Category: Publisher Research Council (PRC)

Consumers are interacting with media content in more sophisticated ways, accessing content on multiple devices, and at any time. In light of this, a new reading currency was required to achieve an accurate measurement of reading behaviour across multiple platforms to enable the buying and selling of advertising.

“The outcome is PAMS (Publisher Audience Measurement Survey),” says Peter Langschmidt, consultant to the PRC (Publisher Research Council). “The most complete measurement of reader audiences, across all platforms, to inform effective advertising investment.”

The reading currency provides an innovative “Gold Standard” Reader Audience Measurement founded on global multi-country best practice, combined with 40 years of local trial and experience.

An important inclusion is a tighter new reading measure, in PAMS, known as Core readers. According to Langschmidt, this unique measurement developed by the PRC, is included to reduce the effects of memory decay inherent in the recency method, where the longer the issue period, the harder it is for respondents to remember whether they read the publication within the issue period. In other words, it is easy to remember a newspaper read yesterday, but it is much more difficult to remember if a monthly magazine was read four or five weeks ago.

To qualify as an “Average Issue” reader of a paper/print publication, a person has to have last read or paged through any copy of the publication within a period no longer than the issue period of that title. For example, to qualify as an average issue reader of a weekly print publication, a respondent has to have read or paged through an issue of that title within the past 7 days.

The Average Issue Readers (AIR) are arrived at using this “within issue period” recency method. Core Readers are calculated using a formula based approach called the FOR (Frequency Over Recency) methodology, which overlays frequency of reading (number of different issues normally read) on top of the traditional recency AIR. Many different approaches using time spent reading, origin of copy and other metrics were examined, but the ‘frequency of reading’ question provided the best solution as it reduces monthly magazine readers by more than it does dailies, where the recency method works better.

The table below shows the average reductions in PAMS 2017 between the recency method calculated AIR and CORE Readers.

The FOR method uses the number of issues frequency probability and multiplies these by the number of AIR readers at each frequency level as shown in the table below. For example, daily newspapers use a 5-issue frequency scale as there are five issues published from Monday to Friday.

THE FOR METHODOLOGY

 Frequency Group

“We’re confident that the new reading currency PAMS, and in particular the CORE readers measurement will provide an accurate reflection of reading behaviour across all platforms that the media industry can use in research and of course, the buying and selling of advertising,” concludes Langschmidt.

To find out more information regarding definitions and the questionnaire, please see:

http://www.prc.za.com/wp-content/uploads/2018/04/PAMS-2017-Technical-Report.pdf 

PAMS documentation and all bureau contacts are available on the PRC website at http://www.prc.za.com/pams/.

For additional information and more in-depth data of titles, please make use of the PRC’s website http://www.prc.za.com/ or contact the PRC on 011 326 4041.

After a period of intense data validation, The Publisher Research Council (PRC) is pleased to announce that the first ever Publisher Audience Measure Survey (PAMS) is live and available to all research and software bureaus. The data researches titles and brands.

PAMS is the new reading currency designed to achieve an accurate measurement of reading behaviour across multiple platforms to enable the buying and selling of advertising.

The research, conducted by Nielsen, measures around 150 newspapers, magazines and websites readers. This title specific media currency survey is based on first world methodologies and best of breed practices. With many research ‘firsts’ and innovative changes, PAMS is the first reading study worldwide to use ‘flooding’. Additionally, global AIR (Average Issue Readership) is included and the recency method is used. The PRC has introduced a tighter new reading measure, in PAMS known as *CORE readers.

“For the media industry, the beloved brands are back in PAMS!” says Peter Langschmidt, consultant to the PRC. “From motor vehicles to financial institutions to retail, including furniture and groceries, the survey has determined who’s responsible for category and brand purchases, where they buy, what they read and on what platform. With a random probability sample in excess of 17,000 respondents, results will generally fall within a few percentage points margin of error.

A word of caution though, the survey’s data (including the paper measure) is not comparable with any previous reader surveys.”

The PRC took the decision to include some key brands from the major adspend categories in this first ever PAMS. These include cellular networks, financial institutions, automotive, and retail – food and grocery, furniture and clothing.

