Importance of question sequence in MR questionnaire design

One of the most important aspects in designing a market research (MR) questionnaire is the sequence of the questions. Each question that you ask poses a potential danger to sensitize or condition the respondent, and thereby bias the respondent in the subsequent questions. One example is that asking a question like ‘Have you heard of Brand X?’ itself raises conscious awareness of the respondent who may not be consciously aware of the brand. Another example is that if a respondent is asked to indicate which brand s/he buys and later if s/he is asked to rate the brands, then there is a danger that the respondent might try to be consistent with his or her earlier answers, and hence will give higher ratings to the brand the respondent buys.

At a high level, the general rules are:

1. Ask the most important questions first when the respondent is more active.

2. Ask those questions which are most sensitive to conditioning such as attitudes and preferences earlier.

3. Ask factual and historical information towards the end as respondent becomes less enthusiastic and fatigued.

In a well-designed questionnaire, the respondent should not know the brand of interest (the brand for which the research is being conducted) up to a desired stage, thus avoiding any respondent bias. If the respondent comes to know that the research is being done for Brand X, then the respondent may become biased towards the Brand X. So, any question that has a danger of revealing the brand of interest must be delayed until all the information that is prone to conditioning is retrieved.

The most popular way of designing a questionnaire is the funnel approach. ESOMAR defines the funnel approach as ‘A way of ordering questions in a questionnaire so that general questions are asked before specific questions. This ordering avoids the responses to specific questions biasing the answers to general questions.’

Typically, the questions regarding awareness of the brands in the marketplace must be asked first. In fact, if the awareness of brands is being measured, then awareness must be the first question which must be asked when the category is mentioned.

Purchase Intention (PI) is one of the most important measures and is very sensitive to conditioning, so it should be asked immediately after the awareness question or as soon as possible depending on the research objective. In controlled experiments, the purchase intention should be asked immediately after exposure to the treatment.

PI should be followed by the attribute ratings – which attributes (category) are considered important by the respondent?. Brand Evaluation on the attributes should be asked next. All the brands that featured in the earlier purchase intention question must be individually evaluated against the attributes. Moreover, it helps to quantify the Fishbein Model. This can be followed by questions on brand behaviour, category behaviour, psychographics, and demographics.

Question Sequencing is a huge research area and there is a lot of interesting research regarding the right position of a question, the right way to frame a question, and the right scale to be used. One example is that some experts say that the consumers tend to be biased towards the left side in a Likert Scale. Another example is that some experts say that demographics should come at the start, while others say that demographics should come at the end. Some people take the middle path by asking the key recruitment demographic questions early and then postpone the rest of the demographic questions until the end.

All the above mentioned factors together make questionnaire design a very interesting and a crucial work in quantitative market research. But due to very demanding timelines, practising market researchers may not always be able to devote enough time for the questionnaire design.

 

Advertisements

The Smartphone Boom in India

India has about 600+ million mobile phone users with about 800+ million subscriptions (SIM Cards base). About, 60% of these users are in Urban India. Now, imagine converting all those mobile users to smartphone users.

Thanks to its population, India is a huge market for smartphone manufacturers. In the coming 2 years, smartphone manufacturers look to cannibalize a large pie of the feature phone market in India, and supply trends like the narrowing price-gap between the feature phone and smartphone, and the entry of more handset makers are accelerating smartphone adoption in India. Smartphone has a very high aspirational value, and people will adopt it in the first opportunity. It is only time that smartphones will be everywhere in Urban India.

So, how is a smartphone defined, how many smartphone users are there in India, and what is the growth?

According to Nielsen’s report released in Sep’2013, there are 51 million smartphone users in Urban India. Nielsen’s definition of smartphone is ‘phones with operating systems that allow installation of applications’. According to IDC APEJ Quarterly Mobile Phone Tracker, companies have shipped 12.8 million units in Q3’2013 and 10 million units in Q2’2013. There is a 229 percent y-o-y growth compared to the Q3’2012 number of 3.8 million units.

