EMV: MVP Metric Or Total BS?

Twenty years after the launch of Facebook, questions about the ROI of social media marketing haven’t been settled. In an effort to suss it out, influencer management platform Tribe Dynamics applied the metric earned media value (EMV) to social media content about a decade ago. EMV measures the value of promotion or coverage from third-party sources, including influencers and news publications on digital channels, and is usually calculated by multiplying the impressions a piece of content receives by cost per mille (CPM), the amount a company pays to reach 1,000 people.

Brands use EMV to gauge awareness, particularly in relation to their competitors, and help in the allocation of marketing dollars, but the metric, like the influencer marketing it tracks, has drawn haters. They argue it’s based on faulty assumptions that make it misleading and possibly detrimental to a business. Beauty Independent spoke to one such hater, Pierre-Loïc Assayag, co-founder CEO of influencer marketing software company Traackr, about the problems with EMV, how he thinks it can hurt emerging brands, and what he relies on instead. 

What’s EMV, and how is it used?

A huge challenge of influencer marketing and social media at large is the distance that there is between measuring your outcomes and measuring the results of your work. Outcomes are measured by way of market share sales. You can even use proxy metrics like volume of Google searches or traffic to website where DTC brands, particularly with brands in general, have figured out the worth of a visitor to the website as an example. These are business outcome metrics.

When you work on a social campaign and creators produce content for you, it has a lot of engagement and things seem to be going well unless you do nothing else, but connecting that campaign to sales is a nightmare. In today’s world, with a bunch of social platforms and a bunch of formats and creators, that direct attribution is very complicated.

Where EMV has established itself as a useful tool is in creating a proxy for success in saying, if I have a bunch of creators talking about my brand and there’s lots of views and engagement, I’m going to get the higher EMV. That’s the reason why it asserted itself quickly in this space, and the fact that there was a dollar amount in front of it gave the impression that it was valuable.

We can talk about the reason why it’s opposite to value, it’s a practical cost, but that’s how it asserts itself because there’s a need for that proxy metric for success that then you can figure out a way to hopefully connect to yourself. You still haven’t got an answer to that question on how you connect them, but at least you get closer to some kind of a performance metric.

What do you see as EMV’s fatal flaws?

There are so many unstated assumptions that you’re making that offer a very murky view on what performance looks like. At the same time, because you don’t really know how it works, you end up using all kinds of shortcuts as a business in order to affect the optics of EMV instead of affecting the outcome of your business. The intent behind EMV is great. You need a proxy metric. In our world, we call this an output metric.

There are three layers of KPIs you can use. There’s an activity level KPI, that’s inputs, the stuff you control. How many posts am I going to write today? How many people am I going to connect with? How many outbound emails am I going to send if I’m in sales, etc.? I’m in control of my input.

On the other side of input is the output, the thing that is a direct correlation between my activity and the success of my activity. I would deem EMV an output metric. Then, there’s the outcome metrics, which is the business impact of all the output across my business that I put together. 

I used to work in management consulting. The framework I just described I did not invent. It’s been around for many, many years as a very simple way of understanding how to measure your business from the things you control the most to the things you control the least and making assumptions along the way. 

In my world, when we started analyzing EMV, we realized that not only is it opaque, but it’s also very easy to game. You end up affecting the behavior of the team that don’t know how to improve their market share or sales in a very straightforward way, but they know how to improve EMV and when the behavior changes, even though it was a loose correlation between your EMV results, so your output metrics and your outcome metrics, that correlation disappears because you treated a correlation as a causation that was not there. 

We’ve seen this firsthand with some of our clients. That led us to the path of helping them find a better way. We saw clients that would tell us, “I am investing in this creator because he has a high EMV. I know in my gut that it’s going to have zero impact on our business, but my boss is holding me accountable to an EMV target this quarter that I need to meet otherwise I don’t earn my bonus.” 

People have EMV-dependent bonuses?

On influencer of marketing teams, yes. So, you end up doing all the wrong things, even though you know they’re wrong because you feel that you have no other choice. You need to move away from this.

In addition to that, in the way the organizations tend to get structured, the EMV targets are more assigned to the upper funnel teams. The brand team tends to handle PR, social, etc., and the lower funnel that do e-comm performance marketing are held accountable to much more tangible business metrics like the outcome metrics.

There’s this circular reasoning around EMV that doesn’t really yield as much results as you could on the other side. The less money you’re investing in upper funnel because you can’t demonstrate the actual results, the more money you putting towards a down funnel, on conversion and transactions. At the same time, intuitively, particularly for indie brands, the magic happens with the brand.

You need your true north, you need your purpose, you need to define what you’re about. So, invest in things that you don’t expect an immediate return on, but if you’re using the wrong proxy for this, you end up moving all of your investment down front on conversion because that’s all that you can control.

How do you respond to people who say Traackr is biased because it has a competing measurement?

I have an alternative to EMV that today is not the industry standard. So, we do have a vested interest in clients thinking differently. Now, there’s a reason why this system came to life instead of us just simply doing whatever everybody else was doing. Most of our key competitors have EMV as the asserted framework that they’re using. They infer much of everything they’re doing through the lens of EMV. 