One of the most interesting facts to emerge from the first PAMS is that Capitec has overtaken   the traditional “big four banking institutions” in terms of number of customers.  And they have not merely just surpassed the others, but at a 23% penetration vs. 13% for the second placed ABSA and FNB, they have blown away the competition.

33% of South African households have a motor vehicle according to the study. The PAMS data closely matches Stats SA 2016 Community Survey (33%). Top manufactures mirror NAAMSA and car park figures.

“We wanted to simplify the complexity of previous research questionnaires, and so within the food and grocery section, where consumers generally visit at least two retailers monthly, we asked only ONE key question – Where do you spend most of your money on groceries.”

Shoprite completely dominates the mass and lower sectors, while Pick n Pay is the leader, to a lesser extent, in the top end SEM 8-10 segment. “This is just the beginning of category and branded data, the next step is survey and panel fusion. Fusion is no longer the vision, but rather the preferred solution internationally,” concludes Langschmidt. “We are well advanced in discussions with Nielsen to fuse PAMS with their Homepanel which tracks actual purchases of over 2000 brands amongst 4000 representative households.

Within a few months we will have more brands linked to the readership of our titles across all platforms, while publishers will be fusing PAMS data with Effective Measure online data in the near future.”

*http://www.prc.za.com/wp-content/uploads/2018/04/PAMS-2017-Technical-Report.pdf

PAMS documentation and all bureau contacts are available on the PRC website at http://www.prc.za.com/pams/.

For additional information and more in-depth data of titles, please make use of the PRC’s website http://www.prc.za.com/ or contact the PRC on 011 326 4041.

This week, a new milestone in reading currency research was reached when the PRC (Publisher Research Council) released the first PAMS (Publisher Audience Measure Survey) results that measure reading across multiple platforms.

“The main objective for the PRC was to achieve an accurate measurement of reading behaviour across multiple platforms to enable the buying and selling of advertising,” says Peter Langschmidt, consultant to the PRC. “We feel that this objective, to a large extent, has been achieved and we’re thrilled to present the findings of the very first PAMS to the industry.

The purpose of the launch in Johannesburg and Cape Town was to provide a broad overview of the study with some salient results and findings. We started with a clean sheet, and even the definition of ‘reading’ was relooked, with the question adapted to include the phrase ‘reading for one minute or longer’ across all platforms and devices, namely paper, cellphone, desktop/laptop and tablet.”

The survey made use of global best practices. A brand first approach was adopted, with respondents self-selecting titles, then measuring paper and online formats, or inserts. New visual frequency scale and multiple pick-ups were amongst many innovative changes. PAMS is the first reading study worldwide to use ‘flooding’. This allows for multiple interviews in the same household, resulting in 17 386 interviews conducted in 10 000 households.

The sample for PAMS was designed by Dr Ariane Neethling and achieved an astounding national coverage of 223 of the 224 municipalities. A disproportionate sample was used to up-weight areas of economic importance and population groups, achieving a weighting efficiency of 83%.

Although the survey’s data (including the paper measure) is not compatible with AMPS (All Media and Products Survey), the results are reasonably consistent with AMPS and some comparisons between the two surveys have been made.

Paper is still the dominant platform for reading at 82%. The top ten ranked daily newspapers reflect similar titles and readership between PAMS and AMPS with Daily Sun, Sowetan and Isolezwe taking the top three spots. Weekly newspapers display similar matches with Soccer Laduma, Sunday Times and Sunday Sun as the top three, albeit that they have changed places. Magazines also show a similar pattern with nine of the top twelve titles being the same in both studies.

The PRC also included two placebo (non-existent) titles namely The World and Tshepo. These were included to model over-claiming. As it happens, only six respondents claimed to have read Tshepo and 56 out of 17 386 (around 1 in 300 respondents), The World (The World claim being confused with World News online).  “This incredibly low level of claimed reading of fake titles validates the PAMS methodology and fieldwork,” comments Langschmidt.

In any research, it is important that no glaring inaccuracies are uncovered, and PAMS validates well against other large sample research, namely Statistics South Africa’s, Community Survey 2016.

