Why There Won’t Be A Standard Digital Metric

Many people believe that it is important to develop a standard audience metric for the internet.  GRP’s have worked for TV, what should replace them on the web?    In fact, nothing needs to.

The rationale for a standard metric stems from the perception that one standard works very well for offline media, is what advertisers expect and if interactive media could produce a single standard it would attract more ad dollars.  This is clearly a misperception about how media is bought and sold.

To get a more accurate picture of what the future will look like, let’s take a realistic look at how media markets actually function.

What is GRP?

GRP, which stands for Gross Rating Point (also just called “ratings”), is basically the number of contacts expressed as a percentage of the population.  That’s it, really. For all of the technical talk, GRP is really just counting exposures.

This simple truth should clue us in to why a standard digital metric might be a bad idea.  GRP could easily be calculated for digital media now. If it’s really important, why doesn’t anyone do it?

In application, it does get a bit more complicated.  For instance, sometimes the population is really the entire population, but usually just a specific target group (or “universe”) is utilized.  So the same programming can have different ratings values depending on how it’s being evaluated.

In effect, GRP is itself a multiple standard.

Often ratings for a particular target group are referred to as TRP, but for our purposes we will call all ratings “GRP’s”.  In either case, GRP can be calculated simply by dividing contacts by the population whether that population is defined by demographic, geographic, psychographic or some other criteria.

Planning Metrics vs. Trading Metrics

While TV stations charge advertisers for contacts (expressed as GRP), what’s really important is to reach as many people as possible who can buy their products.  In turn, marketers expect a measurable effect on brand awareness that will translate into sales.

Although the primary goal of advertising is to reach people, coverage isn’t practical as a trading currency.  The reason is duplicating audiences.

For example, if we would buy two TV spots that both reached 10% of the population, we would not reach 20% of the population.  Some people would watch both spots so our total reach would be somewhat less.

There is no way to know how much less without an estimation of all the channels in the campaign.  That’s why coverage is used by advertisers and their agencies for planning internally, but not for negotiations.

However, two spots that reach 10% of the population will always be 20 GRP’s.  GRP’s can be added across different channels, brands and campaigns. It’s that mathematical quirk which has made ratings such a popular trading standard (just as impressions are in digital).

Because the web isn’t neatly divided into regulated markets, GRP isn’t as practical.  However, if anybody actually wanted to use GRP online, they surely could.  Nobody does.  If they did, it wouldn’t change much.

And, just like in TV, smart planners try to arbitrage between the metrics they use for planning (i.e. clicks or registrations) and the impressions they buy from suppliers.  The fact that suppliers can actually offer them those metrics online is an advantage that media owners never enjoyed before.

TV Trading without GRP’s

As useful as ratings are, it’s possible to trade without them.  In the UK, clients negotiate only their budget, their target group and their discount (everybody got a discount).

Here’s how it works:

An advertiser goes to the TV station and declares that she wants to spend 100 pounds on female audience.  The TV station would adds the  money to all the other budgets and divide it by the number of female GRP’s to get an average CPP.

It’s a bit tricky for people who are unfamiliar with TV, but two points are crucial to understand:

1. GRP’s are announced after the negotiations.
2. Everybody gets a discount off of the average CPP in their target group.

I like this example because it makes two good points about media markets:

Metrics are convertible: They UK system would only work if clients had a good idea what they were getting for their budget.  While TV buyers negotiate based on money rather than audience, they are able to trade effectively because they know that a certain amount of money is equal to a certain amount of ratings and coverage.

Multiple metrics work because everybody wants something different: The concept that everybody can get a discount of average CPP is a bit tricky, but crucial to understand.  It works because there are active markets for a variety of target groups.

If everybody wanted to buy the same target group the system would fall apart.  Yet, the system has effectively functioned for decades.

Audience Research Arbitrage in Magazine Markets

Magazines metrics are even less standardized than TV.  In the US negotiations are based on guaranteed circulation.  In Europe researched audience is more commonly used.  In both cases cost per thousand (CPM) is standard rather than GRP.

Again, most marketers use a different standard for planning.  Because magazines target niche readerships, audience attributes are the primary factor in magazine planning and CPM’s vary widely by magazine category.  For trade magazines, the size of the readership isn’t much of a success factor at all.

One reason the CPM standard persists in Magazines is because it favors marketers.  Through a mathematical quirk, magazines appear to be more expensive when measured by CPM.  By confusing a trading metric with a planning metric print publishers lose money.

Media agencies like to present CPM comparisons to magazine publishers in negotiations.  However, it is generally known among marketers and agencies magazines are up to three times less expensive for commonly used target groups.

This information is even available publicly through the WARC based on data gathered by the media agency OMD.  For some reason magazine publishers have just never bothered to check. So, often what looks like a metrics problem is just a failure to pay attention.

The reality is, everybody promotes the metric that benefits them the most.  Developing all-encompassing digital standard is neither practical nor desirable.  Moreover, it doesn’t accurately reflect the offline world that digital media supposedly wants to emulate.

What Will Be Important

It should be clear by now that metric standards are, for the most part, a false issue.  There will not be a standard way to present online audiences any more than there is a standard way to present offline audiences.  Everybody will, and should, use whatever metric works best for them.

However, there are some very important issues that do need to be resolved.

Research Methodology: While data on the web is abundant, standards for gathering data are weak.  As the digital market grows, this problem will only get bigger and more untenable.

Effective Frequency: One of the great advantages of digital media is that frequency can be directly controlled.  When an ad is shown too often, budgets are wasted and audiences complain.  However, there is very little understanding of how much frequency is optimal for online audiences.

Who to exclude? In offline media, target groups are largely about efficiency.  While a marketer might focus on a specific audience, it is clear that a lot more people come with a campaign than is specified in a media negotiation.

However, in digital media, targets are defined precisely – all others will be excluded.  Yet, there is negative value in making media targets smaller than the pool of potential consumers.

Marketers are going to need to understand buying decisions much better if they are going to get value from the new technology.

Who Implements? Another consequence of the technology is that advertisers are able to have an increasing role in campaign implementation.  This could be a big win for advertisers.

Who Wins from Multiple Metrics?

Ironically, while it is media owners that are mostly crying out for a standard digital media currency, multiple metrics are a big win for publishers.  They can package their offer to whatever benefits their inventory.

Social Media with low demand inventory can focus on performance metrics like CPA (cost per action).  Sites with strong brands can demand high CPM’s for advertisers to associate with valuable content.  Moreover, different areas of sites can be packaged and sold by different metrics.

Just like with the TV example in the UK, media owners win with multiple measurements because different people want different things.  It’s for the same reason that banks break mortgage loans into separate tranches with different interest rate behavior.

Most of all, because of the inherent properties of digital data, media owners can know more about audiences than advertisers do.  It will be much harder for agencies to play media off against each other (as in the TV vs. Press example shown above) when everybody has access to the same data pool.

So while digital media research is an exciting area with many important developments still to come, a standard digital metric won’t ever happen – nor should it.

– Greg