Microsoft word - volume of media cove#209718.doc

Exploring the Link Between Volume of
Media Coverage and Business Outcomes
Angela Jeffrey, APR
Vice President Editorial Research, VMS
Dr. David Michaelson
David Michaelson & Company, LLC
Dr. Don W. Stacks
Professor, School of Communication
University of Miami
Members, Commission on Public Relations Measurement & Evaluation
Published by the Institute for Public Relations Exploring the Link Between Volume of Media Coverage and Business Outcomes By Angela Jeffrey, David Michaelson & Don Stacks Copyright 2006, Institute for Public Relations Angela Jeffrey, Dr. David Michaelson, and Dr. Don W. Stacks Angela C. Jeffrey, APR, is Vice President Editorial Research, VMS, and a member of the
Commission on PR Measurement & Evaluation. Jeffrey spent two decades in PR, advertising
and marketing with the J. C. Penney Company and major national agencies before starting
ABSTRACT
Jeffrey Communications. She founded PRtrak to provide automated media analysis reports.
In 2002, she sold the firm to Surveillance Data, Inc., which enhanced the platform by Since the founding of public relations, practitioners have intuitively believed in the power adding “Share of Discussion” analysis. In 2005, New York-based VMS (Video Monitoring of positive unpaid media to affect behavioral goals. But does it really work? If so, how Services) acquired PRtrak. She has won several PRSA Silver Anvils and IABC Gold Quills, much coverage is needed? This paper suggests a strong relationship between coverage and in 2006 was named a finalist for PR News “Measurement/Research Expert of the Year.” volume and business outcomes, with three case studies looking at volume alone; In addition, her firm shares in Porter Novelli’s 2006 Jack Felton Golden Ruler Award for tonality-refined volume, and message-refined volume. excellence in public relations measurement and evaluation. Angie graduated summa cumlaude from Southern Methodist University with a BFA in Journalism and BBA in Marketing.
David Michaelson, Ph.D., has over 20 years experience conducting high quality,
actionable research for Fortune 500 companies including Johnson & Johnson, Merrill Lynch,
Coca-Cola and AT&T. Michaelson won two Silver Anvils from the Public Relations Society of
America and two honorable mentions for best use of research from PRWeek. Prior to
founding his own firm, Michaelson was Managing Director and Head of Research for Ogilvy
Public Relations Worldwide. He served as Managing Director at GCI Group and Director of
Research at Burson-Marsteller. He holds a Ph.D. from the New School for Social Research
and is a Phi Beta Kappa graduate of the University of Massachusetts. Michaelson is a
member of the Commission on PR Measurement & Evaluation.
Don W. Stacks, Professor and Director of the University of Miami School of
Communication Program in Public Relations, received his Ph.D. from the University of
Florida in Communication Studies in 1978. After graduating from Northern Michigan
University (B.A. in English/Speech), he served in the U.S. Army as an intelligence briefer
and production specialist. During this time he co-founded the Journal of Applied
Communication Research. In 1975 he received an M.A. from Auburn University (Speech
Communication). Stacks previously taught at the University of South Alabama and the
University of Alabama-Tuscaloosa, with visiting professorships at the University of Georgia
and the University of Alabama at Birmingham. Stacks is a member of the Commission on
PR Measurement & Evaluation, and a Trustee of the Institute for Public Relations. Teaching
honors include the University of Miami's Professor of the Year Award. Research awards
include the University of Miami's Provost Award for Scholarly Activity, the Institute for
Public Relations' Pathfinder Award, and PRSA’s Jackson Jackson and Wagner Behavioral
Science Prize.