“Like software, we refer to this as PAMS 1.0. While we have made great strides in terms of reducing RPC (readers per copy) with newspapers, there is still work to do in the magazine environment. Much of this relates to specific vs. average issue measurement which will be addressed using our upcoming online panel,” concludes Langschmidt.

“The validity and credibility of the data released is of the utmost importance to us as the PRC and vital for the industry at large. To ensure this, in close cooperation with our research partner, Nielsen, we are in the final stages of final validation of all data.

The release of data to the agencies and bureaus will take place once Nielsen and our members have concluded these final checks and validations.

PAMS goes back into the field next month – March 2018.”

For additional information and more in-depth data of titles, please make use of the PRC’s website http://www.prc.za.com/ or contact the PRC on 011 326 4041.

In on-going, large sample studies over the past two years, London based research company Lumen has conclusively proved that the actual viewing of, as opposed to potential exposure to adverts (no matter the reading platform of digital or print) leads to almost 9 times more sales.

Lumen is an attention technology company that uses eye tracking cameras to measure the actual visual engagement readers have with various media types¹.  Just because consumers can see an ad, does not mean that they will look at it. Advertising typically gets looked at for very short periods of time – regardless of how much detail advertisers include.

“The technology is brilliant, and by using existing web cams in computers drastically reduces costs and eliminates the biggest problem in all research, that of claimed vs. actual behaviour,” says Peter Langschmidt, Research Consultant to the PRC (Publisher Research Council). The software passively monitors the respondent’s web browsing as well as eye movement on every page and plots a ‘heat map’ of whatever content and advertising is being viewed.”

While print adverts are more viewable and achieve longer dwell times than their online counterparts, this is not the entire story, as quality content plays an important role as well.

There are wide variations in how different media perform,  some sites and ad formats perform far better than others. The best sites do a much better job of delivering attention for advertisers than the average, with as much as 56% of viewable impressions being noticed. In fact, on average, ads on Newsbrand platforms are almost twice as likely to be seen than those on other websites, and they achieve longer dwell times.

Alignment between the advertising and the content also helps facilitate ad attention. This contextual relevance matters! Experimental studies have found that placing ads in a contextually relevant environment can boost attention dramatically. Testing shows just how much context matters,  as ads placed within the right context get 692 times more attention than those simply placed in any environment.

The fundamental to any advertising message working is first, actually seeing the advert and secondly, dwell time that allows for cognition to impart the message. The longer consumers look at ads, the more likely they are to remember the message and act upon it.

Anyone who has ever studied for an exam or test, knows that reading is the best way to learn, and that the longer one studies the more familiar one becomes with the course material.

Does this also apply to advertising and brand recall? Well there’s a simple way to find out. Just like exams we can simply ask respondents to do a prompted advertising recall test.

These results, gleaned from many studies over several years, conclusively prove the theory of dwell times precisely linking back to recall.

The final, and most important step that advertising must accomplish is to encourage purchase, for it is here that the advertiser receives their Return on Investment (ROI). To complete the purchase journey and prove this, Lumen installed their eye-tracking technology onto a sub sample of an existing FMCG consumer panel belonging to Nectar in the UK.

By linking the initial online advertising interaction of the panel with their subsequent offline shopping behaviour, we can see the dramatic effect that actual viewing of advertising has.

As can be seen, across these 10 Nectar brands, the average upliftment of sales during the campaign was 2%, but amongst those who actually looked at the advert this was 8.5 times more at 17%.

Media planning is traditionally based on possible exposure, an opportunity to see, but this metric is becoming more and more irrelevant, because if consumers don’t even view an advert, how can it possibly change behaviour? With this new measure we can track the difference in sales between those who actually looked at the advert and those who did not.

 The Lumen data shows that the best ad inventory facilitates communication with consumers when they are in the right mood to be receptive to advertising messages, and then creates the time and space to enable advertisers to get their messages across.

“This is dynamite,” concludes Langschmidt. “Until now we have been unable to measure actual viewed impressions.

To summarise, no matter the reading platform, to maximise advertising exposure, the advert must firstly be seen for a period of time. Different platforms perform differently, the media environment affects a reader’s attention and finally, context matters.”