Image

If we assume a quarter on quarter growth of 20%, the number of smartphone users in Urban India will reach 105 Mn by Sep’2014.

Deloitte’s technology, media, and telecommunications predictions revealed on 31 Jan, 2014 say that the number of smartphone users in Urban India will cross 104 million in 2014. 

City-wise breakup of the smartphone users in India

According to Internet and Mobile Association of India (IAMAI), Mumbai has the highest number of smartphone users followed by Delhi.

Image

Where does India stand in the global smartphone market?

According to a recent report by the market research firm Mediacells, there will be 1.05 billion smartphones shipped in 2014, and 70% of those smartphones will be bought by new users and 30% will be bought by existing users as replacements.

Mediacells estimates that India (Urban & Rural) will add 172 million smartphones in 2014 to have a total of 250 million smartphone users. Even if  India is going to add about half that number of users (90 million – Urban 50 + Rural 40), India is going to have 160-170 Mn smartphone users by the end of 2014. 170Mn is a huge number, and remember this is a semi-optimistic estimate. India will be the second largest smartphone market in the world, surpassing U.S. and behind only China.

References:

1. http://www.campaignindia.in/Article/371194,deloitte-tmt-predictions-smartphone-users-to-double-non-cricket-sports-to-gain-momentum.aspx

2. http://www.gizbot.com/tech-biz/mumbai-leads-with-highest-number-of-internet-users-in-india-news-014432.html

3. http://www.informationweek.in/informationweek/news-analysis/286525/explosive-smart-phone-growth-cannibalising-feature-phone-sales-india-idc-report

4. http://www.theguardian.com/technology/2014/jan/13/smartphone-explosion-2014-india-us-china-firefoxos-android

5. http://timesofindia.indiatimes.com/business/india-business/Teens-drive-Indian-smartphone-sales-study-finds/articleshow/22406572.cms

6. http://www.telecomtiger.com/PolicyNRegulation_fullstory.aspx?passfrom=PolicyNRegulation&storyid=17660&section=S174

Analysis of New Triers, Repeaters, and Lapsers of a Brand

At any period of time, the consumer base of a brand is comprised of two sets of buyers:  New Triers, and Repeat Purchasers.

Repeats, New Triers

The terms are self-explanatory. To put it simply, Repeat Purchasers are consumers (or households) who repeated the purchase of the brand, and New Triers are consumers (or households) who bought the brand now, but who didn’t buy earlier.

A little more detail

For example, lets take two annual periods 2007 and 2008. Repeat Purchasers of a brand X are those who bought the brand atleast once in 2007 and also who bought the brand atleast once in 2008. New Triers are those who didn’t buy the brand in 2007, but bought the brand atleast once in 2008. Lapsers are those who bought the brand atleast once in 2007, but didn’t buy the brand in 2008. So, it is evident that whenever we refer to the terms New Triers, Repeaters, and Lapsers, we should always have two periods for reference. These periods can be an year, a quarter, a month,  or a week. Similarly, the terms New Triers, Repeaters, and Lapsers can refer to the number of consumers or households depending on the industry. In Telecom or IT, typically it might refer to the number of consumers or users of your device or app, whereas in FMCG it might indicate the number of households that bought the brand. So, whether it refers to consumers or companies or households depends on the industry data, but the philosophy remains the same.

Various Segments of the New Triers of a Brand

So, continuing with the previous example, New Triers are those who didn’t buy the brand in 2007, but bought the brand atleast once in 2008. The important thing to notice is the criteria ‘atleast once‘, which means some number of new triers might have bought the brand multiple times in 2008 (say once in February, June and October of 2008). Don’t get confused with Repeater because the Repeater has bought the brand atleast once in both 2007 and 2008.

