Measurement Flashcards
What’s the measurement process?
- When you’re identifying your objectives, consider HOW you’ll measure it
- Evaluating who your target audience is will help identify and inform your measurement goals so you can see how the campaign has performed in reaching the right people
- The targeting will inform the decisions aroudn what role each channel in the mix will have and in what format it will be placed. We will then be able to evaluate how well they are delivered.
- The decision on the best way to actually buy and activate the campaign will need to take into account how the buy’s success will be evaluated and measured
- While the physical act of measuring a campaign has taken place once it has gone live, it is a function that must be address at every step of the development
What are the two broad approaches to media planning and identifying KPI metrics to be measured?
- Brand - a top-down approach: typically starting with a broad, mass audience and planning revolves around how media can reach as much of that audience as possible. KPIs may include awareness, brand consideration and brand health
- Direct Response - a bottom-up approach - typically starting with a defined audience that may have an established affinity or relevance to a brand, and planning revolves around finding the audience specifically to provoke an action. KPIs might include acquisitions, web visits etc
What 3 things comprise the framework that provides control and accountability across the dual space?
- Viewability - “was the brand viewable on a page?”
- the measurement metric for any given campaign would be In-View %
- tells you what % of your total ad was viewable - Brand Safety - refers to environments or contexts that may have the potential to be harmful to a brand
- the measurement of this would be Brand Safe/Blocked %
- refers to the total amount of ads in your campaign that were served in safe environments - Ad Fraud or Invalid Traffic - refers to the practice of fraudsters misrepresenting human traffic
- could be through bots
- metric is Invalid Traffic Percentage (IVT) %, which refers to the total amount of ads in your campaign that were fraud free or fraudulent
- could be through bots
What are the 2 main Nielsen planning tools?
- Digital Ratings Monthly - Australian market’s monthly, total digital audience, unduplicated, across desktop, smartphone and tablet for both web browsers and apps
- Digital Content Ratings - marks the next phase of understanding consumers’ digital behaviour in a constantly evolving media landscape. And all this with a daily reporting frequency
- Digital Ad Ratings differs from these as it looks at the media environment audience to inform planning rather than assessing activity. The main purpose is
- to track who the ad is reaching
- to track if my ad has been seen
- to measure ad audience delivered on each device
What makes an ad viewable?
A display ad impression is viewable if 50% or more of the ad’s player appears in the viewable space of the browser tab for at least one second, and for 2 seconds for video ads
Who tracks page geometry and browser optimisation?
- Adform
- DoubleVerify
- comscore
- flashtalking
- integral ad science
- moat
- sizmek
- RealVU
What are some limitations to measuring viewability?
- Some mobile apps - due to the lack of standardisation
- Connected TV - which doesn’t allow measurement
- Some formats such as high impact formats or roadblock are difficult to track
What are classifiers? (Brand Safety - Context)
There are standard classifications that are widely agreed to be unsafe environments for advertisers - e.g. hate speech
Look beyond this and consider other variables that might have an impact on your brand e.g. a liquor retailer not appearing alongside a news story featuring alcohol-related violence
What tools should we put in place? (Brand Safety - Context)
- An Exclusion (blacklist) list includes a list of websites that should be excluded
- they contain inappropriate content, have excessive ad placements or deliver poor performance
- An Inclusion list (Whitelist) is made up of approved websites that are
- safe
- have appropriate content
- have quality ad placements
- good historical performance
What are thresholds of risk? (Brand Safety - Context)
Different thresholds can be set for brand-safety monitoring, ranging from very high risk to low risk
There are several methods for identifying brand safety via your vendor. What are 4?
- URL Analysis
- Keywords
- Links
- Metadata analysis
What is ad fraud?
The practice of fraudulently representing online impressions, clicks, conversion or data events in order to generate revenue. And every time money is involved you would have fraud.
It is impressions that result from an intentionally deceptive practice designed to manipulate legitimate ad serving or measurement processes or to create fake activity
Non-human traffic can be classified into two types:
- General Invalid Traffic (GIVT)
Traffic generated by known industry crawlers and traffic generated by bots doing the kind of things that real humans would probably never do. e.g. switching between sites every 10 seconds for hours - Sophisticated Invalid Traffic (SIVT)
Sophisticated invalid traffic is more difficult to detect because fraudsters are actively trying to avoid simple patterns that would raise a red flag. These fraudsters are making an extra effort to mask their behavior as legitimate, so it requires advanced analytics, multi-point corroboration/coordination and significant human intervention to detect, identify and analyse
What are 6 common ad fraud examples?
- Selling inventory automatically generated by bots or background mobile-app services
- Serving ads on a site other than the one provided in an RTB request (known as domain spoofing)
- Hindering a user’s opportunity to engage by frequently refreshing the ad unit or page
- Delivering pre-roll video placements in display banner slots
- Falsifying user characteristics such as location and browser type
- Hiding ads behind or inside other page elements so they can’t be viewed
What’s the difference between delivery and performance?
Delivery is validation that the publisher has delivered what was agreed, whether it’s impressions, clicks, conversions or completed video views.
Performance evaluates whether the publisher met your campaign objectives. e.g. cost per unique reach, cost per conversion, return on ad spend
SO delivery is about validation while performance is about whether your objectives were met.