At this year’s PDRF (Publishing & Data Research Forum) in Madrid, the PRC’s (Publisher Research Council) Peter Langschmidt took top honours for South Africa when his presentation, entitled “The Reading Revolution. React, Reposition, Recalibrate” was judged to be the symposium’s best presentation out of all those submitted from around the world.

 The Publishing and Data Research Forum (PDRF), held every two years, is devoted to best practice and innovation in measuring publisher paper and online audiences around the world, where experts in media and audience research globally share ideas and solutions. This year’s symposium, held in Madrid from 14 to 17 October, looked forward to publishing in the 2020’s and was themed New Opportunities: New Approaches.

“It’s an honour for my paper to have been selected amongst many outstanding global presentations,” says Peter Langschmidt, Research Consultant to the PRC. “There were 42 papers presented by 60 media and research experts from over 30 countries. The paper details the manner of the work completed by the PRC across reading media platforms and a summary of projects handled over the last 18 months in South Africa.

The Publisher Research Council is now truly on the global publishing and media map and validates the principles, methods and manner of our media research. Our world first innovation of RGS (Reverse Geographic Sampling) where we combine an online sample with a face to face one to ensure optimum spread and household type was particularly well received.“

Further to the winning the award, the IAB (Interactive Advertising Bureau) has requested that this paper be presented at their annual conference in March 2018.

On 3 October 2017, the two Peters, Peter Langschmidt and Peter Storrar from the Publisher Research Council (PRC) and Kantar TNS respectively will be presenting the Establishment Survey’s (ES) full year results, which for the first time includes household expenditure.

 The ES is a multi-purpose, multi-media survey providing context for all media platforms and all media currencies in South Africa. It is a collaborative partnership between the Publisher Research Council (PRC) and the Broadcast Research Council (BRC). The aim is for media owners, marketers, and advertisers to use the survey as a strategic inter-media planning tool.

The presentation will take place at Bryanston Country Club on Tuesday, 3 October 2017 at 15h00 until approximately 17h00. A Q&A session will take place immediately after the presentation to address all concerns and queries regarding the ES and the presented data. “The results, off a robust 25 000 sample, comprising of two waves of 12 500 each, includes household expenditure for the first time,” says Motlalepula Mmesi, researcher at the PRC. “We look forward to seeing all of our media, advertising and marketing colleagues there.

The two Peter’s will be available afterwards to discuss any questions personally and of course, everyone is invited for some informal networking.”

The first PAMS (Publisher Audience Measurement Survey), with individual published titles and major ad-spend brands included, will be released in February 2018, and subsequently fused with the ES.

Peter Langschmidt, consultant to the PRC (Publisher Research Council), investigates how print’s OTS (Opportunity to See) of around two per reader effectively halves the medium’s Cost per Thousand (CPT).

 Print is the only medium where rate cards are based on one OTS (Opportunity to See) per reader, however new research consistently proves that the same issue of publications have multiple reading occasions by the same reader. This results in additional opportunities to see an advert, which is usually around two. These multiple OTS effectively halve print’s Cost per Thousand (CPT) compared to broadcast and other media.

As seen in the table below, this thoroughness/multiple reads translates into advert OTS which increases amongst the older, more affluent, more influential consumer. TABLE 1. OPPORTUNITY TO SEE ADVERT BY SEGMENT

The research allows us to apply the OTS data to planning principles to get to a more accurate CPT for print versions of newspapers and magazines. By understanding this multiple OTS occurrence by segment and type of publication it can now be applied to media planning principles.

An advertising budget divided by each medium’s CPT will deliver that medium’s GRP’s (Gross Rating Points), a media weight measure (reach x frequency). According to IBIS media manager the CPT for newspapers is R117 and R103 for magazines, when we take these multiple OTS into account it effectively halves the CPT as shown in the table below:

TABLE 2. MULTIPLE OTS

REDUCE PRINT CPT

Let’s take a budget figure of say R100,000 and see how many ratings we could achieve with each medium based on its CPT. Outdoor, having the lowest CPT, delivers the most ratings (GRP’s) for the same budget.

We also have to take into account the inherent strengths and communication ability of each medium relative to one another. We summarise this in terms of a simple impact figure. All media types are rated relative to the most impactful medium – cinema, which has a captive front facing audience in a dark cinema with a 30m screen and Dolby stereo sound.