The system we’re using that we have a patent on is public, and everybody can use it for free. It’s not a secret sauce. We’ve made it as easy as possible for anybody to use. It’s called the Brand Vitality Score, and the acronym we’ve used for this is VIT, and it’s visibility, impact and trust. We compile the score around visibility, the number of eyeballs, impact is proxy for engagement, and trust is, how much trust do consumers have in that person’s content? 

What are the main ways you’re trying to remedy what’s wrong with EMV?

First and foremost, the fact that we have a completely transparent open view into the way things are compiled and structured allows our clients to understand every single lever behind the score. Just the visibility and the transparency is hugely helpful so that they can connect input and output.

The second is that, when we launched this score, and we still do this on a regular basis, we have connected it to outcome metrics. We can’t, for example, [say] this post has X number of points of VIT. How many sales is that? We can’t tell you that, but what we can tell you is that there is a very strong correlation to the way the score is compiled and outcome metrics that we’ve studied with our clients and outcomes by way of sales, of traffic to one’s website, of additional followers on their social channels, Google searches, the KPIs that these clients are using to measure their business performance.

We have tuned the weights in VIT so that it correlates. Again, we have to be careful of the causation jargon on this, but it correlates. The transparency and the constant reviewing of this matters because the platforms keep launching new algorithms. As a result, our weights ought to change as well to continue to reflect the connection with the down funnel outcomes.

We structured the deconstruction of VIT into modules that our clients can use and say, “OK, I have X amount of performance in VIT, how can I improve it? How do I tally up to my competitors? Where do they do better than I do? What’s special about my brand?”

As an example, if you’re a brand that is looking to improve your performance around purpose, instead of looking at your very direct competitors, you’ll be looking at brands like Dove that have been doing this for a while. What is the deconstruction of their performance around the VIT? Where is my ability for the brand, what can I map to on this and which partnership should I be pursuing, what strategies? They’re able to use as a way to set strategy and monitor the impact of the execution of the strategy. 

Can you give an example of how gaming the system in regard to EMV would look? 

Every creator in a platform is going to have a set EMV next to them. If you have no experience in working in influencer marketing, you’ve built no partnerships, the shortest path to increasing your EMV is to pick the top 10 creators, pay them a hell of a lot of money so that they publish on your behalf.

As a result, your benefit with some of their EMV, it’s probably going to be subdued, but it inflates the cost per post because everybody’s looking at the same data with the same people. It just offers none of the details that will tell you whether or not they’re a good partner to you. So, that’s the shortcut you can find to faking good performance. 

The opposite is much more insidious, and this is not ill intent on the part of the practitioners, it’s just a lack of understanding and knowledge on how things get calculated. EMV is the stepchild of advertising value equivalent, the old PR metric, where it basically gets measured by the size of the article. How many columns, and how it was a physical thing from the newspaper that you multiply by the advertising cost depending on real estate for that newspaper?

That’s your AV, debunked 15 years ago by the application of analytics, but it was for its time an OK metric because, how else are we going to measure PR? In today’s world, EMV has inherited a lot of the criteria from AV. At its core, [the brand] gets points per post no matter the viewership. 

When you think about the impact on measurement, and it’s particularly true for Stories because there aren’t many data points around Instagram Stories by way of engagement or even views that you track, if you get a million Instagram Stories posted, you’re going to get millions worth of EMV from Stories that nobody’s going to see. How do you do it? You have to blast a whole bunch of creators, offer them free products, ask them to post, and you don’t care about whether or not it worked because the mere fact that they published it, even if nobody watched it, gives you the performance data that you’re seeking. 

You end up creating these massive seeding programs that all beauty brands have to work with. For what outcome? If you’re a small brand and you send product every other week to a thousand creators, crowd your office space with boxes, have a bunch of people who spend their full day doing this, what’s the ROI on this? It’s the EMV points I just mentioned on Stories. There’s a level of absurdity on how it drives behavior just because of that calculation. 

Is EMV better than nothing? 

Probably yes. In the past you would measure the number of followers that a creator had to know whether or not you wanted to work with them. Is EMV better than that? Sure, but it has so many shortcomings and the practice has become so complicated over the years that I don’t think that it’s a measurement framework that has adapted to the age that we live in.

Influencer marketing started as an awareness tactic of flooding the space and getting as many posts as possible very similar to PR where it was do as much as you can volume, activity. That metric made sense then. With the complexity of influencer marketing, it doesn’t make as much sense.

And you have people that have built these elaborate practices and systems around a flawed metric, back to the seeding campaigns to thousands of creators. By no means do I want to take a dig at the people that receive free products. Maybe they should, but EMV is not going to tell me whether or not they should.

How can brands start to measure influencer marketing initiatives in a more effective way?

We have a bunch of people that are realizing, “The work that I’m doing, I don’t think it’s working, but my boss is telling me I have to do it.” It’s just the way it goes. Be able to take a little bit of distance between the actions and your campaigns, educate your management that the data that you’ve been looking at and sometimes pitching for the past decade is actually flawed.

It’s not an easy transition, and this is where I think it’s incumbent on the key people in the industry—investors, leading brands—to change the narrative and say this is BS. We are not using that KPI, we are looking at another set of KPIs. This is how we’re going to measure the value of your business from this point onwards. I think there’s a mind shift that needs to take place that we are just at the beginning of.