The Evolution Of AMPS

Over the past few months a lot has been said about AMPS and what it should be, mainly by people who have very little to do with SAARF, and who seem to think that the date is 1974.

Well I have news for them – the world has changed and 2006 is not 1974!

It is important to realize that one of the strong points of AMPS is the intense industry involvement through the years which resulted in an ever-changing product that has kept up with the fast changing South African landscape.

Right from the start of AMPS in 1975 the study, as its name reflects, covered all traditional media as well as products and services. This concept of an All Media and Products Study (AMPS) has proven to have been one of the best decisions ever taken by our industry.

Over the years the study has changed and grown from a fairly modest offering to an extremely comprehensive and sophisticated product. In doing this, the simplistic model of 1974 had to change because the world that we live in and the markets that we operate in have changed irrevocably. Thankfully the leaders of our industry saw this and have supported the necessary changes in our AMPS study to meet the needs of the modern fragmented and complex market of 2006.

As far back as the eighties, SAARF and media researchers in many countries realized that large changes were going to be required to keep the industry currency surveys relevant in a world of increasing media fragmentation and where the power of demographics was waning. It was clear to many users that the use of demographics alone in targeting was just not adequate anymore and that new thinking was required.

In South Africa this lead, inter alia, to the creation of the SAARF Living Standards Measure (LSM) which gave a new tool to marketers, media owners and advertising agencies to use in the segmentation of their markets. However, although a great step forward, this was not the alpha and omega of what was required by users.

At the same time SAARF was faced with another dilemma namely that the length of the AMPS questionnaire was becoming problematic and something had to be done to solve this problem.

This was the situation in 1998 when SAARF started considering alternative methodologies to enhance and compliment the trusted face-to-face methodology that is still today regarded as one of the most suitable methodologies for this kind of research.

The core of the problem was that the measurement of products and brands on AMPS have been neglected due to the time constraints imposed by an already long questionnaire and therefore a new way of gathering product and brand information had to be developed. This was crucial in order to ensure that users of AMPS would have access to more product categories and brands to enhance their capacity to do target marketing in a market that was getting more complex by the day.

During 1999, SAARF decided to investigate the possibility of using a self-completion leave behind questionnaire for the measurement of products and brands. This would reduce the length of the AMPS interview by 12 minutes and would enable SAARF to add more product categories to the survey.

It was also decided to develop a questionnaire that was more suitable for South African conditions than those used in other countries and the first pilot study took place later during 1999. The concept was piloted both at a category and brand level and was found to be very successful. Indeed it was found that the addition of the branded information using SAARF’s unique “brand logo approach” actually improved the quality of the data.

It was decided as a first step to implement the unbranded questionnaire in AMPS 2001A and that this would take the number of product categories from about 60 to 155 in one quantum jump. The first fully branded questionnaire was planned for and implemented in AMPS 2002B thus providing the first major source of branded information for use by the industry in the history of South Africa.

At the same time two other major changes were effected in AMPS:

Firstly, SAARF introduced Computer Assisted Personal Interviewing (CAPI) to replace the old pen and paper methodology. At that stage, South Africa was only the third country in the world to have taken this important step with relation to its national currency survey.

Secondly, SAARF started to report the AMPS results on a 12-month rolling database. This change was a necessity for providing the industry with larger samples without huge increases in cost and to stabilize the data.

The period 2001 – 2002 was also the time of the big race debate which led to the SAARF Hearings on Race on 27 March 2002. At these hearings it became clear that SAARF would have to spend more time on the development of segmentation tools. In addition, the AAA (now ACA) did a key presentation which again impressed upon us the importance of branded information and gave an even bigger impetus to the move towards the inclusion of many more brands on AMPS than in the past.

Dr Nina de Klerk and Ian Shepherd spoke on behalf of the AAA and said:

“In advertising planning (both media and message), truly useful segmentation should in any event be derived, bottom up, from brand usage and should not be conveniently replaced by general, predefined segmentation techniques (whether race, age, gender, occupation, life styles, home language, etc). Whilst these generalised descriptions of users and non-users of product categories are valuable in media planning, strategic considerations must be founded on brand usage and behaviour”.

Although they were not thinking of AMPS specifically in this regard, it just confirmed what we already new and encouraged SAARF to redouble its efforts to ensure that this important change would be implemented as originally planned during 2002.

At this stage it is important to ensure that everyone is very clear in their minds about what Branded AMPS really is. Branded AMPS is a term that was introduced to use for marketing purposes when the SAARF Board decided to allow SAARF to market the product to users that were prepared to pay for it.

It must be stressed that the self-completion questionnaire (sometimes referred to as Branded AMPS) is an integral part of AMPS. Without it, virtually all product data would disappear from AMPS and the single source nature of AMPS would be destroyed. It is thus not a new or separate survey but an essential change that had to be introduced otherwise AMPS would have been in a dead-end.