ABSTRACT
Since the founding of public relations, practitioners have intuitively believed in the powerof positive unpaid media to affect behavioral goals. But does it really work? If so, howmuch coverage is needed? This paper suggests a strong relationship between coveragevolume and business outcomes, with three case studies looking at volume alone,tonality-refined volume, and message-refined volume. Exploring the Link Between Volume of Media Coverage and Business Outcomes By Angela Jeffrey, David Michaelson & Don Stacks Copyright 2006, Institute for Public Relations INTRODUCTION
Since the founding of public relations as a discipline, practitioners have intuitivelybelieved in the power of positive unpaid media1 to affect the cycle of awareness,knowledge, interest and intent2 that is needed to achieve behavioral goals. EdwardBernays, who is often referred to as “the Father of Public Relations,” demonstrated theway to manipulate public opinion was through the indirect use of “third partyauthorities,” such as the media: “If you can influence the leaders, either with or without their consciouscooperation, you automatically influence the group which they sway.” 3 While the field of public relations has matured far beyond this early one-way model ofcommunication, influencing the media remains the largest focus of the $5.4-billionworldwide PR industry today.4 But does it really work? And if so, how much media coverage is needed to affectchange? Like the old adage, “if a tree falls in the forest, and no one is around to hear it,does it still make a sound?”… is a company successful because it builds a terrific productor behaves as a good corporate citizen, or because people know about these thingsthrough the communicated word?” What good is the most provocative message if no onehears it? This paper will explore the importance of media coverage volume in the effectiveness
of a public relations program. According to the The Dictionary for Public Relations
Measurement and Research, “Public Relations Effectiveness” is “the degree to which the
outcome of a public relations program is consonant with the overall objectives of the
program as judged by some measure of causation5.” While the challenge is truly finding
‘proof of cause’ in any PR campaign, we can still look for a “preponderance of evidence”
of a meaningful relationship (correlation) between two variables. In this paper, the
relationship would be between coverage volume and business outcomes6, such as
medical procedures, sales and prescription volume.
1 “Unpaid media” or “earned media” refers to news or editorial coverage that has notbeen purchased as advertising or advertorial.
2 A New Model for Media Content Analysis by David Michaelson of David Michaelson &Company and Toni L. Griffin of MetLife, available through the Institute for PublicRelations, www.instituteforpr.org .
3 Edward Bernays (1891 - 1995) as quoted in his 1928 book Propaganda. 4 Council of Public Relations Firms 2002 Industry Fact Sheet; www.prfirms.org .
5 The Dictionary for Public Relations Measurement and Research by Dr. Don Stacks ofthe University of Miami, available through the Institute for Public Relations,www.instituteforpr.org .
6 Outcomes are defined in The Dictionary for Public Relations Measurement and Researchas “quantifiable changes in awareness, knowledge, attitude, opinion and behavior levelsthat occur as the result of a PR program or campaign.” “Business Outcomes” would referto behavior changes such as an increase or decrease in sales, store traffic, productqueries, intent-to-purchase, etc.
Exploring the Link Between Volume of Media Coverage and Business Outcomes By Angela Jeffrey, David Michaelson & Don Stacks Copyright 2006, Institute for Public Relations This topic may appear to be very basic and fundamental, but it has not been addressedwith specific evidence to date.
The case studies presented in this paper are among almost 200 executed over the past
four years by a leading media research firm7. These studies involved an analysis of more
than 10 million news articles that compared the quality and quantity of client media
coverage with business outcomes8. While most of these studies focused upon share of
qualified coverage between competing organizations, the data yielded insights about the
subject of volume itself, and the variables that affect its impact.
MEDIA COVERAGE VOLUME METRICS
In order to better understand the issue of media coverage volume, it is essential to lookat the various ways it is measured today.
Clip Counting – This is the most common, and most basic, way to measure volume.
Articles are simply collected, sorted by subject or by date, and counted regardless of thesize or quality of the story itself. Because it offers no qualitative information, it can beseverely misleading as a measure of success. According to a 2003 PR News survey,“Attitudes toward Public Relations Measurement & Evaluation9, a full 84% of respondentscited clip counts as their main method of measuring.
Audience Impressions – This is the second most commonly used method today, andconsists of collecting print circulation figures, broadcast gross impressions and Internet“daily average visits,” and totaling them up as an estimate of audience reached. Variouspermutations exist, such as multiplying circulation figures by estimated pass-alongrates. New research shows that using audience impressions to gauge effectiveness maybe 11.6% more accurate than story counts10 when correlating to outcomes; however,impressions can also mislead, since the size/duration of the story, and quality of theplacement itself, is not considered. In the survey mentioned above, 51% of respondentsclaimed utilizing this method.
