Virtually every NewFront presentation this season talked about the fact that linear TV has lost 40% of its 18-49 rating points over the past four years. And, naturally, that their CTV and streaming media platforms were perfectly positioned as replacements in budgets of TV advertisers wanting to reach those younger-skewing audiences' premium, high-engagement video environments.
There's no question that connected TV generally, and ad-supported video streaming services specifically, are perfectly positioned to help advertisers fill that void. Same audiences; same kinds of content; viewed on same devices.
The short-term challenge with CTV today, as we all know, is that premium, fraud-free inventory is very scarce. Advertiser demand exceeds supply. The vast majority of streaming video time is on services like Netflix and Amazon Prime that carry no ads.
Fortunately, the media industry is working hard to fix that. Recent or coming ad-supported streaming launches include Peacock, The Roku Channel, Discovery+ and an ad-supported tier of HBOMax.
But another branch of the media industry is also moving fast to fill this void. Keep a close eye on developments in the console-driven video game industry.
In parallel with video companies’ efforts to build ad-supported streaming platforms, video game publishers are beginning to introduce TV-like ads into console-driven gameplay to tap already massive audiences.
According to Morgan Stanley research, 310 million people globally play console-driven video games on their TVs and PCs every month, and 124 million of them play daily for an average of two and a half hours per day -- and almost one-half of that group is in the U.S.
It makes sense. "Fortnite" regularly delivers movies, movie trailers and sponsored concerts to its players. Why shouldn’t other games offer their players permissioned, skippable 15- and 30-second TV ads that give them reward points they would normally have to pay for?
Late last year, Brian Nowak of Morgan Stanley wrote that console advertising on premium games could reach $2 billion per year in a few years, even if only 45% of gamers opt in.
For sure, linear TV is losing audience share among younger viewers, and for sure, ad-supported streaming services are going to capture a significant portion of those ad budgets over time.
But watch video gaming too. It doesn’t just reach audiences under 40; it absolutely owns them. As TV ads become mainstream within video game environments, the TV dollars will follow.
What do you think?
Dave, several points. As I keep explaining, the so-called national TV "targeting" currencies---adults/men/women 18-49 or 25-54 are not marketing targets for the brands involved. these are merely outdated GRP tonnage guarantee metrics from a time way in the past when these "demos" represented half or more of the broadcast TV networks' primetime audience delivery and the median age of their viewers was in the low 40s---when teens and kids were included. Now their median age is hovering around 58-60 years---The CW excepted ----and the vast majority of their average minute "viewers" are aged 55+. The cable figures are better and, unlike the broadcast situation, there are many opportunities to reach younger viewers at low CPMs on cable.
Interestingly, it has finally dawned on the sellers---with A&E leading the way---that it's stupid to guarantee audience tonnage on metrics that represent a shrinking minority of the audience so movement is underway to go to adults 18+ or persons 2+ as national TV's new GRP guarantee currency. This will greatly reduce the"rating attrition" that we see with 18-49 and 25-54. So "the problem" will be largely "solved". A little humor, here.
As for videogames being an alternative due to limited GRP availability in CTV and AVOD, that's true for now but if you are targeting mainly young males aged 10-30 and have little interest in those aged 40+, which seems to be the typical heavy user profile of videogame players, you are already there---I would hope.
Your premise seems to be that videogames might be a viable alternative for many other types of advertisers---those targeting the great mass of adults of both sexes and most age groups to some degree. Maybe---but I would think that a more likely alternative would also be whatever can be had at reasonable CPMs via CTV and AVOD plus, more cable and less broadcast. A different daypart mix might also help.
Ed, very good points. As you know, even if we don't agree on all things, we do agree on many things. For sure, buying different daypart mixes - with precision - across cable can get TV advertisers younger audiences. It is unfortunate that more don't yet use data-driven linear approaches to do that. However, we can't get away from the gross losses of audiences on linear TV and the high costs for marketers to reach some of those who are there. To get 18-34 Males at any real scale, even on cable, costs an effective CPM of well over $100, and there are fewer and fewer of them. For sure, video game ad dleivery will make a big difference here. Also, since console-driven video games are connected TV enviornments, you don't have to buy on sex/age demo's. Marketers can buy their targets and use first party data matching to do cohort targeting. That is where these new platforms will make the biggest difference.
