Mr. Magazine

This week we focus on one of the more delicate subjects in the publishing industry: the relationship between data analytics and editorial. Luke Magerko, from MagNet, will walk us through how he foresees the relationship between these groups and how it can increase newsstand sales.

DO YOU HAVE ANY COMMENTS ABOUT LAST WEEK’S AAM REPORT SHOWING YET ANOTHER HALF OF SALES DECLINES?
There were outside factors that caused sales declines, including weather in the first quarter and Source Interlink closing in the second quarter. National economic trends and supply chain disruptions are uncontrollable; publishers must focus on the things that produce positive results.

WHAT DO PUBLISHERS CONTROL AT NEWSSTAND?
The magazine cover is, without question, the most important marketing piece in publishing and editors deserve straight analytics to help understand newsstand trends and how those trends affect all parts of consumer marketing.

WHY IS NEWSSTAND SO IMPORTANT TO THIS PROCESS?
Consider this: newsstand sales and subscription direct mailings are similar to a simple survey with the implied question: “do you want to buy this product?” For our purposes, there are two relevant terms relevant in survey methodology: simple random sample and voluntary response.

Newsstand is similar to a simple random sample. A typical grocery store provides a sample representation of the local community. To sell a magazine, the editor must compel that person to pick up the product and make a choice. The ability to accomplish this tough task makes newsstand sales results extremely important to understand.

Direct mail has more in common with a voluntary response survey. Anyone familiar with internet surveys know what a voluntary response is; the respondent is volunteering to provide information. Subscription modeling has turned into a form of volunteer response because consumer marketers are very effective at identifying their target audience and selling only to them. While this is an efficient business model to acquire subscriptions, the consumer insights might be skewed.

HOW DOES THAT AFFECT A PUBLISHER?
If an editor believes the magazine’s audience signifies a certain cluster of society, the product will reflect the cluster. Newsstand can similarly identify customer clusters at retail. My question for you: what if editorial believes their audience is a high-end suburban Target shopper but they sell most successfully to rural Walmart shoppers?

THEN THEY CONFRONT THEIR NEWSSTAND STAFF FOR PLACING THE COPIES IN THE WRONG STORES!
Exactly! Editors have an understanding of their audience. Newsstand analytics is designed not to change their beliefs but to confirm or deny those beliefs based on newsstand sales.

EDITORS (RIGHTFULLY) ARE SUSPICIOUS OF ANALYTICS WHEN IT COMES TO COVERS!
And they should be! I have read a dozen articles on what makes and effective cover and heard over 20 newsstand veterans opine on what makes a good cover. EDITORS: DO NOT LISTEN TO ANY “EXPERT” EXPLAIN WHAT WORKS ON A COVER! YOU ARE THE ONLY EXPERT ON YOUR TITLES!

I suggest editors read an article by Antoine Boulin. He wrote this week about the relationship between data and editorial. I appreciate his point of view. He writes a very important paragraph in his column:

“Data informs editorial decisions. It shouldn’t define them. A content strategy needs to be shepherded by content creators — those with expertise in creating high-quality content that traffic drivers such as Google or Facebook reward. If you let data lead editorial, you might see some short-term gains in scale, but, long-term, you’re more likely to be punished.”

HOW DO YOU FORESEE A RELATIONSHIP BETWEEN EDITORIAL AND ANALYTICS?
MagNet has a repository of analyses that can be used determine the effect of cover attributes on sales. We recently worked closely with a top 100 publisher on cover treatments. Here are our best practices in the editor/analytics relationship:

1. Let the editors speak – Editors understand nuances on their covers and point out what they were trying to accomplish. This insight deeply affects what should be analyzed.

2. Let the editors play with covers – MagNet designed a Cover Analyzer to encourage an editor to peruse all competitive covers. It is remarkable what patterns an editor can detect just by seeing all the covers in one place.

3. Editors have questions, answer those first – After using the Cover Analyzer, our publisher had ideas on what worked for specific covers. Those ideas should be the foundation of the cover analysis.

4. Editors “tag” cover attributes – Tagging is defining each cover attribute. After editors provide questions, the art director or editor must sit down and walk an analyst through each component of the cover to ensure the analyst is looking at the right attributes.

SEEMS SIMPLE ENOUGH. CAN YOU SHARE FINAL RESULTS?
Out of respect to the publisher we cannot but I will show you what one of multiple exploratory data analyses (“EDA”) results look like.

The publisher was interested to know what types of blurbs succeeded on the magazine cover. The publisher identified eight types of main blurb theme and MagNet compiled three years of results highlighting overall sell-through percentage:

Results show average sell through percentages and also the maximum and minimum sell through percentages. Theme Type 3 is the weakest and Theme Type 2 is the strongest, with one issue doing exceptionally well.


SO THE EDITOR SHOULD FOCUS ON THEME TYPE 3!
That question is why editors get nervous around the data. There are multiple factors not included in this analysis but this is a good starting point to discuss the impact of the main blurb. MagNet suggests using this as the beginning of a conversation, not an end.

WHAT HAPPENS IF EDITORS DISAGREE WITH YOUR FINDINGS?
I created over 150 in-market tests in my career designed to address just that point. Sometimes editorial intuition and data do not match and in those instances, MagNet suggests in-market testing to get a better understanding of the results.

HOW WAS THIS INFORMATION RECEIVED BY THE PUBLISHER?
The data confirmed the intuition so adjustments will be made on an ongoing basis.

ANY RECOMMENDATIONS FOR EDITORS?
Yes, trust in the data if it is provided with no agenda and guide the process of analytics because you are the experts and also the analyst’s client.

THANK YOU LUKE!
MAGNET WILL PROVIDE A FULL DEMONSTRATION OF COVER ANALYTICS AT THE MAGAZINE INNOVATION CENTER’S ACT 5 CONFERENCE AT THE UNIVERSITY OF MISSISSIPPI OCTOBER 7 – OCTOBER 10. TO REGISTER OR CHECK THE AGENDA CLICK HERE.