Media Value – Formerly referred to as “ad value equivalency,” “media value” refers tothe practice of multiplying the space or time occupied by a story by advertising costs,and totaling them as a measure of success. While there is no ‘equivalency’ between theimpact a story might have versus that of an ad, new research11 shows that theinformation embedded in market-driven advertising costs do improve correlationsbetween media coverage and business outcomes by 25.5% over clip counts, and 12.5%over impressions. However, while this method does capture the size/duration of a story, 7 PRtrak actually pioneered this research in the late nineties. It continues today underthe auspices of VMS, which purchased PRtrak in September of 2005.
8 See definition of “business outcomes” in Footnote #6.
9 “Attitudes toward Public Relations Measurement & Evaluation,” a survey sponsored byPRtrak and published by PR News; designed and executed by David Michaelson, Ph.D, ofDavid Michaelson & Company, LLC, and October, 2003. www.instituteforpr.org. 10 “Close Correlations” white paper summarizes research done to compare the accuracyof using clip counts, audience impressions and media value when correlating tooutcomes, from VMS, available at www.vmsinfo.com.
11 “Close Correlations,” as above.
Exploring the Link Between Volume of Media Coverage and Business Outcomes By Angela Jeffrey, David Michaelson & Don Stacks Copyright 2006, Institute for Public Relations and is a good measure of prominence12, it lacks other qualitative factors such as tone,message appropriateness, etc., so is not an adequate measure alone. Media value wasused by 45% of respondents in the PR News survey.
Qualified-Volume Measures – More progressive and accurate measures of mediacoverage volume include counting and comparing only those articles that have been“qualified” by factors such as: Contains messages placed within proper context While qualified-volume scores will always be lower in absolute value than non-qualified,they usually correlate much better to business outcomes than raw counts.
Media Analysis Indices – In an attempt to marry the best quantitative measures (clipcounts, impressions or media values) with the best qualitative measures (tone,prominence, message, etc.), more and more analysis firms are offering single-metricindices such as the VMS Media Prominence Index™, the Delahaye Impact Score and theCarma Favorability Rating System. While no metric is perfect, media indices do go along way toward quantifying coverage volume in qualitative terms that are meaningful,and certainly toward equalizing it for competitive analysis.
Share of Media Coverage – Finally, we must consider share of media coverage volumeas a metric growing in importance today. As we will explore in a subsequent paper,overall coverage volume may have less to do with business outcomes than relative shareof volume among competitors.
ISSUES TO BE EXPLORED
While this paper will not attempt to address every variable discussed above, it willexplore the importance of two of them: Tonality and Message, and how they modify theeffect of pure volume.
METHODOLOGY
The case studies that follow were executed through an advanced linguistics, artificialintelligence tool with human-like text analysis capabilities13 that enabled the automatedscoring of tone. Because of this technology, studies that would have normally beencost-prohibitive due to large clip volumes were done quickly and economically. The goal 12 “Advertising Value Equivalency,” an IPR Commission on PR Measurement andEvaluation white paper, by Bruce Jeffries-Fox; www.instituteforpr.org .
13 This analysis tool is part of VMS’ editorial measurement system; www.vmsinfo.com .
Exploring the Link Between Volume of Media Coverage and Business Outcomes By Angela Jeffrey, David Michaelson & Don Stacks Copyright 2006, Institute for Public Relations of these case studies, along with almost 200 others, was to determine how changes inpublic discussion (unpaid media) correlated to changes in business outcomes. Thesecases utilize sales, prescription volume and medical procedures as examples of desired,and undesired, business results.
Print, broadcast and Internet articles were imported into the Ai system from Lexis Nexis,Factiva and other electronic sources. Only the portions of each clip that were actually‘owned’ by a client organization were counted for credit, and scored qualitatively andquantitatively. The examples that follow utilize Clip Counts, Impressions, and the MediaProminence Index™, with Pearson Product Moment Coefficient (r)14 correlationcomparisons to results.