It has also proven to be the most cost effective way of leveraging the industry’s investment in AMPS as with a very small additional investment in AMPS, nearly 2 000 brands as well as numerous activities and interests have been added for use by all industry players.

For target marketing it is essential that the currency survey must contain extensive information on the characteristics of the users of mass media, their media consumption as well as information on their usage/purchasing of products, brands and services. This is the minimum required by users to do segmentation or audience profiling in the markets of today.

During the last AMPS Tender further important decisions were made by the SAARF Tender Committee. Firstly the Tender Committee felt that it was in the long-term interests of the industry to award the tender to two research houses so that the pool of skills for media audience and products/brands research could be expanded.

Although there is a price to be paid for this, the Tender Committee felt that the creation of skills inside other research suppliers would eventually lead to more competition for SAARF contracts thus leading to keener pricing in future and that this must thus be seen as an investment in the future.

The second important decision that was taken was that one of the proposed two contractors must be a BEE company, to assist the transformation process which as everyone knows, has not really happened in the research industry to date (especially as regards quantitative research).

It is clear that the financial implications of these decisions were not fully and clearly communicated to all stakeholders resulting in a funding debate early in 2006.

These issues have now been resolved, but still needs to be debated again by all stakeholders to ensure that a consensus is reached before the start of the next AMPS Tender Process in 2007.

Although this was not the reason for these decisions, the two contractor approach has had some additional benefits for SAARF as it made comparisons between the work of the two contractors possible for the first time.

However, it has also lead to an increase in the work load of the staff at SAARF and an increase in the size of the technical component of SAARF will have to be considered in the near future.

The 2003 investigation into SAARF resulted in a thorough overhaul of the Memorandum of Association as well as the Articles of Association of SAARF to ensure that the main business and the main object of SAARF were in line with modern demands.

The new Memorandum and Articles were adopted by the Board during 2004.

During the period 1998 to 2006 big changes also took place as far as the currencies of Television, Radio and (shortly) Outdoor are concerned.

Radio moved to a new methodology called flooding which enabled SAARF to increase the size of the sample to approximately 47 000 diaries a year. The frequency of publication of the data also increased to 6 times a year.

As far as Television is concerned, numerous technical changes were made to improve the survey leading up to the biggest single change in the survey since its start in 1989, namely the change to overnight ratings which was implemented in the middle of 2006. The full implications of this change will only become clear during 2007 once the industry has fully adjusted to this new methodology.

Over the past few years SAARF in conjunction with Nielsen Outdoor in the USA have been working on a new electronic measurement for Outdoor to bring the measurement of this medium in line with those of other traditional media. The implementation of the new methodology started with AMPS 2006 and the first results will be released to the industry in about March 2007.

As far as we know, SAARF is the first Joint Industry Committee in the world to have taken this step.

It is clear that we have entered an era of change as new technologies will increasingly be employed to meet the needs of users. The electronic measurement of Radio for example is just a question of time as competing technologies sort themselves out and costs, which are currently problematic, come down from the stratosphere.

These changes will have implications for the whole of the Media, Marketing and Advertising industry and the future challenges with reference to AMPS must be evaluated in this light.

It is clear in the light of the above, that the introduction of the previously mentioned improvements to AMPS does not mean that AMPS is now complete or above further improvement.

The industry investigation in 2003 that lead to a reduction in the sample size (virtually to 1975 levels!) and a reduction in the frequency of publication of AMPS to one a year resulted in great unhappiness amongst users of this data. It is of utmost importance that these issues be addressed by SAARF stakeholders as soon as possible.

In addition, all indications are that industry currencies in other countries are adapting to changing markets by introducing more and more information in order to provide their users with the necessary information to be able to function effectively in a complex and more fragmented market.

Industry currencies, to a greater or smaller extent depending on the country, now contain combinations of the following information:

1. Media audience figures

2. Demographics

3. Product information

4. Brand information

5. Information on services such as, for example, Insurance, financial, retail etc.

6. Attitudinal information

7. Activities and lifestyle information

8. Life stages

9. Socioeconomic indexes such as the SAARF SU-LSM

10. Interests, attitudes and opinions

11. Values

12. Psychographics

13. Wide ranging qualitative information on media and products.

The basis of AMPS which by nature and design is a behavioural study is however uncontested as it has been proven over and over again that the best predictor of future behaviour is current and past behaviour. It is thus not a question of fundamentally changing the way that we conduct AMPS, but rather how we are going to handle the increased demand for information and accountability.

The above developments must not be regarded as nice to have as it is seen by many leading researchers as a necessary change to meet the challenges of the future. Luckily SAARF and AMPS are in a position where these challenges can and will be met to the extent that our stakeholders require.