TABLE 3. FROM CPT TO

GRP’S TO ERP’S

Once this impact factor is applied we’re able to, more effectively, compare each medium’s Effective Rating Points or ERP’s. Television emerges as the most powerful medium while magazines and newspapers take the other two podium positions and deliver almost twice the effective audience numbers compared to radio and outdoor.

The dominance of print over radio is even more pronounced in the upper SEM/LSM 8-10 market and the perfect information foil to pin down and complement the transitory TV message.

The Media Comparison Calculator attempts to go beyond the numbers and PRC members are supplying the calculator to all clients and agencies so that they can apply it to any target market.

For additional information and research on reading, visit www.prc.za.com.

The ‘Holy Grail’ of media research will soon become reality in the form of Fusion, offering multiple hub and donor studies. The Publisher Research Council (PRC) and Nielsen recently held an industry workshop to discuss exactly how Fusion works and all its possible survey options. 

 The days of one dimensional AMPS numbers are long gone and in media planning and buying, the ideal has always been to know as much as possible about the consumer to enable more accurate and timeous branded messages. These include demographics, geography, mood, activities at time of impact, touch-points during the day, retail behaviour, brand repertoire, media consumption and much more. Much of this information is available, but it is scattered all over the place, contained within multiple studies, panels and sales data.  Fusion provides the key to consolidating all these insights into one easily accessible place.

Apart from assisting in media planning, by fusing with ongoing sales data from scanner panels these new systems can be used at the implementation planning stage and afterwards, when linked to panels and sales data, in the post campaign stage to measure results and determine how the campaign performed.

“This is not some futuristic pipe dream that only exists in first-world markets,” says Peter Langschmidt, consultant to the PRC. “We are talking to many database and survey owners, and we could be fusing with the Nielsen FMCG Homepanel as early as the second quarter of next year after the release of PAMS in February 2018.”

Data fusion is a statistical technique that creates a single data set from the attributes from two separate data sets. All that is required is “fusion hooks” or relevant common variables on both the hub and donor studies.

The industry has moved incredibly fast since the last AMPS in 2015. In September 2017 the first full year of the Establishment Survey will be released. And within a few months there will be a full set of currency donor surveys, that could be linked together to create an essentially “single source” fused data set.  These could include published titles on PAMS, Radio and TV on BRC RAM and TAMS, Online on EM (Effective Measure), and OOH (Out of Home) on ROAD.

PAMS will have branded data and with a 20 000 sample will make it an ideal hub survey for Nielsen’s FMCG Homepanel. “The incorporation of multiple demographic, industry and lifestyle variables is at the heart of Fusion,” says Motlalepula Mmesi, researcher at the PRC. “If you have the correct “hooks” in your data, even customer databases, you can fuse with our PAMS survey. The result is much deeper research insights into your brands.”

It brings the discussion back to the differences between Fusion and Single Source. All markets fragment as the world becomes more complex and there can be no denying that the world is becoming more complicated and fragmented on a daily basis. Inevitably single source data/research will no longer be viable as there are fewer respondents per survey question, and more questions are needed to understand the complexity of life and brands, which means longer questionnaires and poor quality of responses due to respondent refusal and fatigue.

“When combined with ongoing sales and panel data, it becomes obvious why data fusion is a preferred solution worldwide,” says Terry Murphy, Head of Nielsen Watch in SA. “It is now possible to get a far more granular view of consumers, interrogate cross platform behaviour, calculate incremental unduplicated reach and define target markets based on last month’s purchases.”

PAMS could be fused with Nielsen’s FMCG Homepanel. Homepanel measures actual brand scanned purchases (not survey claimed behaviour) across 200 FMCG categories and over 2 000 brands nationally. The data is longitudinal (updated every month) and is weighted to the same IHS population data as PAMS and the ES.

Best of all this data is already paid for by subscribers, so it could be made available to agencies and advertisers for media planning purposes at an affordable cost.

The fused data will provide answers to questions about cross platform content consumption; longitudinal actual product and brand consumption like which titles (digital and print) are best suited for my target? What’s most effective in driving unique reach for my plan? How can I evaluate the effectiveness of secondary targets beyond demographics, including segmentation?

