Despite its increasingly prominent role in the media mix, social media, and in particular the major walled garden platforms, can be somewhat opaque to advertisers. In our ongoing quest to foster transparency across the digital ecosystem we sat down with Brittany Richter, VP Head of Social Media at iProspect, to get some insight into how social is bought and sold, the metrics that matter, and where transparency is helping to foster growth.
What metrics do you most commonly buy social media on?
Brittany Richter: We don’t buy on one specific metric. It varies depending on the client and their specific business objectives. We primarily try to stick to the desired user behaviours that are leading indicators of the business results our clients are looking for. For instance, optimizing on CPC or ROAS to get the highest conversions is something we might do for an e-commerce client. In some cases, where we’re trying to focus on new customer acquisition, it’s highly qualified traffic then we might be bidding on CPC instead. In any of those scenarios where video is an element, then we’re buying on the cost per video view.
Are those the same metrics that your clients are focused on?
Yes, for the most part there’s not much of a disconnect with our clients, especially on Facebook and Instagram.
There’s dispute among agencies as to whether YouTube should be primarily bought by the video or social team. Does your team handle YouTube buys?
We don’t have a separate video team. Typically our display or search team will buy YouTube since it’s bought in a similar way and sometimes in the same interface, but the social team does buys it as well. It also depends on how the client views YouTube [as a social or video product]. We tend to take our lead from them. In any case, all teams are trained to handle.
Are YouTube buys most commonly viewed as a social buy or a video buy?
It’s most often viewed as a video or display buy.
Can you tell us which social platforms you see as most important, and perhaps which are more second tier or niche?
There is some variance, but generally speaking we see the strongest performance against all key metrics on Facebook and Instagram. Every once in a while we have a client or a particular audience that we are running against for which video performance, CPM or CPC, might be slightly stronger elsewhere. But that performance is not usually available with the same scale that we can get on Facebook or Instagram, so that’s generally where we see the best performance. After Facebook and Instagram it varies widely by client whether we invest in Twitter, Snap, Pinterest or LinkedIn.
Do you find that clients are usually aligned with your assessment of the platforms?
Typically, we are aligned. If we’re not aligned early on then we get to a place, through strategy and education sessions, where we become aligned. But that does come up a lot. We’re asked why we aren’t using a specific platform or tactic, but then we explain it and then we are aligned after that. With that said, sometimes there’s still a desire to spend on specific platforms. It may not drive the best result from a pure performance perspective, but there are other things to be gained from running there. Sometimes it’s as simple as people not wanting to put all of their eggs in the Facebook basket. In other cases it’s a desire to make sure that we’re testing the other platforms at all times so that if we do see performance start to increase, or an opportunity for a particular activation, that we are able to adjust to that.
Do media quality metrics like brand safety, viewability, and fraud factor into your ranking of platforms?
All of those things are a factor. Luckily, the majority of our placements are only served if they’re in full view or 50% view within the newsfeed so we feel there’s a pretty low risk when it comes to viewability as there’s no such thing as an impression firing below the fold on social, so that’s a topic that comes up but it’s less of a concern in evaluating platforms. Fraud and brand safety also come up, but with the platforms making the moves they are to allow for different types of exclusion targeting–whitelisting or blacklisting different publishers and placements on the different buys–it’s a prominent topic, but something that feels relatively handled in social.
There’s seems to be a spike in concern around brand safety, does that map to your experience?
Yes we’ve seen an increase in attention paid to brand safety, and that’s promoted a deeper evaluation and more questions asked of the platforms. But, [on social] the content lives within a feed and it’s generally not adjacent to other brand content. Instead it’s adjacent to user generated content that we, and our clients, feel is low risk for misattribution. A post from your best friend below a post from a brand, for instance, isn’t perceived to be related to that brand. We have the conversation more often recently, but there doesn’t seem to be major concern with in-feed placements for most advertisers. However, when we look at things like in-stream placements, both Facebook and Twitter started making strides to offer pre-roll type media buys last year. In those placements there seems to be even more concern around the ability to scan and properly tag video. Whether brand safe or not, it still feels risky to our clients, so there we do spend more time with the publishers or the platforms trying to evaluate before making a buy.
How important is it for social platforms to submit to third-party verification?
I think we’re going to get to a place where it’s going to be required. If we’re doing it everywhere else, then it doesn’t makes sense that we’re not also tracking for it in social. I think that the risk is lower other than in those in-stream video placements but I do still think it’s going to become a standard across all media. Even though fewer clients are pushing for it right now because they feel like the placements are lower risk. Right now they understand that you can’t have view tags or you can’t always have viewability third-party tracking, but it’s only going to increase.
Broadly speaking, which metrics or KPIs are most important to you in a social campaign?
It varies, but it’s not based on social it’s all about the client’s business objectives. We feel very strongly about not doing social for the sake of social. So most of the metrics that we would use to measure social campaigns should be the same or similar metrics that we’re using in display, video, search or any of the other media that’s in market. A client may want to drive online sales vs. offline sales vs. change in perception vs.leads, so you can’t really rank the metrics as a whole across our client set.
Are there any metrics that you think you need that you aren’t getting from the major platforms?
Would a single viewability measurement standard across platforms be useful or should platforms be evaluated individually on performance?
I think it would be incredibly useful if there was a set of basic metrics that were each defined the same way on every platform. I think where it becomes really tricky, and frankly impossible today, is that the consumer behaviors on each platform are different. The types of action that happen on Snapchat versus an Instagram story versus Facebook are different, and we have to evolve and adapt which metrics we’re using based on the options and the actions that are available to consumers. So, as long as they continue to be different I don’t think that one standard could work but it would be great if certain other basic metrics were all defined the same way simultaneously.
Do you feel that any of the platforms have specific measurement blind spots?
In-store sales are typically really hard to track. If the clients don’t have the infrastructure in place for that, then it can be a hurdle. Sometimes it’s possible and easy, other times not so much.
Anything that’s attached to a brand study can be challenging because it can be just one element of the campaign that skews the results. There’s not a lot of flexibility, which is understandable because they need to hit a certain sample size to be statistically significant and they can’t be tampered with or you’ll introduce compounding variables. But it does make it difficult to run brand studies when you’re also trying to be flexible and swap out creative and do other types of testing for performance.
You’ve been in social for most of your career, can you point to any trends that have emerged that surprise you or any missed opportunities on the part of platforms?
I am surprised that there hasn’t been more effort around programmatic. Facebook sunset their FPS product, and although Twitter tried to go programmatic last year, but there just doesn’t seem to be a lot of focus on that. I think it would be a huge opportunity and it would cut down on the labor of managing campaigns on these platforms.
We also haven’t seen a lot of dynamic creative optimization products emerge they way they have in display which I think is a missed opportunity.