So,  a New Trier of a Brand X in 2008 comprises of all consumers (or households) that have:

– Not bought the brand in 2007, and bought the brand in Jan’2008 and never bought the brand again in 2008.
– Not bought the brand in 2007, and bought the brand in Jan’2008 and repeated the purchase in Jun’2008
– Not bought the brand in 2007, and bought the brand in Feb’08 and Aug’08 and Dec’08.
– Not bought the brand in 2007, and bought the brand only in Dec’08
– Not bought the brand in 2007, and bought in ………………

So, regarding New Triers of a brand,  the marketer is interested in finding out:

– How many New Triers have bought the brand in the year 2008?

– Out of the New Triers of 2008, how many consumers (or households) went on to repeat purchase my brand in the next 12 months? For example, if a New Trier purchased the brand in May 2008, then did he repeat purchase my brand in the next 12 months or in 2008. You can define the period as you wish. This shows us the effectiveness in understanding if the problem is in converting the new trials to repeat purchases or is the problem of the brand not getting enough trials? (Please note that these repeaters are different from the brand repeaters in 2008).

– How many First Time Ever Buyers?  If you observe carefully, the new triers in our example are consumers (or households) who didn’t buy in 2007, and bought atleast once in 2008. So, the consumer (or household) could’ve bought in 2006, but didn’t buy in 2007 and then bought in 2008. So, these type of consumers are also New Triers in 2008, but they are not buying the brand for the first time.

So, First Time Ever Buyers of Brand X in 2008 are those who didn’t buy the brand anytime before, but bought the brand X in 2008.

– Among the New Triers (consumers or households) that my brand got in 2008, how many of them are category entrants (consumers or households that were not using the category before, but entered the category with my brand), and how many of them are brand entrants (consumers or households that were using some other brand in the category, but not using your brand). This is especially important for SKUs that are launched to drive the category and brand recruitment.

– How many of the New Triers of my brand in 2008, were using some specific brand ABC before. For example, if a user was using a brand Cinthol in 2007, but now she bought the brand Dove of the same category in 2008.  So, this will help the marketer understand which are the brands that I am pulling consumers from?

– What is the Average Revenue Per User (ARPU) or the Average Volume Consumption of the New Trier? Am I recruiting the high category volume consumers? Do my New Triers increase the volume consumption along the line?

– Is my New Trier also buying some other brand? Is he buying both Cinthol and Dove ?

Similarly, there are a lot of things that can be done on the New Triers, Repeaters and Lapsers. So, one can slice and dice the data in anyway we want to look at and analyze for key insights. I will write down more details in another post.

Thank you.

Entertainment in Television Advertisements

As I mentioned in earlier blog posts, to communicate something to a recipient one has to command the recipient’s attention and then be relevant to the recipient.

Communication:  Command Attention (Clutter breaking) ->  Be Relevant

This holds true even for communication among two individuals or two groups of people or for television commercials (TVCs). For the rest of this blog we will discuss it in the context of TVCs.

Though the rules seem simple, commanding attention is itself a very daunting task in this fragmented and cluttered world of media. On top of it, the message is driven home only if you are relevant to your fragmented consumer segments. Currently, we shall focus on the first part of the problem – clutter breaking and commanding attention.

What have commercials been doing to break clutter?

Historically, entertainment has proved to be one of the most effective ways to command attention of people. Entertainment is a very pervasive element of television ads today. Research shows that creative entertainment increases the attention to view the entire ad, reduces the resistance to persuasion, and has positive effects on purchase intention.

Wikipedia defines entertainment as – “Entertainment is something that holds the attention and interest of an audience or gives pleasure and delight.”  Psychologists define entertainment as “attainment of gratification of senses”.

Though people have different personal preferences of entertainment, it has been observed that across cultures and time there are recognisable and familiar forms of entertainment such as story-telling, music, dance, drama, sex, sports, horror etc. So, most ads today have atleast one form of content used to entertain consumer such as humour, music, and creative stories, etc.

The answer to the question is – Commercials have been using entertainment as one of the effective ways to break clutter and maintain attention levels, increasing people’s interest to view the entire ad, and research shows that creative entertainment has positive effects on persuasion and purchase intentions.

If all is well, what is the problem about entertainment in commercials?

