8 Dimensions of PR Analysis

Hisham Elzubeir
7 min readApr 29, 2021

In both roles I have played, from an agency side and a client side, I found that PR media analysis is unfortunately usually applied at highly superficial levels. Most organizations/companies look at 2 metrics only; reach and value. To them, this sums up the performance of their PR activities or PR mentions. PR agencies bank on the value aspect, whether AVE (Advertising Value Equivalent), or PR Value/Editorial Value, usually multiples of AVE on grounds that editorial coverage has higher credibility than advertisements do.

I’ll first dwell on those 2 metrics further, before I go on to other often unexplored dimensions:

VALUE

Monetary values, though used all over are highly misleading, and that is the case for many reasons. First, while its based on ad rates, those change often due to seasons or discounts or otherwise, which makes assigning value to editorials based on fixed ad rate inaccurate. Secondly, for Editorial Value being multiples of AVE on the premise that it is more credible — the first question that comes to mind, what scientific method was used to decide that it is 3 times as valuable?

Digging even deeper into it, let us take an example. Assume Health Co. gets 3 full page editorials on page 3 of Medicine Magazine in January, February and March issues respectively. Same magazine, same page, same size. A simple AVE/Editorial Value might assign each of those articles a fixed $15,000.

Ok.. so the January issue is talking about an art initiative that is taking place around the streets and bridges in town. In the last paragraph it mentions Health Co. as a location marker to be near one of those art installations — a side mention. The February article mentions Health Co. in the headline with the subject being its receipt of a prestigious award. The article only talks about Health Co. with colored visuals — highly prominent and highly positive. The March issue also has Health Co. in the headline, but this time the subject is on its products causing health problems to its customers, and the article goes on to mention that the company is facing legal action with a possibility of getting shut down — highly negative. The question that follows - are all 3 articles worth $15,000? Clearly, not.

REACH

Now if we dig deeper into reach, we find that it is a far more reliable metric, but it still has some flaws and its also generally looked at superficially as well. Most publications are unaudited in terms of circulations, which is what reach or OTS (Opportunities To See) is based on for print. It is a ‘claimed’ number. Pageviews for websites are usually not updated or relied upon by 3rd party data that is seldom accurate.

When looking at the ‘reach’ aspect of performance, it is useful to consider your ‘target’ and link those 2 together. Admittedly, there is no accurate and bullet-proof way in doing so, but it still helps to understand the efficacy of reach. If you are targeting people within a certain geography, gender, age, etc, then what was the reach for that demographic? If its print, and you are targeting the business community, then having lots of coverage from sports and lifestyle & general interest publications might not be the most effective target demographic reach as opposed to business publications.

SIZE

This is an accurate and straight-forward metric. For print its measured in cc (columned centimeters), for magazines its measured in pages and portions of pages, for audio or video its measured in duration. It does get tricky for digital mentions though. Looking at a page layout and the specs is one to go about it by using grid layout systems.

I tend to associate size with placement, position, layout, etc. For print, half a page means little without knowing on which page, for a 30 second radio mention, it means little without knowing on which day and which program, etc.

Size and position go hand in hand.

PENETRATION

This is broken down to various types of categorial breakdown of the coverage, and from it, more can be learned about the target reach. Penetration can be the breakdown of the coverage by languages, or markets, media types, clipping types (PR, feature store, interview, letter to editor, photo caption, etc), and other classifications.

This is very useful. Instead of just knowing you had X amount of articles in this time period, its good to know where and in what format your company is being covered. Diversification is sometimes essential and you won’t know if you are achieving that without knowing your coverage penetration.

SENTIMENT

It is important to always keep monitoring, particularly for negative mentions, but it is also good to know the level of positive mentions received within the overall coverage. Sentiments or tonality is usually segmented into 4; positive/favorable, neutral, negative/unfavorable and balanced (contains both positive and negative in the same article/post). Understanding where you stand helps you better target to achieve greater positive sentiments in your coverage.