CASE STUDIES
Study One: Pure Volume
The graph below illustrates a study for a healthcare organization that had extensively
publicized the importance of mammograms, but didn’t know if the effort had achieved its
objective of driving patients to doctors’ offices for medical procedures. The campaign
included no paid media or promotional activity, so was completely dependent upon
media coverage.
The five-quarter study analyzed 47,000 articles, all of which carried clear messages ofthe importance of mammograms for the early detection of cancer. This was aparticularly simple study in that the message itself was essentially neutral and factual,and therefore resonated well with the media. Also, there were no competitors withdissenting messages. The volume measurement used in this study was simple ClipCounts.
As illustrated in the chart that follows, the correlation between editorial coverage for thiscampaign and mammograms was a very high r=.89. Thus, it appears that there is astrong relationship between media coverage and business results, since proceduresincreased as the volume of press went up, and decreased when it went down.
14 The Pearson Product Moment Coefficient is a measure of association that describes thedirection and strength of a linear relationship between two variables; it is usuallymeasured at the interval or ratio data level. Definition is from The Dictionary for PublicRelations Measurement and Research by Dr. Don Stacks of the University of Miami,available through the Institute for Public Relations, www.instituteforpr.org . PearsonCorrelations were preferred by the authors of these studies since the qualitative data insocial science research is inherently less precise.
Exploring the Link Between Volume of Media Coverage and Business Outcomes By Angela Jeffrey, David Michaelson & Don Stacks Copyright 2006, Institute for Public Relations Correlations between Editorial Coverage for Mammography
and Corresponding Medical Procedures (CPT Codes)
2,000 # of Articles
Patients CPT Codes 290,000
Study 2: A Closer Look at Tonality
What we’ve demonstrated above is relatively simplistic: that in some cases wheretonality and message are neutral and straightforward, a clear correlation exists betweeneditorial volume and outcomes. But such an ideal situation is rarely the case in the realworld.
This next study deals with the issue of negative news, which has an inverse correlationto desired outcomes. The old adage of, “there’s no such thing as bad publicity,” isclearly untrue. Bad publicity happens, and with it, undesired results.
On January 6th of 2006, The American College of Chest Physicians released a reportsaying that cough medicines were ineffective: "There is no clinical evidence that over-the-counter cough expectorants or suppressants relieve cough.” The articles went on tosay that allergy medications were effective, however, due to their drying effects. Thisnews story was immediately covered by all the major TV broadcasts (World NewsTonight, Today, Early Show, Good Morning America) and most of the daily newspapers.
Exploring the Link Between Volume of Media Coverage and Business Outcomes By Angela Jeffrey, David Michaelson & Don Stacks Copyright 2006, Institute for Public Relations Regarding other marketing efforts, there was no change in any marketing mix variableduring the short time period during which this news hit and sales dropped.
Negative News on Cough Syrups Coincides with Severe Drop
in Cough Segment Sales
The chart above illustrates the immediate impact of the surge of negative stories (blueline) on cough syrup unit sales (red line) in a very short time period. (Note: cough syrupsales had already started downward in December, which is standard during the HolidaySeason; but the rate at which they dropped after the news broke, and the degree of theloss, was unprecedented). It also shows a slight rebound in sales once the volume ofnegative news eased up. The measurement metric used in this study was Tonality-Qualified Clip Counts, and the number of stories was several hundred.
Interestingly, one client’s bad news may be another’s good news. Correspondingmeasures to this negative publicity included a 12% surge in physician visits forrespiratory illness for the week ending January 14th; and, allergy medicine sales soaredreflecting the positive news in that same story. Bottom line, whether the tone ispositive, negative or neutral, it appears volume links closely with business results –though the desirability of those results is another issue.
Study 3: Is Tonality Enough? What about Message?
Here is an award-winning study from Porter Novelli15 illustrating how media coveragethat simply mentions a brand, but does not deliver meaningful or accurate messagesabout that brand, may not have much impact. As with the other studies discussedabove, this product was promoted solely through unpaid media.
15 Study was entitled PRoof - Porter Novelli Key Message Assessment & Optimization”for Watson Pharma, Inc. on OXYTROL, with VMS’ editorial measurement system.