One observation that always intrigued and puzzled me is that the commercials that are very entertaining and enjoyable don’t always drive home the intended purpose. There are many commercials that are enjoyed a lot and has high ad recall, but they just become only a source of entertainment for the audience.

My observation of several ads and people made me come to the hypothesis that the entertainment provided in the ad actually fulfils the consumer and conflicts with the consumers’ process of synthesizing the brand/product message.  This negative influence of entertainment is especially seen when the brand purpose is not weaved into the story provided for entertainment. For example, in ads where the entertainment part comes first and the brand is shown very late in the ad and they are not so well connected. If entertainment is used to break clutter, then it is important that the brand is shown as a part of the entertainment at the beginning of the ad, else there is a risk that the TVC may be very entertaining but not serving the objective of the ad.

Harvard professor Thales Teixeira has conducted interesting research on this regard and wrote a paper – “Why, When, and How much to entertain consumers in advertisements?” This is based on a facial tracking study (software used to track the facial emotions) in response to the TVCs. This is a first of its kind study and is the latest (dated January 2013).

One of the key hypotheses for the study is – Does high entertainment in advertisements have detrimental effects on persuasion and purchase intent, while having beneficial effects on a person’s willingness to watch the ad?

Key Results from the Study:

1. Entertainment can overcrowd your product message.

2. Viewers tend to pay less attention to the message associated with the brand once they’re already entertained.

3. If entertainment is not brand-associated (brand comes first and then the entertainment part starts or both at once), then it works only as an attention capturing device.

4. An excessive amount of entertainment is ineffective because it reduces the ad’s persuasiveness, as the entertainment conflicts with the persuasiveness.

5. Medium level of positive entertainment leads to a higher intent to purchase the advertised brand than low or high levels.

Entertainment plays both a co-operating and a conflicting role

Prof. Teixeira found that entertainment plays both a co-operating and a conflicting role, depending on its type (i.e., location in the ad). Entertainment that is associated with the brand is co-operating, as it acts as a persuasion device both in the interest and purchase stages. Entertainment that is not associated with the brand acts predominantly as an attraction device at the interest stage, thus indirectly cooperating but also directly conflicting with the ultimate goal of the ad.

The paper talks about the role of the location of entertainment and brand in the ad and its effects on the purchase funnel. If the ad is solely intended to induce purchase from previously aware or interested consumers, early placement of the brand is recommended. This might be the case for established brands or mature products. Yet, if the purpose of the ad is to generate awareness and interest, for example for new brands or products, and other marketing tools will be used to trigger purchase, then placing the brand later in the ad will be more effective to increase its attractiveness. Lastly, for ads intended to increase interest and purchase, ad persuasiveness and attractiveness should be balanced.

The study shows that entertainment, while increasing interest, can hurt purchase intent, especially if it appears before the brand, and can help purchase intent, when it occurs after the brand. So having the brand appear later may work if the objective is more towards building awareness. But still I am not a strong supporter of entertainment coming first and then brand later. If you want to be safe, make sure that the brand has an appearance somewhere in the beginning of the ad (especially when entertainment is used for clutter-breakthrough).

Thank you.

Similar Posts on this blog:

https://brandalyzer.wordpress.com/2012/09/20/centuryplys-latest-gorilla-tvc/

https://brandalyzer.wordpress.com/2012/01/15/television-advertisements-and-relevance/

https://brandalyzer.wordpress.com/2011/09/03/flipkart-attacks-the-online-fears-with-its-new-set-of-tvcs/

References for this post:

https://brandalyzer.wordpress.com/2012/01/15/television-advertisements-and-relevance/

http://www.hbs.edu/faculty/Pages/event.aspx?num=387

http://www.affectiva.com/assets/Entertainment-effects-on-Purchase-Funnel.pdf

Choice based Conjoint (CBC) and Brand Price TradeOff (BPTO)

Choice based Conjoint

Choice based Conjoint (CBC) is a research technique based on the observation that consumers always choose products among a set of products in the marketplace, and a simulation of it is the closest to the real consumer behaviour. CBC is a technique wherein the respondent is shown a set of concepts (with specifications) and is asked for his/her preferences. This technique hopes to simulate the tradeoffs that consumers make in their daily buying experiences; the tradeoffs could be among the attributes of the product or among the products and brands listed. This technique is generally used to understand the interaction among the attributes, and for pricing studies.