These "new platoforms" have been around for almost 20 years. Nice that the industry is finally catching up as linear TV continues it's slow, predictible decline.
Very true Dan. The advertising industry likes to see itself as on the cutting edge of innovation and consumer behavior changes. Actually, it's pretty far back on the trailing edge of change.
Here's a thought.
Advertisers are prone to briefing out on a very specific 'demographic' for there multi-media campaign. For example, people who have purchased knitting wool and/or knitting accessories, or have knitted in the past 12 months. In essence they are bespoke demographics.
No medium can reliably and accurately provide audience measuremnent at that level of granularity for all advertisers. Nor could they afford it were that possible, and after all, who pays the bulk for the audience measurement. (Would the cost increase be worth it for the granularity requested?)
Online is the closest to 'bespoke', because it does a better job on recent behaviour (leaving aside the myriad issues around privacy, bots, frequency, recency and dwell time). The most effective is action-based metrics such as an on-line purchase, so online is a great direct sales tool. But targeting recent purchasers is like selling snow to eskimos ... sorry, already got some. And if a brand continues with its communications laser-targeted at recent purchasers then they are not effectively (or efficiently) introducing new purchasers to the brand resulting in brand usage growth and hopefully long-term brand and purchasing loyalty.
John, regarding targeting, if I were at an ad agency right now, I could take a client brand's product category and define the audiences of just about every nationally aired TV show by heavy, moderate and light or non users of the broduct and, in many cases I could single out which shows indexed high--or low---among current brand buyers as well as major opposing brands. More important, I could refine these product usage distinctions by the mindsets of the viewers---who is price conscious, who is health or fitness conscious, who cares about their image, who is a big foreign traveller, Who most often buys high ticket brands, who is on a diet, etc. etc. and appropriate combinations of these mindsets or behavior patterns ---all of which goes to the critical way the brand's particular positionning strategy is formulated and what message or sell it is using in its commercials. And I won't have to use Nielsen or the internet to do all of this. Yet, incredibly to me, here we sit debating the need for more refined metrics and how---or if---- the sellers will provide them.
My source would be an ongoing service that has been around for decades. It's now called Simmons/MRI and, yes, it's not electronic but is based on human responses. Nor are the measurements perfect---especially when it comes to TV program viewing claims which tend to be overstated relative to what Nielsen is reporting. However, directionally, the two services closely parallel eachother and there are ways to use the data if one is careful, that makes it very useful. So why isn't this resource used more often to deal with the targeting question? Is it because it's not electronic? If so, that seems rather stupid to me.
I totally agree Ed.
Here in AU we have two similar 'consumer' data sets that are heavily used.
The more frequently used (buyer, seller and marketer) is Roy Morgan Research which is a very broad questionnaire with a national random sample size of 1,000 people per week rolled up into a rolling annual data base of 50,000 (to dampen any 'noise' from week to week). It is a single-source data base - in essence product categories, demos, behaviours, ownership, income, politics. media usage etc are collected for all the individuals to get both profiles as well as duplication.
The other is Nielsen's CMV. It uses a smaller sample and fuses the industry 'currency' onto the respondents based on 'hooks'. It is not as broad but for those who require the precision of the currency it is used. It is not uncommon for both data sources to be used or 'cherry-picked'.
My guess is that all large media agencies and vendors use Morgan more than CMV.
John, great points. I totally agree that just segmenting past purchasers on an index basis to network isn't enough (yes, I agree Ed that it might be better than what is being done by many in the status quo). The online ad world has shown us the limitations (and annoyance) of too much retargeting of custoemrs. Building behaviaorl models that extend campaign reach into potential buyers who fit certain target audience characteristics but who are not already buyers is where TV helped build so many brands historically and where I believe that TV-like ads on video games will do now and in the future.
Research will be important. On the gaming front, we haven't seen a lot of mainstream media research, but I expect that to change fast, particularly since their is so much social media feedback from those audiences on Discord, Reddit and other channels.