Exploring the Link Between Volume of Media Coverage and Business Outcomes By Angela Jeffrey, David Michaelson & Don Stacks Copyright 2006, Institute for Public Relations The product studied was a prescription medication for the treatment of overactivebladder, and the outcome measure was prescription volume. Some 3,000 articles wereanalyzed using a proprietary media index16 that accounted for tone. The overallcorrelation between the primarily neutral and positive coverage and prescriptions wasmodest (r=.51), suggesting that volume alone is not sufficient to produce desiredresults.
However, when coverage was limited to only that which contained at least one key
message, and was then correlated to prescriptions, the r value soared to .97. Of further
interest, only two of the six key messages appeared to be driving the results.
Incidentally, this Porter Novelli study won the 2006 Institute for Public Relations
Jack Felton Golden Ruler Award for excellence in PR measurement & evaluation.
News with Key Messages Coincide with Prescription Volume
Correlation of Oxytrol Key Message Discussion and TRx • The value of articles containing key messages correlates very well (r=.97) with Oxytol prescription volume.
• The chart below shows the net media value of Oxytrol media coverage containing one or more key messages versus Oxytrol prescription volume.
PN Key Message Optimization
These Porter Novelli findings dovetail with work done by David Michaelson, Ph.D, andToni Griffin in their paper, “A New Model for Media Content Analysis.”17 Michaelson andGriffin found that “between 60 percent and 85 percent of articles on the key issues ofconcern to MetLife included errors in reporting, a misrepresentation of key information 16 VMS’ Media Prominence Index utilizes Media Value refined by Tone and Prominence(of client in article).
17 “A New Model for Media Content Analysis,” by David Michaelson of David Michaelson &Company, LLC, and Toni L. Griffin of MetLife, 2005; available at www.instituteforpr.org .
Exploring the Link Between Volume of Media Coverage and Business Outcomes By Angela Jeffrey, David Michaelson & Don Stacks Copyright 2006, Institute for Public Relations or an omission of basic information that should have appeared in the context of thearticles.” Both studies illustrate the fallacy of considering a campaign a success if one gets lots ofoverall pick-up, and even “key-message” pick-up, without checking that the messagesdelivered were accurate and resonated well with the targeted audience.
INITIAL CONCLUSIONS
As stated earlier, this paper only skims the surface of exploring the relationship betweenmedia coverage volume and related business outcomes. It cannot lay claim to proof ofcausation in any of the cases above, since experimental research would be required tomake that leap.
But it appears there is a preponderance of evidence that suggests a strong relationshipbetween the two. Certainly Bruce Jeffries-Fox, former chair of the Institute for PublicRelations’ Commission on PR Measurement and Evaluation, found similar results in aseries of studies he did for AT&T several years ago18. At the conclusion of his work, hewrote, “These data suggest that exposure to prolonged and highly positive newscoverage can positively impact consumers’ attitudes toward and perceptions of acompany.” IMPLICATIONS AND DISCUSSION
So media coverage volume matters; and, depending upon your business goals, tone andmessage also matter. Implications for the practice of PR seem obvious: 1. Establish clear, measurable business goals and objectives for your campaign.
2. Utilize formative research to establish key messages that resound with your 3. Secure as much positive, on-message media coverage as is possible.
Implications for PR measurement and evaluation would include: 1. Measuring the volume of media coverage with Clip Counts may be sufficient for correlating relatively neutral, factual campaigns to business outcomes.
2. For more controversial, complex or commercial campaigns, volume can be better determined with more qualified measures that are sensitive to Tone andMessage, if correlations are to be seen.
A second paper will follow exploring coverage volume in terms of competitive share, tosee how relative volume plays into the mix.
18 “Toward an Understanding of How News Coverage and Advertising Impact ConsumerPerceptions, Attitudes and Behavior;” by Bruce Jeffries-Fox of Jeffries-Fox Associates;www.instituteforpr.org .
Exploring the Link Between Volume of Media Coverage and Business Outcomes By Angela Jeffrey, David Michaelson & Don Stacks Copyright 2006, Institute for Public Relations

Source: http://www.prnewsonline.com/Assets/File/whitepapers/volume-of-media.pdf