One needs to list down the attributes and the levels for each of the attribute. For example, to conduct a CBC to understand the importance of the features of a smartphone; an example of an attribute could be “RAM Size” and the levels could be 512MB, 1GB, 2GB or whatever options you would like to present to your consumers. The options should be as close to the actual product as possible and the attributes and the levels should be given an extra-ordinary amount of thought. CBC should ideally be done on a sample of around 300-600 respondents who are aware of the products and the category.

One of the issues I faced while deciding on the attributes and the levels is that it is a little on the easier side for a very functional product like a smartphone or a car, where you can easily distinguish between different engines or processors, (different features like power steering, windows, etc…). The features and levels in functional products are easily distinguishable and conceivable. On the other hand, for products such as biscuits, toothpastes, sanitary napkins, etc. I am not sure how well people can distinguish and conceive different product benefits in such categories where you know the product only by experiencing it.

History of CBC

Limitations of CBC

  • Not all brands are  equally known to the consumers, and there is a risk of popular brands mostly being preferred in a CBC study.
  • CBC doesn’t take promotions and distribution into consideration, and it assumes that all brands are available and have enough media spends.
  • It assumes that the consumer has the ability to buy the product.
  • The number of questions involving different choice sets could easily increase, causing respondent fatigue.

Brand Price Trade Off

BPTO is a simpler version of a conjoint analysis where a set of brand/price combinations are shown to the respondent. As the respondent choses a particular brand, the price of that particular brand is increased and the consumer is again asked to choose among the new set of brand/price combinations. This technique helps us understand how the consumer trades off the brand and price, and what is the best  price point or price band for your product.

The one biggest advantage of this method is its simplicity, while it has quite a few critics in the market. One of the disadvantages of BPTO is that consumers may become conscious and may start playing around with the lowest price, or consumers may be protective of their brand and may always prefer a brand and take it to unrealistic pricing levels.

Technorati Claim Token: 9GYQZA35NG7P

Significance Testing

Significance Testing lies at the heart of all the inferences that we do from a sampling exercise. We always start with a ‘Null Hypothesis’ in the jargon. A test of significance is a test of that hypothesis. We analyse the data from the sample and try to estimate what would be the probability of getting that data if the hypothesis were true in the universe.

For the below reading, it is first important to understand the difference between accuracy and precision.

ACCURACY

Accuracy is the proximity to truth. If we knew the truth we would totally not estimate it altogether. So, accuracy of an estimate is a totally useless concept altogether for population statistic estimation.

PRECISION

Suppose you have the task of adding up long list of numbers – perhaps your daily expenditures over a month. You do your sum and get a particular result. But you’re not sure whether you got it right. You may have made a mistake in adding or punching in the numbers if you were using a calculator.

What do you do? You do the sum again. And if you’re a cautious accountant you might even do it a third time. If you get the same result every time you feel you have got it right.

Lesson: When in doubt, repeat. Repeatability of the result generates confidence in it. Repeatability is reliability.

Actually, our example of adding up a list of numbers is not a good one. Because, in this case there is only one true answer and we shall get it every time we do our sum correctly. But, the real life situations that we are interested in are the results that we get from measuring a sample of people from some universe. Again, we are not sure if the results are true. So, in line with our commonsense philosophy, we should be repeating the sampling exercise. If we did, it is highly unlikely that we would get exactly the same result, because different people would be included this time. In fact, if we repeated the sampling exercise many times and measured the same thing on different samples of people, we would find that most of the results fall within a range.

We would be entitled to come to a conclusion that, most probably, the truth that we are trying to estimate must lie somewhere in that range.

If we had a method of being more precise and if we could say, for example, that after repeating the sampling exercise many times, 95 percent of the results would fall within a certain range, then there would be a 95 percent chance that the truth would lie in that range.

