When Stats Come Up Short
October 26, 2011 by Donnie Clapp
A few days ago a coworker sent me an eMarketer article titled Social Media Comes Up Short for Agency Prospecting. For me, the piece fell well short of supporting that title or even moving the conversation forward very far, and I'd like to discuss the reasons why.
The post was an attempt to take some interesting new research, in this case from RSW/US and RSW/AgencySearch, and distill it into some sort of new take on the state of the world. A noble goal, if done well.
Full disclosure: I’m not a statistician or a researcher, but I am a person who very much appreciates good statistics and research. Both can give insight, change how I think about my job, and drive my strategy for clients. And this research from RSW is well-done and potentially very useful. It is only this particular analysis of that research that has me frustrated.
The premise of this eMarketer piece was that social media is not being used effectively by ad agencies to generate new business. A topic with a fairly narrow audience, but potentially with insights that could be extrapolated to other uses of social media. Here are my problems with eMarketer’s execution of the idea:
No Frame of Reference
The author opens by saying that despite increases in online U.S. ad spending, some portion of U.S. ad agencies are having trouble generating new business:
eMarketer expects US online ad spending to reach $31.3 billion this year, growth one imagines would position ad agencies for gains in new clients and business expansion overall.
But according to new business relationship management firms RSW/US and RSW/AgencySearch, more than a third (35%) of US ad agencies are having trouble generating new business even in spite of such spending growth.
That’s super interesting, but what were those stats the last time we checked? What was online ad spending last year? What percentage of U.S. ad agencies are always in decline or struggling to generate new business? If less than ⅓ of all agencies have historically been on their way out at any point in time, that’s one thing. But I could see it being the case that 30-35% is just the normal portion of agencies who are performing poorly during any given period – indicative of industry turnover, if you will.
Also, is there any historical data that ties ad spending to this “Struggling to Generate New Business” metric? Is it possible that ad industry growth has never shown any correlation with online ad spending, for instance because so many small companies buy their own online ads?
These questions basically boil down to one criticism: It is the author’s job, literally, to add context to the data so that it is more useful to me. That has, to a large extent, not been done.
Biased Analysis
It seems pretty clear to me that someone at eMarketer saw this new research come across their desk and thought it was interesting enough to warrant a post. They surmised that this data could shed some light on the question of social media ROI. They were probably right.
However, they made a mistake in assuming that the data conclusively proves either that social media is an effective tool for new business acquisition in the ad industry, or that it is not. There is a third option - that the data is inconclusive, or merely inches the story in one direction or the other, without providing a black and white takeaway to use as a juicy headline.
Take this quote:
Most agencies attributed 10% or less of their new business revenue to social media outreach. In fact, 88% of agencies are seeing 30% or less of their total revenue coming from social media outreach.
Again, these stats are interesting, but is the premise really supported? One way to read this is that 12% of U.S. agencies attribute more than 30% of incremental revenues to social media. That seems like a lot. It’s not hard to imagine a separate article by another analyst claiming that this data proves that social media is becoming a huge part of the new business pipeline.
The real problem with these numbers, though, is that all we are given to compare them to is this graph:

So, what portion of U.S. agencies spend 30% or more of their new business time and budget on social media outreach? We have no clue.
In fact, if the data actually contained this comparison, it would be amazingly useful. A direct (if subjective, in the eyes of survey participants) correlation between the percentage of available resources dedicated to social media and the perceived benefits. For instance:
“While 12% of agencies reported 30% or more of new business revenues resulting from social media, 23% of agencies said they spent more than 30% of available time and money on social media outreach.”
Headline Baiting
Here is the closing paragraph of the article:
Overall, agencies have room to improve how they approach and initiate conversations with potential clients. Though the majority of marketers prefer to make contact with agencies through email, agencies shouldn’t abandon their social media-based lead generation practices. Rather, agencies would be wise to increase the quality of their email content and social media conversations in order to deliver what marketers want through multiple touchpoints.
Does this support the headline - Social Media Comes Up Short for Agency Prospecting? No. In fact, the author is backtracking and then giving advice so general it could accompany virtually any research on the subject, positive or negative.
I don’t mean to pick on the author of this eMarketer article, and I do not think that this and other articles like it come up short because the people that write them are not talented or capable. I think that they are under pressure to produce one thing, while their audience is actually after something different.
The 24-hour news cycle has decreased the quality of television news at any given moment to somewhere below the threshold of usefulness. I think the infinite uphill battle of SEO has done the same thing for the myriad of blogs and industry analysis websites out there. None of their content is thought about long enough before the “Publish” button is clicked to be of any use to anyone. There is an infinite stream of articles and posts, constantly filling our RSS readers and email inboxes and LinkedIn feeds – enough to make us all feel like we could never have enough time to keep up with it all. But that guilt is misplaced. If we were truly getting smarter and better informed every time we took the time to read an eMarketer article, we wouldn’t feel so drained at the end of the day; we’d feel energized - ready to apply these new-found insights to the future.
So, analysts of the world: unite. Quality over quantity. Tell me a complete story. I’m looking forward to it.
