YuMe took a wide range of data including Nielsen TV/Internet Data Fusion, which integrates Nielsen's National People Meter Panel with the Nielsen Netview panel to obtain a clear picture of cross-platform reach, along with the following: Nielsen National Television Ratings, Nielsen Monitor-Plus, and Nielsen IMS software. They used it in an analysis of Online Video Share-shift to see how a particular brand benefited from moving ad dollars away from TV and into online video.
Now remember, this is a whitepaper by a video ad network. Much of the data is from following a CPG food brand. CPG is consumer packaged goods, FYI. So Doritos, Oreo, whatever.
Demographic Reach by Spend Shift Percent
The report looked at three specific demographics: Women 18-54, Women 25-34 and Men 25-34 and tried to determine the effects of shifting ad dollars to online video and its effectiveness versus TV spending.
The point of the whitepaper was to show how much more reach a company could have if they shift some money away from TV to online video. That's not saying you will if you do as there are numerous factors to take into account.
Here's a nifty graph that shows the brand reached an additional 1.8 million women aged 18-54, and half a million men and women aged 25-34.
Now in this exercise they didn't increase or decrease spending at all. They sued the same amount and simply shunted some over to online video. The results look promising, albeit marginally. The question here is: would it be worth the investment in time, energy and adapting of TV ads or creation of new ads for online use?
They then looked into several variations of spend-shifting in 5%, 10% and 15% range. They found that as the spend-shift increased so did the campaign reach, mostly.
Here's a breakdown of demographic percentage reached when spend shifted online.
- Women 18-54: 75.4 - 0% | 76.8% - 5% | 77.4% - 10% | 77.5% - 15% (overall: 2.1% change)
- Women 25-34 72.6% - 0% | 74.2% - 5% | 75% - 10% | 75% - 15% (2.4% change at 10% & 15%)
- Men 25-34 65.9% - 0% | 67.5% - 5% | 68.2% - 10% | 68.2% - 15% (2.3% at 10% and 15%)
They don't really say if that's the overall demographic in the US or if it's just of TV and online video viewers. I would guess that since they're using Nielsen data they are extrapolating that out to the full US population.
Brand Recall by Ad Type
The whitepaper states that brand recall rose 20% from 62 to 82 when ads went from TV only to TV and online combined. However, it said nothing at all about the ad spends (increase/decrease/ percentage shifts), the number of exposures or anything else aside from the fact that it was a 30-day window for adults over 18-years-old. So it's a statistic in a black box and we don't know any of the other factors. I've added it here just as a point of interest.
Budget Allocation Decisions
The YuMe whitepaper took a look at ad spend shifting to decide upon a recommendation but then said that it "is a question unique to each brand."
They did then go on to say that for the CPG food brand placed an ad on both TV and online in front of 20.6% of women (18-54) at the 15% spend-shift point. That's up from 11.7% at the 5% mark. However, exposed to TV only never dropped below 54% and exposed only online never reached over 2.8%.
For 25-34 the results were slightly better. Consdering that demographic is probably one of the most active online that comes as no surprise. Exposed to both ads started at 12% and rose to 20.7% with top exposure online only was 3.3% at the 15% mark and TV only dropped to 51.1% there.
The same held true for men but in a diminished capacity with the numbers being 9.1% to 15.6% on both, 3.5% online only at 15% and TV only 49.1% at that same shift.
This tells me that men are more likely to be online than women who still prefer television. These are broad stroke generalizations of course and are all dependent on this data from YuMe and Nielsen. I know many women where that is completely wrong and they don't even own a TV. Of course, Nielsen families are probably all TV owners, right? Obviously, if they looked at people who didn't own a TV it would skew the numbers more in favor of online ads.
Exposure, Reach and Drive
Almost sounds perverted the sub-header. The paper talked about how many numbers of exposure were good for a campaign to be successful. They cited 3 as some magic number with the graph (below) but offered no basis for that number.
The paper gives reach extension for 3+ and 6+ exposures and the 3+ is far higher, some 20%. From TV only at 54.8% to 58.2% (4.6% increase) at 15% spend shit for Women 18-54, 49.7% to 54% (4.3%) for women 25-34 and 39.8% to 43.3% (3.5%) for men 25-34. The graphs (below) had some strange growth numbers on them which I'm not quite sure what they meant.
They then went on to cite lower CPM rates across the board. Considering the price of TV advertising, that's just a given, right?
Not surprisingly, the YuMe whitepaper states:
Overall, the optimal scenario for the CPG food brand was to shift 15% of TV dollars to online video.
Here's their final analysis. Remember the odd numbers above when you read it.
Now, If you're looking to save money, then 10% shift wins out. If you're looking at average frequency, hey 10% wins out again. If you're looking at 3+ reach...yes, 10% is just as good (0.1% difference in two demographics). Overall reach differences are negligible between 10% and 15% ranging from 0.2% to 0.3%. In fact in two of the three demographics, 2-platform reach growth was highest at 10% spend-shift. Total GRPS, also best at 10%.