The width of this range is a measure of the precision of our estimate – narrower the range, higher the precision. Our objective is to narrow this range as much as possible, because that would bring us closer to the elusive truth. Precision replaces the concept of accuracy. We will never be able to say how accurate is our estimate of the truth, but we can say how precise it is.

But how do we get a fix on this range? Taking just one sample in real life is problematic and costly enough. Repeating the exercise many times may be conceptually brilliant, but completely undoable in practice.

Actually, you don’t have to repeat the sampling exercise. This is where the science of inferential statistics comes in. By analysing the data in one sample that you have taken, specifically the variation contained in it, and by making some assumptions about the pattern of variation in the total universe, it can calculate the 95 percent or 99 percent or any other precision range that would actually come to pass if you did take the repeated samples. The whole purpose of inferential statistics is to save you the trouble of actually repeating the sampling exercise by inferring what would happen if you did.

It sounds like magic, but it is only logic. This logic completely depends on a crucial aspect of reality, namely the ‘Laws of Chance’, more commonly known as ‘Probability’.

So, the whole stuff is all about how precise are we in our estimate of a population statistic. After all, we all know the statistics of the sample. The problem is to understand the average height of the population in India, if you have a sample whose average height is known. This is where it all starts, and this is the role of the Central Limit Theorem (CLT).  CLT assumes the population to have a normal distribution, else the ‘n’ value has to be a minimum of 30.

CLT says that if you have a sample mean (x-bar) and the standard deviation of the sample is σ, then the probability that the population mean(µ) lies between the confidence intervals for a desired confidence level (z) (read it as a confidence level for now, I will come back to it later)

 

 

 

which is nothing but

 

 

 

For now, understand that CLT will provide with a confidence limit for the population statistic if you know the sample statistic and the standard deviation. Understanding the nuances of how CLT works and what its details are decently complicated and I will come back to it later.

Let us take a practical requirement for our understanding. Take for example, we have done a product test among men and women in a population and we asked the purchase intention of a product. Let us say, the results look as follows: (the numbers quoted are just for understanding the concept and may not hold the law of statistics)

Since we want to examine the differences in scores between men and women, we formulate the ‘null’ hypothesis that ‘there are no differences in the real scores in the population among men and women’ implying that the differences in the scores in the sample have come about by chance, and if we had repeated the sampling exercise, the differences would have disappeared.

The first thing to do is to calculate the confidence belts for both the scores by analysing the ‘variance’(using CLT) in the sample scores among men and women. Various situations can arise as follows:

Situation 1

Sample Size: 100 each

95% of range of scores of men is: 4.5———-4———–3.5

95% of range of scores of women is:                                      3.4———–3————2.5

There is only a small chance that men’s scores will be lower than 3.5 and women’s scores higher than 3.4. Therefore, the statement that ‘Men score higher than women’ has only a 5% chance of being wrong. Scores are significantly different at 5% level.

So if the 95% confidence belts don’t overlap much, then we can say that the scores are significantly different and cannot come by chance. So, we reject the null hypothesis. Here the degree of risk in rejecting the hypothesis is 5%.

Situation 2

Sample Size: 100 each

95% of range of scores of men is: 6———-4———–2

95% of range of scores of women is:    5———–3————1

No evidence for believing men score higher.

Scores are not significantly different at 5% level. We therefore don’t reject the null hypothesis.

Scores are not significantly different at 5% level. We therefore don’t reject the null hypothesis.

We can make this case to be significantly different by taking decreasing the confidence level or increasing the sample size as follows:

90% range of men: 4.6———–4———–3.6

90% range of women: 3.5———-3———–2.5

Scores significantly different at the increased risk level of 10%

We can also increase the sample size

Situation

Sample Size: 200 (As we increase the sample size, the range for confidence level will decrease which may lead to significant difference even when the confidence level)

95% range of scores of men: 4.2———-4————3.8

95% range of scores of women: 3.3———–3————2.7

Scores significantly different at 5% level

Therefore, increasing sample size will make smaller differences significant.