Better crisis management is reliant upon early detection of negative coverage. Often a crisis begins with a negative mention whether on social or PR that the company might not be aware of. This signals immediate action, and earlier action is prompted when there’s a negative coverage monitoring that is actively ongoing. It is also important to look at sentiment from a comparative standpoint. How does your company fare relative to its competitors. It might present opportunities to capitalize on your competitors’ downfalls.

PROMINENCE

When I used to work for Mediastow, we developed a formula to assign a prominence index value to each article/clipping/post as a substitute for AVE. It is one of the most important metrics, as it provides a far more accurate understanding of the value of coverage albeit not a monetary one. We looked at different attributes of each article/clipping/post, such as the location of the top mention m (headline, 1st paragraph or marginal), the number of mentions n (3+, 2, 1), the relevant visuals or lack of v (colored, b&w, none) and the level of clutter c (exclusive mention, 1 other competitor, 2+ competitors). Plugging those attributes into a formula gave us an index value that provided insight to how prominent was the mention of an entity.

For each of the attributes the scores were based on a max of 3 (3, 2, 1). The Formula is:

P = (m+n+v+c)/(M+N+V+C) => P = (m+n+v+c)/12.

For example, if Promina LLC was mentioned in the headline (3), there were 5 mentions of it (3), there was a b&w visual (2) and it was exclusively mentioned (3), then the value is:

P = (3+3+2+3)/12 = 0.92.

RESONANCE

This is yet another overlooked metric. While your PR department or agency is doing the work getting your coverage all over the place, not every mention is a direct result of their effort. Being a talked about company organically in the PR media space is something to aspire to and constantly analyze.

Resonance can be looked at from a number of perspectives. The first I’ll talk about is ‘Manifest Vs. Latent’. That’s the obvious one — breaking down the coverage by those that were the result of PR effort vs. organic mentions. It goes deeper by looking at which publications or authors provided for the organic or ‘Resonance’ mentions. It might be worthwhile reaching out to them to ensure continuity or to understand the reasons for such organic mentions to insight more elsewhere. It could also be a case of negative organic mentions, which might lead to a crisis management set of actions.

REPUTATION DRIVER

This is rarely used, and it really depends on the type of organization you are applying this to, but it adds a quite useful insight. I believe this is a combination of a number of other metrics mentioned previously. To understand what aids or damages your company’s reputation, you have to look at ‘Sentiment’, you have to consider ‘Prominence’, but you also have to look at the sources in terms of authors, accounts, publications, etc. As sources quality, credibility and popularity differ, they could have different impact on reputation of mentioned entities.

An even deeper study looks at ‘Language Patterns’. This is primarily used for political based organizations. Words and phrases can be subtle in how they convey a message and how its perceived. For example, addressing the same group of people, one author might call them rebels, another freedom fighters, a third protestors, etc. Understanding the language patterns is understanding perception gained by the audience and thus an important factor in understanding what drives your reputation and in which direction.

The Power of Graphs

While not one of the 8 dimensions, I wanted to touch on the power of graphs. Using the right visuals goes a long way to helping you obtain insights that might not scream at you from a spreadsheet or from using the wrong type of graph. Seeing relationships between multiple variables enables you to understand cause and effect relationships that could lead to actionable insights. In a way, using the right visual acts as another dimension to PR analysis.

I grew fond of an analytical software called InZight developed by the University of Aukland. The graph below illustrates one such visual I created when I used to work for Mediastow. It looks at the coverage of many brands under 3 categories; Beauty, Fashion and Jewelry. It then looks at the volume of coverage VOC, the magazine coverage sizes MCS and the opportunities to see OTS each category gets over a period of a year.

As you can see from a single visual, we can compare those 3 categories and clearly see that fashion by far receives the greatest sizes, followed by jewelry, and that while the beauty segment attains greater opportunities to see or reach, and greater volume of coverage, jewelry has far more coverage sizes. It also says that despite the jewelry segment gaining greater sizes, it still falls short in terms of reach when compared to beauty, which could mean that beauty has smaller articles but in far more high reaching publications, while jewelry gets bigger articles in less reaching publications. A single graph shows 3 variables across 3 different categories, and the insights can keep coming the more you look into it.

--

--