Cookies alone distort result:
● 58% – overstatement reach
● 135% – understatement of frequency
● 51% – accuracy in broad age & gender targeting
Metrics like clicks aren’t a good proxy for results
● 91% – of people who could buy your product don’t click your ads.
Models like last-click attribution can inaccurately value your touch points
● 22% – of incremental revenue could be misattributed due to last-click models
● 54% – could be misattributed when mobile spend is high
Ensure only 1 variable is being tested across cells/test in order to isolate cause
Advanced Measurement – Facebook and Instagram Conversions
NBA Season Example:
We saw a spike on September 7th, where daily ticket sales reached a high of 225 conversions. This could be attributed to the pre-sale of tickets for the Miami Heat game on Nov 6.
Since the start of October, we’ve seen a healthy increase in daily conversions through Facebook. This is obviously due to the dawn of the new NBA season. We’ll continue to track the daily conversions as the season progresses. We are positive that the current high of 255 conversions in a day will be far eclipsed, especially with the game-by-game ticket strategy.
Advanced Measurement
NBA Season Example:
Some fantastic results from both Facebook and Instagram, recording a ROAS of over
1,000% on each platform. What’s interesting here is to compare the two channels.
There has been over 3x more Ad Spend used on Facebook than Instagram, but
Facebook generated:
- 5x more revenue than Instagram
- 5.5x the number of impressions
- 4.2x the number of conversions
While Instagram is a fantastic channel and compliments our tactics with Facebook, this
cements Facebook’s superiority over Instagram and gives us a clear focus on how we
should allocate future ad
Where people viewed your ads vs. Where people
converted
64% of conversions were from a desktop, yet 67% of ad views were from a mobile
device.
We expect these figures are due to the ad being viewed on the mobile platform, then
having to coordinate plans with friends or wait for their paycheck to come through too,
then, subsequently purchase the tickets on a desktop. This is where they feel more
comfortable inputting personal details and making sure their purchase is submitted
correctly.
Touchpoint to Conversions
A real anomaly in the chart is from 80-90 days. There seems to be a significant
increase, considering between days 63-79, it’s an almost dormant period of
conversions. This could be attributed to ticket sale releases for games in the next 2 to 3
months, with fans scoping out price and availability when our new campaigns are
launched. By the time our last-minute sales ads are delivered, the fans purchase tickets
a day or two before the game with Facebook giving the initial touchpoint credit for the
conversion. Another possibility is the 85+ day conversion is a carryover from last
season’s playoffs.
The immediacy of purchases between a 1 and 5 day period is great. If we can stimulate
this urgency on all ticket sales, we can continue to increase Facebook revenue and
allocate both time and budget regularly throughout every individual home game
campaign.
Although 52% of conversions occur within one day, people are viewing and researching
many days prior to this. 58% of the clicks are beyond 7 days, so we would think that we
are planting a seed in the fan’s mind in preparation for the conversion closer to game
day.
A click means that the fans are interested in the ad and therefore not passively blasting
impressions at them. The more touch points we can achieve, the more assists we can
execute