Interpretation of Significant Results

The fact that a survey result is found to be significant, by carrying out a statistical significance test, often leads to confusion when such a result is presented to people unfamiliar with recent methodology. The layman, when told that something is significant, often assumes that the researcher considers the result to be “important”.  Always remember when the researcher says significant he means that the result is statistically significant. In statistical terms, if, for example, a difference between two percentages is declared significant, it simply means that this difference, no matter whether it is a large or small difference, cannot have occurred by chance.

References:

http://dsearls.org/courses/M120Concepts/ClassNotes/Statistics/530_conf_int_mean.htm

http://dsearls.org/courses/M120Concepts/ClassNotes/Statistics/530G_Derivation.htm

http://lssacademy.com/2007/07/16/explaining-the-central-limit-theorem/

http://www.southalabama.edu/coe/bset/johnson/lectures/lec16.htm

http://www.socialresearchmethods.net/kb/stat_t.php

AdStock GRPs

AdStock is a simple mathematical model of how advertising builds and decays. It is invented by Simon Broadbent as he studied Milward Brown’s ad awareness data.

AdStock helps to:

  1. Optimize your advertisement scheduling
  2. Used in marketing-mix modelling to come up with advertising ROIs, etc.
  3. Helps you decide when to be off-air and when to be on-air
  4. Helps you understand the advertising decay behaviour

How advertising builds and decays?

Let us take awareness as a parameter to understand the concept of AdStock.  As a consumer watches an advertisement for the first time, let us assume that consumer gains certain awareness of the brand, category, etc. Now, when the same consumer watches the advertisement for the second time, the advertisement builds on the awareness. The advertisement hopefully will strengthen the awareness, recall, preferences, etc. So, advertising builds on itself and that is why we call it as a campaign building.

Similar to the way it builds, an advertisement also decays in similar fashion. If a consumer has seen an advertisement A1 10 times in a week and the same consumer has seen an advertisement A2 only once in a week, then the way the consumer forgets the advertisements is very different.  The decay rate of an advertisement depends on various parameters such as: the strength of the advertisement itself, media plan, media vehicles chosen, category involvement of the consumer, etc.

The normal GRP data doesn’t take into account the build and decay rates. So it doesn’t take into account the residual effect of advertising, though a company doesn’t advertise in a specific period. AdStock is nothing but the GRP data taking into account of the build and decay of advertising, which is more sensible in marketing applications.

Optimize your advertisement scheduling

As explained, the AdStock GRPs are the GRPs weighted for the advertising build and decay rates.

Let us look at case to optimize the scheduling strategy for an advertisement. For this case, the advertisement is assumed to have a half-life of 6 weeks (hypothetical). This will come out for a decay rate of 12.24% as shown in the table below.

We have four options of scheduling, each using roughly the same (1200-1500 GRPs) amount of GRPs. Once we translate these raw GRPs into AdStock GRPs, it will help us decide which scheduling strategy is the most optimum as explained below.

The AdStock GRPs are adjusted based on the decay rate.  For example, the number 469 in Wk 2 is arrived by: (250 of Wk2) plus (250*87.8) (decayed GRPs of Week 1) = 469.

Similarly, 662 in Week 3 is arrived by: (250 of Wk 3) + (250*87.8) (decay of Wk 2) +(250*77.0)(decay of Wk 1)= 469

From the above, it is clear that Option 1 gives the maximum ROI. The other parameter important for selection of an option is the off-air time. Which of the above options gives me the maximum off-air time (when you don’t air the advertisement)?

From the above table, it is clear that Option 1 gives the maximum off-air time for the advertisement by still maintaining more than 500 GRPs. In the above example, 500 GRPs is considered as the threshold and if it goes below, then the advertisement has to come on-air.

To sum it up, AdStock helps marketers understand ‘When to advertise‘? AdStock is commonly used in scheduling, marketing-mix modelling, etc.

Any comments on this regard are most welcome.

Thank you.