“A hub or establishment survey with donor currencies and purchases is based on global best practice, and produces vastly more insightful and real data,” concludes Langschmidt. “Although we can fuse with any survey, provided that we have the relevant common variables in both, the process has to start somewhere and it’s great for our publishers, that we, together with Nielsen, are pioneering brand and media fusion in the advertising industry.”

For additional information and research on reading, visit www.prc.za.com.

With PAMS (Publisher Audience Measure Survey) in field this month, the PRC (Publisher Research Council) discusses the advantages of the survey’s inclusion of brands, large sample size, innovative flooding technique, multi-platform measurement and other highlights.

 The purpose of PAMS is to provide an innovative “Gold Standard” Reader Audience Measurement founded on global best-practice and local experience.

“The primary objective of the survey is to measure and track audiences across all reading platforms to enable the buying and selling of advertising. We (the publishing industry) felt that the high RPC (Readers Per Copy), so prevalent in AMPS, had to be addressed in order for us to get closer to the truth,” says Peter Langschmidt, consultant to the PRC. “Other elements in the brief to Nielson included the ability to fuse our currency survey to ES (Establishment Survey), IAB data and the PRC mobile reader panel.”

The face-to-face interviews in 10 000 households will be flooded to produce a sample of around 20 000. One of the biggest costs in any survey is actually getting to, for example, a household in Viljoenskroon, so why not take advantage of interviewing everyone in the house while the interviewer is there? Hence flooding which doubles the sample size, but not the cost.

To achieve a new and improved reader currency, newspaper prompts will be regionalised, so as not to waste time with publications unavailable in an area. For example, what’s the point of asking a Johannesburg based resident if they have read The Argus? Available publications are represented in both print and online formats, allowing respondents to simply select read titles.

Traditional readership methodology had the interviewer asking the respondent if they had read any of a long list of publications. In order to appear well read and informed, the respondent would reply in the affirmative to at least a few, normally unread, titles. This verbal barrage led to ‘respondent fatigue’ and the ‘halo effect’ of answering at least some titles positively contributed to over-claiming. The PRC has addressed this by handing the tablet, containing the mastheads of the titles, to the respondent allowing them to select titles in complete privacy.

“We believe that this, together with other changes we have implemented like reading or glancing at titles for longer than one minute, will go a long way in addressing the readers per copy issue,” comments Langschmidt.

Other improvements in the survey include ‘time spent reading’ for each publication to gauge  loyal and regular reader measure, as well as a new visual frequency scale based on actual publishing intervals, and not a standard 4 or 5 point scale.

One of Print’s biggest strengths is that a single issue is read on multiple occasions, unlike other channels, like radio for example, where there is only one opportunity to hear per spot. PAMS asked readers about the number of times they pick up a single issue. “We’re hoping that the PAMS results will mirror other studies we have conducted this year where the average OTS of two is achieved in a single issue advert. This effectively halves the CPT of newspapers and magazines,” says Langschmidt.

PAMS also measures if and how inserts are used by readers.

With the exclusion of brands from the Establishment Survey the PRC took the decision to include many brands from major adspend categories. These include cellular networks, financial institutions, automotive, and retail – food and grocery, furniture and clothing.   For detailed information regarding the brands and products, please contact the PRC or download the PAMS questionnaire from the website.

All the usual demographics (Language, Literacy, Gender, Age, Ethnic Group, Income, Education), geographics (Province, Community, Metros) will be available complete with LSM and SEM measurements. These together with other frequency and recency questions form the fusion variables, or ‘hooks’ into the ES, IAB and other data, including the Nielsen CPS household panel.

“Obviously the questionnaire is set, but the PRC looks forward to engaging with everyone regarding the questions asked in PAMS and looking at possible additions to be added to the next survey,” concludes Langschmidt. “We want to achieve the most complete measurement of reader audiences, across all platforms, to inform effective advertising investment.”

The survey fieldwork began on 11 July 2017 and the full PAMS data release and launch will happen in early February 2018.

For additional information and research on reading, visit www.prc.za.com.