An Open Letter to YouTube

Subject: An Open Letter to YouTube
From: Phil Nottingham
Date: 12 Mar 2015

Dear YouTube,

I’d like to spend more money with you. A lot more money. My clients are spending significant amounts on TV advertising, which is becoming increasingly expensive and the value is hard to quantify.

AdWords has been a revelation for our digital media spend. Where once we were stuck to buying bespoke packages based on estimated impressions, real time bidding has allowed for a more efficient, targeted use of spend that means we can get our content in front of the right people at the right time. Alongside AdWords, Google Analytics has allowed us to track the performance and value of our adspend alongside our referring traffic from links, social media and organic search, meaning we can better understand our customer conversion funnels; and thereby create better ads which more directly serve user intent - meaning our money goes further and your users have a better advertising experience.

But, for some reason, we’re not there yet with video. The toolsets aren’t sophisticated enough, the tracking isn’t good enough and as a consequence, the content isn’t good enough and spend isn’t high enough. The majority of spend still sits with TV, where the creative fits in a tried and tested, broadly formulaic model. Therefore, with the little bit of online video advertising I’m working on, I’m just recycling formats made for TV, to fit into a pre-roll model that feels outdated, unnecessarily intrusive and increasingly ineffective.

If you could offer me a different model to this, an auction based, real time bidding alternative that allows me to break free of interruption ad breaks with the flexibility and accountability of AdWords, you can have all my money.

TrueView and AdWords for video are not good enough. The tracking is too low quality, the advertising options too limited and the quality of placements available not sufficiently exciting. However, with Chromecast bringing YouTube to the TV in a meaningful way and more and more people around the world holding a device capable of streaming video in the palm of their hand, I think you can get there.

I truly believe that with a few minor adjustments to your offering, YouTube can be the place where the biggest brands of tomorrow are built and the lion’s share of ad dollars spent.

So, without further ado, here is my relatively short list of requests...

Provide Full Integration with Google Analytics
While there’s currently some level of integration offered, the only pages we’re able to track are the channel pages: i.e.. http://www.youtube.com/user/DistilledSEO/videos & http://www.youtube.com/user/DistilledSEO/about, and this data is largely useless. What I really want is data on the traffic to video watch pages and playlists; pages where the majority of activity happens, and to be able to chop up and segment these views in exactly the same way as I do with traffic to my website. While YouTube Analytics offers a very broad overview of some of this data, it’s nowhere near sufficient.

Here’s what I specifically want to be able to do….

Segment demographic data against drop off rate and retention.

Right now, while YouTube Analytics will show me the age, sex and location of my viewers, I am unable to compare engagement and retention against these demographics. Until the available data is comparable, this is of limited value.

See the level of (anonymised) user data available in AdWords in GA.

I don’t care about keyword referral data, but I do want to know a bit more about the people watching my videos. You’re willing to offer this data in Adwords for video through the targeting options, so please make this data available in GA. Learning more about the interests and demographics of the audience I pick up organically will allow me to better understand who I should target for my AdWords campaigns, meaning I can then spend money more effectively, which will encourage me to spend more with you!

Provide full URLs in referring traffic report.

YouTube Analytics currently only shows details on referring sites and embedded players at a subdomain level. As I’m sure you’ll appreciate, having “Google”, “Disqus” or “plus.google.com” listed as top referrers isn’t terribly useful information…

Be able to accurately distinguish paid from organic referrals to my site from annotations & ad overlays.

It’s very frustrating to see people clambering for workarounds for what is a solved problem in all other areas of AdWords.

Be able to work out which videos have delivered referring traffic to my site and through which links (links in descriptions, annotations, ad overlays etc..).

This may be a caused by a site architecture problem - namely having all videos living on pages defined by parameters under the /watch subfolder… but, you’re Google! This isn’t tricky to fix! Having YouTube.com/watch listed as a referrer in GA is largely useless.

Allow us to add a rel=”canonical” tag to videos.
It’s extremely common that we’ll host a video on YouTube and then embed a video on our own site. While I think a lot of businesses take this as a default approach when they shouldn’t (i.e. with product videos) there are, nevertheless, completely sensible reasons for doing this - for example, with an informational video that we might create for a blog post.

