From targeting verification to privacy concerns to real-time optimization decisions, there are seemingly hundreds of factors that brand marketers must consider when it comes to brand campaign measurement. We’ve compiled a list of critical questions to ask your measurement partner to ensure you’re getting the most accurate, efficient, and privacy-compliant data when analyzing and optimizing your campaign investments.
How Accurate is Your Methodology?
For an accurate picture of multi-channel campaign performance, cross-media measurement should be limited to channels where deterministic measurement is possible, while other channels (such as OOH or some unmeasurable social platforms) should rely on MMM, as is explained here. Rampant contamination of exposed and control groups can easily make an effective campaign look ineffective, and vice versa.
- Do you provide an overall control for each study or do you provide a control group for every cut of data to ensure maximum accuracy?
- Do you use overall or cut-level controls for audience insights? Creative insights? Frequency insights?
- Are all media exposures deterministic?
- Can you provide deterministic cross-channel data for digital and TV in all its forms (CTV, OTT, Linear TV, Addressable TV)?
- Are TV results in your dashboard updated daily, or segregated out from digital results?
- Do you employ OTS (Opportunity to See) measurement as a stand-alone methodology or mixed with deterministic exposure?
- How do you account for the data contamination with OTS?
- Can you account for seasonality and market position (whether my brand is a newcomer, an incumbent, etc) in benchmarks?
Do You Provide Metrics That Support Optimization Decisions?
As we explored in a recent article, traditional brand lift reporting displays multiple metrics – observed lift, statistical significance – that can each tell a different story, making it hard to know how to interpret and act on insights mid-campaign. While brand lift statistical significance is appropriate for evaluating results at a couple of fixed points (e.g. mid and post-campaign reads), brand marketers need metrics that will more accurately inform real-time optimization decisions (such as reallocating budget or traffic between tactics) throughout a campaign. To ensure that you get these metrics from your measurement partner, ask:
- Do you have metrics that enable us to optimize before cuts reach statistical significance so we don’t have to wait until campaigns are nearly over?
- Can you surface daily optimization opportunities for media (creatives, publisher & DSP partners etc.)?
How Are You Dealing with the Massive Loss of Industry Identifiers?
The quickly changing privacy landscape — Chrome’s erosion of 3rd-party cookie support; Apple’s changes to IDFA — coupled with the reality that 1:1 tracking has always been probabilistic, has further heightened the need for a future-proof, privacy-safe campaign measurement solution that delivers person-level exposure probability and deterministic, uncontaminated control groups. It’s a lot to ask, but shouldn’t be impossible. Here are some questions to ensure your measurement partner is on the same page:
- Is your solution in place now or do you have a multi-year plan to deal with the changes that already impact the industry?
- Are you using email as a proxy for the loss of MAIDs/IDFAs and cookies?
- If yes to email: Apple has repeatedly made it clear that emails can’t be shared in any way unless the consumer has opted-in to sharing their IDFA. Do you think publisher partners will risk being kicked off the app store to share email data captured without IDFA opt-in?
- Emails are based on logged in users who are roughly 10% of internet traffic—How will you provide granular data let alone overall results when you’ve lost 90% of your exposed responses?
- How can you provide any data accuracy when your control group is massively contaminated because you’re only measuring the 10% of logged in users?
- How will you measure programmatic media if you rely only on the 10% of respondents who are logged in users with email and Google has said no email-based identifiers will be included in the 50% of bid requests running through AdX?
- Is fingerprinting your solution?
- If yes to fingerprinting: How will this work when Google and Apple have repeatedly come out against this approach and have vowed to shut this down at every opportunity?
Is Your Dashboard Updated Nightly with Customizable Views?
Data is only as valuable as it is accessible and the table stakes of campaign measurement are constantly being raised. The multi-channel complexity of cross channel measurement requires more than static graphs in periodic powerpoint slides to fully understand. Without an interactive, frequently-refreshed, alway-on dashboard even the best data in the world will fail to deliver insights.
- Do you provide just an end of campaign report and possibly one interim read or more frequent data updates?
- Do you increase the features on which you weight your sample nightly, or only when an interim read is purchased? If the latter, how do you advise customers interpret dashboard lift numbers when they change significantly only when interim reads are purchased?
- Does your Dashboard provide a Tableau-like exploratory environment, or just online access to status reports?
- Can you combine any combination of cuts and KPIs in a chart?
- Can you create a custom dashboard of just the KPIs and cuts I want to monitor?
- Can you permission access to a dashboard by limiting what cuts other companies can see?
- Does your dashboard provide trend charts of a campaign’s brand lift over time?
- Can you save and export data in both graphical and tabular formats?
- Is your Dashboard an automated SaaS platform or is data updated manually?
- Can you create on-demand nested cuts of data like audience and creative by publisher partner and do you have the scale to show results for these more granular cuts?
Do You Provide Targeting Accuracy Analytics?
Despite the recent explosion of CTV, legacy measurement platforms are not equipped to support this new landscape of video targeting beyond age and gender demographics. Not to mention, targeting verification has historically been fragmented, with coverage gaps from several major media partners. Aside from simply trusting a publisher or a data provider, there is little that a marketer can do to verify audience accuracy. And without verifying your audience across all digital channels, it’s difficult to justify your brand campaign investment. Make sure you’re getting what you pay for by asking:
- Do you provide targeting accuracy data on media and audiences?
- Do you provide a profile of my exposed audiences by campaign, publisher and audience? Is this profile limited to age and gender, or also include brand-specific traits?
- Do you profile my entire campaign, include all CTV partners, Addressable TV and Linear TV, in one place?
- Do you provide a census baseline for my target audiences, including brand-specific audiences, so I can see my campaign is indexing relative to census?
How Are You Reducing the Cost and Time Required for Measurement Setup?
Agencies today are faced with mounting pressure to streamline their ad operations from both economic drivers (tightening budgets, workforce shifts) and industry drivers (increasing amounts of data, multiplying media channels). Measurement partners must support automated ad ops to ensure lean agencies can meet these demands.
- Do you provide same-day setup of measurement?
- Do you integrate directly with DCM, Sizmek or other ad servers so I can generate tags self-serve?
- Do I have to wait for an SLA to get tags, or can I generate them entirely self-serve?
- Do you automate impressions comparisons and tag troubleshooting or rely on manual, managed services for this?
We hope you find these questions productive when working with your campaign measurement partner. Be sure to browse our Resource Library for additional expert insights into brand campaign measurement.