation has likely been modified in an unsupported way.
  • Ad requests greater than 2,083 characters are truncated. If the &adk= and &adks= parameters are lost, the request/impression is "not eligible" for Active View.

Improve measurability

  • Check for cross-domain iframes, creatives that break out of the iframe or tracking pixels being served. The &bs= parameter of the Active View pingback in this scenario will be a negative number.
  • Ensure that the DOM specifies measurable sizes for creative objects. For assets hidden upon page load and some custom interstitials, the DOM may report the creative to be 0x0, 0x? or ?x0. These are not measurable since there are no pixels to measure. You can check the &p= parameter of the Active View pingback to see how the DOM sizes the creative.
    Even if you see a creative on screen, the DOM may report a zero-pixel size because of rich media or engagement creatives. Work with your development teams to resolve this issue.
  • Creatives that are expandables, wallpapers or other custom implementations often cannot be measured because of how they appear in the Javascript DOM. Determine their size with the &p= parameter.
  • CSS styling such as display:none; can cause the DOM to report an non-measurable size upon page load. To hide a page that opens upon load, render it outside the viewport, then bring it into the viewport at the desired time.

Improve viewability

  • Render all requested line items when using GPT SRA.
  • Break down your report by position targeting. If you have a 728x90 at the top and bottom of the page, the bottom slot may lower your overall average.
  • Consider creative placement and user habits. Creatives at the very top of a page generally have lower viewability than creatives just above-the-fold. Users often scroll down immediately as a page loads, causing top-of-page slots to be out of view before 1 second elapses.
  • Ensure that you’re viewing "Active View % viewable impressions," and not comparing "Viewable impressions" against "Total impressions."
  • Lazy loading can affect viewability. Try to load GPT slots as quickly as possible so that viewability measurement is ready when the slot comes into a user’s viewport. Users often scroll quickly through a page; if the slot only loads when it comes into view, it’s probably too late.
  • Floating elements over the page with CSS (e.g. float:) can interrupt viewability measurement because the JavaScript DOM may not reflect the positioning that you see.
  • An auto-refresh function on your page can accumulate "measurable, but not viewable" impressions if a user leaves your page open within a tab of their browser for a long period of time.

Compare third-party viewability reports

Below are the most common reasons DFP reports will differ from a third-party's report.

Many discrepancies are specific to one size or one ad position
When you compare reports, make sure both reports are broken down by creative size. Position targeting is also very useful, if possible.
  • Different total measured impression values used to calculate % viewable metrics
    Active View is built directly into the ad server, so it can often measure a greater number of impressions than other non-native viewability solutions. Always compare the third-party’s raw measured impressions totals against Active View’s Measurable impressions total; not percentage numbers.
  • Third-party creative rendering latency
    Active View begins measuring as soon as the ad slot loads and a creative is returned from DFP. Most third-parties can only measure after the final creative asset renders, which may be significantly later. This can cause the Active View Measurable impressions total to be higher than a third-party’s report if users navigate away from the page between the time an ad slot loads and the final creative renders.
  • CSS styling nuances
    Active View determines measurability upon page load. Hiding elements with CSS upon page load can block Active View measurement, artificially decreasing your Active View Measurable impressions total.
  • Ad requests and creative rendering
    • GPT SRA allows you to request ads, even if you never display them on the page. Ads requested via GPT SRA, but never rendered, appear as eligible and measurable, but not viewable. This can artificially inflate Active View Measurable impressions and decrease Active View Viewable impressions %.
    • Out-of-page creatives require the viewed impression macro (e.g., %%VIEW_URL_ESC%%) to record a delayed impression. Improper implementation deflates Active View Eligible and Measurable impression totals.
  • MRC flexibility for measuring large creatives
    For large display creatives – 970x250 or larger; such as IAB’s “Rising Stars” formats – the MRC has given viewability vendors some flexibility, as long as the vendors transparently disclose the definition used with their clients in the Document of Methodology. It is acceptable to mark a large display creative impression as viewable when either 50% of the creative’s pixels are in-view or when 30% are in-view.

    Active View currently measures at 50% in-view (or 30% in-view for creatives of 242,500 pixels (970x250) or more). If a third-party measures at 30% in-view for large creatives, the Viewable impressions totals for those sizes may be lower in Active View reports when compared with third-party reports. Please consult your third-party to see if they uniquely measure large creatives.

Was this article helpful?
How can we improve it?