In such instances, I want to be able to be sure the version of the video on my site is the one that will show up in the SERPs - not competing with the youtube.com instance and cannibalising rankings in the process. Just as I’m able to use the canonical tag to consolidate duplication across web properties I own, I’d like to be able to do this with my YouTube videos.

Are there any negative consequences of this for YouTube? Well, it means you’ll have to sacrifice consistently dominating Google search results, driving users instead occassionally to embedded versions of the video, rather than the video on YouTube.com. This may affect the amount of ads you can serve, but if you want businesses to continue to host and deliver the majority of their online video content with your platform, you’ll need to allow us to retain some algorithmic credit for our own content.

Moreover, this would be a wise move, because I don’t think you need to keep web traffic traffic on YouTube.com in order to benefit from it. Related video links allow you to serve more content (including ads) within the embedded player, the links on the YouTube logo and play bar will drive some traffic back to YouTube.com and ultimately, the kind of content that should be canonicalised e.g. product videos and blogging videos, are not the kinds of content you are monetising right now anyway.

Make your best features available to everyone, not just major advertisers.
Right now, the best advertising opportunities are the least accessible. Of the 12 different types of YouTube ad placements detailed on this page, only 4 are available through AdWords, with the rest requiring a dedicated account manager or access to DoubleClick. Practically speaking, this means you need to be a fairly big business capable of paying a dedicated ad agency in order to compete, just like with TV.

If I’m willing to invest in creating rich media display ads, or custom mastheads in HTML5 or Flash, I want to be able to bid for these directly, rather than through deals brokered with advertising agencies. I want to be able to have a Brand Channel custom gadget without spending significant amounts before a single user has seen it. If I want to have third party tracking on my ads, I don’t want to be restricted to a list of pre-approved vendors.

I firmly believe that you can take a big chunk of the budget currently spent on TV advertising, but you won’t do this by following the same exclusivist model. Cut out the middleman and allow all of the ad placements you offer to be bought through AdWords in a real-time self-serve auction that will encourage smaller businesses and challenger brands to experiment, innovate and spend more and more money with you as they grow into the dominant brands of tomorrow.

This was the approach that allowed Google to supplant the print advertising industry and it will be the approach that allows you to do the same with TV.

Give us more premium content to advertise against.
I’d much rather advertise my product or service alongside season four of Game of Thrones than a compilation of funny looking cats. Don’t get me wrong, I love cats, but I don’t particularly want to associate my brand with them. Users are likely to be more engaged when watching premium content, rather than throw-away UGC - and when they see the advertising billboards on the train or talk to their friends in a bar about the show they’ve invested time in watching, I want my brand to be top of mind.

Additionally, I need a non-arduous way of ensuring a quality threshold for the content my ads will be shown against. It’s not just that I want to avoid politics, porn and piracy, but I also don’t want my ad in front of content users react negatively to with high drop-off rates. You have the engagement metrics to discern good from bad (it’s a significant part of your ranking algorithm) and I want to be able to take these quantifiable “quality” factors into account as part of my bidding.

Let us bid for many more types of video placement.
I’m not a big fan of pre-roll video advertising. I don’t think any users are really big fans of pre-roll advertising. It’s ultimately, invasive, annoying and mostly serves only to directly prevent you from doing the one thing you wanted to do when you clicked the “play” button on a video player.

However, I appreciate that while the majority of your advertising inventory remains pieces ostensibly created to fit the 30 second TV ad spot model, pre-roll is going nowhere.

However, a way to get us thinking more creatively and spending more money would be to expand significantly past the in-stream TrueView model, allowing a much wider variety of different video placements.

Can you buy Mirriad and allow real time bidding for composited graphics integrated within the content itself? Can we have post-roll placements? Unskippable 6 second (Vine style) mid-rolls? Can we buy combinations of pre-roll and post-roll placements on the same video? Can we bid for a recommended video slot at the top of comments? Can we bid for ad slots in between videos on a playlist? Can I bid for a complete YouTube homepage takeover, but just for a very specific subset of users?

You have a ton of creative, smart people who can come up with loads more exciting ways for us marketers to think about advertising. If you challenge us to be better, and give us the data and the tools to support it, then I’ll shut up and you can take my money.

With regards and best wishes,

Phil Nottingham

Category: