Tuesday, November 08, 2005

Are pay TV channels like specialist magazines ?

Last year B&T magazine (20/9/04) asked if Australian pay TV channels are like specialist magazines.

And the representatives of these channels replied YES. This is not surprising because they typically pitch pay TV to Australian advertisers as delivering both targeted and highly involved audiences. The argument being that specialist channels attract specialist audiences. And because viewers have chosen to pay for the channel, so the argument goes, they must be interested in the content and they must therefore 'watch more intently/closely'.

While this sounds plausible, it's actually a load of nonsense. Australian pay TV doesn't deliver different nor more involved TV audiences, it simply delivers smaller TV audiences.

The analogy with specialist magazines is inappropriate because TV channels are typically sold in bundles of very many different channels. So most subscribers of a particular pay channel are not people who deliberately selected to buy this particular channel - it just happened to come with their bundle. Whereas, in the case of specialist magazines, a person who buys, say, Decanter magazine undoubtedly has a serious interest in fine wine, so this magazine does deliver its advertisers a targeted (but also very small) audience. In contrast, viewers of Foxtel's FashionTV are not fashion enthusiasts (a few are, but most aren't). They are normal people with pretty much a normal interest in fashion and a tendency to occasionally watch FashionTV a bit.

Accidental viewing represents a huge part of the audience for all small TV channels, including genre specific pay channels. In Australia the reach of a typical pay channel halves when you count those people who watch for at least 10 minutes rather than for 1 minute. Or, put another way, in any given week more than half of a PayTV's audience watched for less than 10 minutes - not a lot of involvement there!

As Erwin Ephron in New York pointed out to me, because the average US household can receive close to 100 channels "accidental" viewing tends to dominate the viewer composition (and audience value) of smaller cable channels (Food, House and Garden TV, Fashion, MSNBC Business, Lifetime for women, etc.). Considering that their average rating is typically 0.2 and the average amount of people-switching at any time is closer to a 5.0, this is effectively a tsunami 25 times as large constantly washing over the channels. The result is that even narrow-casting channels tend to collect undifferentiated (ie normal people) audiences.

Regardless of what PayTV operators say in marketing trade publications, they do know that their small specialist channels do not deliver targeted audiences. Indeed they have testified before US congress that they can not sell unbundled channels because they would then deliver audiences so small that no advertiser would ever be interested in buying space on them. They know they need channel surfing to deliver much of their ratings and, indeed, survive.

It is a common misconception that a genre specific channel must skew to a targeted audience. But in actual fact, the skew is far less than is anticipated. For example, in the UK the pay channel "Men and Motors" runs with the tag line "fast cars and women" - unashamedly targeting men. And yet, 30% of its viewers are women. But more importantly (and this point is frequently overlooked, or omitted from the sales pitch) is that hardly any men at all actually watch it. Only about 5% of male viewers who actually subscribe to the channel watch it at all in a week, and of those who do watch, they devote less than one hour to it (out of their 30 hours of weekly TV viewing). These same men in these pay TV households spend many more hours (18 !) each week watching the big Free-To-Air channels.

The story is the same throughout the world. But is particularly true here in Australia because as yet we have no high rating pay TV channels. Unfortunately many marketers make decisions based on industry mythology & sales patter rather than fact. They need to get (start seeking) the facts.

Byron Sharp

Dr Byron Sharp is Professor of Marketing Science at the University of South Australia where he directs the Marketing Science Centre (MSC). The MSC's research is financially supported by major advertisers as well as both FTA & pay TV networks such CBS, ABC, CNN, ESPN, Network Ten, and TVNZ.

Friday, September 23, 2005

Myer for sale

In previous posts I have been critical of the marketing strategies of Australia's Myer department stores. Particularly their use of loyalty programs and price promotions.

This week Coles-Myer announced it is to sell the Myer department store business, after another year of disappointing financial performance for the stores. The sale decision is public acknowledgement that someone else can run the business better.

Saturday, July 30, 2005

Tesco cola is not the reason Tesco has been doing well lately

Recently both Coles and Woolworths, the two dominant Australian supermarket chains (with combined share of more than 80%), announced a shift in strategy towards private labels. Coles has said that by 2007 30% of its sales will come from its own labels.

Australia is an interesting evolving case study because it has one of the very highest levels of retailer concentration yet private labels (or retailer brands) are only a small part of the Australian supermarket industry, reportedly sitting at 12% of sales or less.

Coles and Woolworths are now trying to mimic the strategy of most/all of the UK supermarkets.

But have they got it wrong ? I suspect that they have misunderstood a crucial aspect of UK retailers strategy, which isn't surprising because I think UK marketers haven't noticed it either.

The real success of Private Labels has been less about taking on the big successful brands, and more about quite simply improving the branding of the supermarket by putting the retailer's brand everywhere it can be, while at the same time not in anyway degrading the shopping experience for consumers.

Let me explain....

This weekend I visited a Coles supermarket to see how their private label roll-out was going. The answer is obviously very slowly. You could very easily completely miss noticing their private labels.

They were present in only a few categories, where they seem to be trying to take on the big brands. For example, here is a picture of Coles Corn Flakes right next to the leading brand Kellogg's Corn Flakes. The Coles packaging is good, and priced substantially cheaper - although marked as on special.


From the amount of boxes missing on the shelf it looks as if it is selling fast - though this could be misleading - Kellogg's Corn Flakes may have been restocked more recently.

I was interested to see how Coles looked in comparison to a UK supermarket like Tesco. And the answer is very different. Tesco brands most, if not all, their fruit, vegetables, seafood, roasted chickens, meat cuts and anything that is not usually strongly branded. Coles makes the occasional, half-hearted effort, but the vast majority of such items carry no name or the name of some farms no one has heard of or cares about. A few things like packaged cuts of meat carry a tiny Coles logo. They seem to be missing the easy and obvious ways of branding their store.

So Coles have a very long way to go, and a lot to learn about branding.

I was also struck by the huge opportunity to replace thousands of obscure brands with the Coles name. There are so many 'brands' in an Australian supermarket that would be barely recognisable to shoppers, even the ones who buy them. These are brands that are never advertised. Some are even from large companies, like the Mars/Masterfoods Promite brand (a competitor to Australian icon Kraft Vegemite).

In short Coles could easily achieve 30% of their sales from Private Label with us (and brand managers) hardly noticing the exclusion of any brands.

But instead they seem to be concentrating on mimicking big brands. Presumably because these are high volume items. And they can encourage big brands to lower their prices - which can be a good thing.

But this comes at the cost of creating tension between major supply partners. And it increases complexity for shoppers (buying Corn Flakes is now more, not less, complicated).

A more "easy win" for a supermarket is to replace as many of the 'non brand' brands as possible with their label. Over many years, in different categories and countries, this is where we have seen Retailer Brands enjoy great success - replacing the differentiated but unadvertised brands. The brand that's only on the shelf because it is the 'super cheap' brand, the 'Australian' brand, or the 'organic' brand. Supermarkets are full of these 'non brands' that take up far more shelf space than their sales justify.

Supermarkets like Tesco have, where practical, cleaned out a large number of these 'non-brands' (or "brands without brand managers") and replaced them with their own label. It has vastly improved the branding of their supermarkets - your bag of tomatoes or punnet of strawberries sitting in your fridge now says Tesco (not R&J Smith fruit farm or some other such name which meant nothing to you). And, importantly, made the shopping experience easier - the number of brands has decreased with almost no loss in variety.

Sure Tesco has also tried taking on some of the big brands. But I think this part of their strategy and success has been vastly overrated. Tesco cola is not the reason Tesco has been doing well lately. They've done well in winning share from other retailers, not from Coca-cola.

The lesson for retailers is avoid diversifying into the business of marketing individual packaged goods brands. Retailers sell shopping experiences, that is how they compete against other retailers, and to take your eye off this ball can be fatal. Better for retailers to take the easy route of using Private Label to improve store branding by replacing brands that aren't doing sufficient branding and advertising. And keep and support brands that are working hard to maintain and grow their consumer franchise.

And the lesson for national brand manufacturers is that if you don't work hard to maintain and grow your consumer franchise, if all your marketing investment is simply in-store sales promotions, then you deserve to be replaced by Private Label.


Dr Byron Sharp. Professor of Marketing Science
Director Ehrenberg-Bass Institute for Marketing Science
University of South Australia
www.MarketingScience.info

Friday, June 10, 2005

Mix modelling muddles marketers

I think econometric modelling is over used in marketing. And routinely produces misleading results. Let me explain...

Warning: people who make their living from such modelling will not like what I have to say.

Astrology, like econometric marketing modelling, has many fans, many of whom are very intelligent capable people. Proponents of astrology and marketing mix modelling each use the same arguments to justify their practice (eg that it is popular, that it is rigorous and highly technical etc), and the documented predictive track record of each is similar.

I think econometric style modelling is over used in marketing. And not subjected to serious enough criticism and scrutiny. I'm talkng about statistical 'best fit' modelling of marketing mix effects, i.e. attempts to quantify the sales effect of different aspects of the marketing mix. This is now a common service offered by all the big market research companies and specialist consultancies.

How your marketing mix affects sales is an important question, so I don't blame marketers trying something that offers a solution. Unfortunately it is very much a second best solution. Instead, or at least in addition, they should be doing experiments, and replying more on scientific models that generalise across a known range of conditions and so can be used to predict and explain.

Here are some of the problems with econometric marketing mix modelling:

Firstly, such statistical modelling works on variation in the dependent variable (eg sales or share) and the independent variables (advertising spend or SoV or exposures, pricing, media strategy, timing, point of sale, sales team emphasis etc). But very often there is little in the way of variation, especially in the dependent variable. Big established brands show little sales reaction to changes in advertising. While even smaller growing brands simply keep on their trajectory. This makes it practically impossible to correctly statistically model the impact of advertising on sales.

And some important sales drivers, like distribution, change very occasionally (new stores/channel) and often in different ways every time. This makes distribution effects difficult to incorporate into the model.

In contrast interventions like price promotions routinely produce swings up and down in sales. So these can be modelled, indeed they show up prominently... but too prominently ???

Secondly this modelling of variation assumes that correlations imply causality. And we all know that is very often not the case. Both sales and advertising spend go up at Christmas but it doesn't mean the advertising is causing (some ? all ? any ?) of the sales increase. And sales increases can cause advertising increases, as well as the other way round. Inferring causality from statistical association can be a dodgey business, but it is the business of econometric modelling.

And it is especially easy to get the strength or direction of causal assumptions wrong when you are dealing with small correlations - and generally marketing mix models are nothing more than a description of a huge host of small weak associations.

Thirdly there is the (subjective art) of what to include in the model and what to leave out. Typically advertising spend and price are included as potential drivers of sales, but what about competitors' advertising and prices ? And what about point of sale ? And numbers of merchandisers and sales people ? And trade promotions ? And publicity levels ? And product placements ? And allowance for media strategy ? And creativity and branding execution ? And new variant/product/brand launches - ours and competitors ? And seasonality ? And what's going on in other categories - complementary and competitive ?

The problem is that if some things are left out then the resulting statistical model may be wrong/misleading. But if too many variables are put in the model will be horribly complex and complex models don't tend to work - see the 4th problem below. How can the modeller know if they got it right ?

We know that modern markets are complex. And that a brand's sales depend on a huge range of influences, many of them interacting. Much of the time these complex forces (some of which are marketer controlled, most not) result in an equilibrium where brands maintain much the same level of sales this year as next and last. It says much about our knowledge of these forces and their dynamic interaction that there are so few known guaranteed ways to increase sales, and even if we know what might increase sales we can't accurately predict by how much. Awkward things get in the way like unpredictable competitor reaction, unanticipated interactive effects etc.

Yet marketing mix models claim to be able to explain sales, ie to show how the complex interaction of forces work. This seems a naive heroic goal.

Fourthly, the modeller has to make many decisions about how the various 'sales drivers' might interact with one another. For instance might a price promotion increase the sale effects of accompanying advertising and/or the other way round ? Might radio advertising work better if it is accompanied by TV ? Or might the effects be independent ?

And then there are decisions about dynamics. If one variable changes will it cause a change in another ? If we drop price what will competitors do ?

Modellers might use 'theory' to guide these decisions (rather than use their own intuition or guesses). However, there is so little in the way of empirically grounded marketing theory that they are really flying blind.

Typically then those modelers with more time and larger budgets will compare a variety of models. But against what criteria ? Typically they use a 'best fit' criteria, ie which model fits the existing (ie historical) data best. Unfortunately, and seldom mentioned to modelling clients, this means picking between a number of different models with very similar fit. There is nothing to say that the one chosen (ie best fit) model is actually the right one (if any are) or best one.

The best fit model simply fits the historic data a tiny bit better than the others, this means it models everything in the data (including all the error) a tiny bit better. But this says absolutely nothing about its ability to model a different set of data, like a future set.

And yet this is what marketers want of course. Not to describe the past but to describe a different circumstance, ie one where there are changes (to their marketing mix, their competitors, and a host of other things that happen as time marches on - like that buyers age... we all get older). Marketers want to know what will happen if they do X, not what happenned to happen last time.

What I'm pointing out is that chosen 'best fit' model is not chosen against the criteria that matters.

In comparison scientists look for simple models that hold under a wide range of conditions. This is the criteria that we use to choose a scientific model (not 'best fit'). Consequntly econometric style techniques don't get anywhere near the amount of use in science than marketers are led to believe. And when such techniques are used it is generally with multiple sets of data covering as many different conditions as possible. And scientists do deliberate interventions and experimentally observe what happens. This is a completely different approach to modelling the real world. It is one that marketers (and marketing academics) should make more use of.

PS It isn't hard to integrate a scientific approach into your marketing modelling. Key ingredients:

Some reformed statisticians
Multiple sets of data
Data reduction techniques
Willingness to experiment

Tuesday, June 07, 2005

New Marketing Science Centre website

The revamped website contains a number of free articles, and complete listing of R&D Initiative reports and seminars.

Wednesday, February 16, 2005

Market research: What do you really think?

25 November 2004
Marketing Week
39
by Alicia Clegg


What a customer thinks, and what a market researcher is told, may not be the same thing. Companies have to ask the right people the right questions at the right time to get the best from their research, says Alicia Clegg

Think back to your last visit to the supermarket. How long did you have to queue for? Was it less than a minute; one to two minutes; or more than five minutes? And what about stock availability? Could you find everything that you wanted and, if you couldn't, were the alternatives acceptable? Now (and think hard), if you were given the option, would you prefer to find all your first choices and queue for five minutes, or not queue at all but have to settle for more substitutions?

Deciding how to allocate a limited budget to best effect is absolutely fundamental to retailers. So the idea of presenting customers who have just shopped in a store with a range of scenarios and asking them, in effect, to tell you which aspect of service to invest in and which to ease up on has obvious appeal. But do such techniques really tell companies anything about the way in which customers, rather than businesses, make choices in the real world?

The question of how best to research customer service has been debated for years. It is a subject that exercises Professor Byron Sharp, director of the University of South Australia's Marketing Science Centre and global director of the Research & Development Initiative (RDI), an industry-funded venture run in concert with South Bank University. In a short paper circulated earlier this year, Sharp argues that companies make investment decisions based on research that is often meaningless. The first problem, he suggests, is that consumers rarely evaluate service experiences in the logical way market researchers would have us believe.

It's the thought that counts

To begin with, there is the assumption, often implicit in research, that people carry attitudes around in their head that determine their buying behaviour. "Most of the time, we don't give much thought to the burger that we've eaten or even the flight that we've made," says Sharp. But, he adds: "Our natural inclination is to be polite and co- operative. If a researcher asks us to give an opinion we will do our best to formulate one on the spot."

Made to measure

It is not just consumers' desire to oblige researchers that makes measuring service experience so hazardous. Other factors are at work. A classic example concerns tracking studies. If customer satisfaction is rising, most managers assume that this is down to their own efforts.

Yet according to Sharp, satisfaction measures "show little or no relationship to service delivery". Instead, they reflect factors that more often than not fall wholly outside the control of the business. To take a couple of examples: when interest rates fall, borrowers feel generally more positive about their bank's service, even though, as customers, they are being treated in exactly the same way as before. Similarly, when the press carries stories about executive fat cats, satisfaction ratings for indicators such as value for money take a turn for the worse.

The pitfalls highlighted by the RDI are issues that research agencies grapple with on a daily basis. Many have evolved techniques that compensate, at least in part, for the worst sources of bias. A case in point is the growing importance attached to looking at research in context.

To take a simple example: discovering that 80 per cent of customers are satisfied with staff friendliness, while only 70 per cent are satisfied with value for money, might lead a company to conclude that it is doing well on friendliness and not so well on perceived value.

No comparison, no results

However, the exact opposite might be the case, according to Research International group business practice director Roger Sant. He says: "To make sense of research findings, you need comparative data, not absolute scores." In this instance, that would mean comparing the company's performance to that of its competitors and knowing that the scores awarded for staff friendliness tend to be higher than those relating to value for money.

On a similar tack, Tim Wragg, global director of NOP World's customer management centre of excellence, argues that any competent researcher should be able to identify and correct distorting factors, such as seasonality or interest rate movements. So if the remedies exist, why do the problems continue?

One explanation may simply be ignorance. "There is a still a good deal of managerial naivety about research," acknowledges Wragg. Another factor, undoubtedly, is cost and pressure of time. Tracking customer satisfaction on an absolute scale is quick and easy to do - even if the results provide a less than reliable guide for service improvement. Indexing performance against competitors is more complicated and pre- supposes that companies can find people with relevant experience of competitor brands and are prepared to buy into a database containing normative data for the whole of the industry.

Even then, as Maritz Research head of analysis Jeremy Griffith points out, the solution is not perfect. "When you make comparisons between surveys there's a danger of comparing apples with oranges - there is no guarantee that the questions asked of competitors' customers exactly match your own."

You do the maths

Some of the toughest research conundrums are being investigated mathematically with the aid of statistical tools such as econometrics.

Such techniques come into their own when a researcher wants to test a hypothesis. For instance, a company might have a hunch that a dip in satisfaction is seasonal, or that consumers are playing down the importance of a perk or freebie, such as free champagne in the flight lounge or the chance to enter a prize draw. "Spotting patterns in data can open your eyes to the influences on satisfaction consumers won't freely admit to," says Ian Brace, director of research at TNS and a spokesman for the Market Research Society.

Another way to probe the subtleties of consumer satisfaction is to undertake qualitative studies. The risk here, however, is of spending more time listening to people talk about brands and services that, in the normal run of things, they would consume without thinking. So what is the way forward?

Liaising with the enemy

One approach is to stop treating all respondents as equal and to focus on those who have most to say about your service. At one end of the spectrum, this means courting people companies have failed to impress - complaint-makers or those who have recently switched to a competitor. Another potentially fruitful line of attack, says Simon Roberts, founder of research-based strategic consultancy Ideas Bazaar, is to spend time with front-line staff. "There's huge value in finding out about the processes that impede employees in their everyday work, because those are often the things that frustrate customers the most," he says

At the other end of the spectrum, researchers are devoting extra attention to clients' most committed customers - so-called evangelists - who buy the brand repeatedly and recommend it to their friends. The hope here is that by talking to advocates, companies will gain a better understanding of the emotional triggers that create brand preference. "Historically, researchers have focused on the rational bread-and- butter aspects of service, such as whether a repair was done properly," says Larry Crosby, chairman of customer loyalty research and consulting company Symmetrics. He adds: "What we under-estimated was the extent to which brand preference and loyalty are influenced by emotional responses."

To illustrate his point, Crosby describes how his own attitude towards US phone giant AT&T was transformed by a directory assistance operator, who provided him with the number he wanted for a restaurant in an unfamiliar neighbourhood and then talked him through how to get there.

The silent majority

But what about the people who consume brands casually, who are neither committed nor disaffected, but who, when all is said and done, account for the vast majority of the consuming public? Even if, as the RDI's work implies, mainstream consumers pay less attention to service experiences than was once assumed, it would seem risky to discount their views. Whenever we purchase something from a store, book a holiday or buy a coffee, we come away with a general impression of whether the service was better or worse than we expected. And - if we are caught quickly enough - we can probably say why.

One way to capture fleeting thoughts before they fade from our memories is to conduct interviews close to the point of purchase. Another line of attack is to encourage consumers to give feedback on service in whatever way suits them. Grass Roots, a company that specialises in "performance improvement" rather than classic market research, favours this approach. Executive board director Nigel Cover says: "It's about opening a channel of communication. If you call a customer while they are watching EastEnders, offer to contact them later. Alternatively, give them the option of calling you and leaving their comments on an automated system."

A question of trust

Marketers are increasingly aware of the limitations of market research, particularly the dangers of a relying upon a single source. But whatever the perils of trusting naively in consumer opinion, there exists one greater danger - and that is not trusting consumers' opinions at all.

by Alicia Clegg

Thursday, February 03, 2005

Why luxury brands really need to advertise

Economic activity in the major industrialised economies is starting to pick up. Demand for luxuries is reported to be rising, which is good news for advertising.

Luxuries tend to use a lot of advertising relative to their sales, and not just because they have the profit margins to allow it.

There is the view that fashion, cosmetic, and spirit brands advertise to build their "image", their cool, their mystique. And this is partly true, symbolic consumption brands (ie when we buy them it symbolises something about ourselves to others) benefit from advertising telling other people what they mean. There is no point in buying X to look cool if no one else knows X is cool.

Some brands have managed to build "image" without advertising, but they usually end up running large advertising or publicity campaigns - it's all very well being known by an "in crowd" but if you want to grow you need to reach more people.

And the advertising industry (and textbooks) tends to overplay this "image" reason for advertising. For competition between brands it matters little; competitive brands Prada, Armani, Versace are all cool (we all know what they mean) so now they advertise for other reasons.

Like all brands they largely advertise to build and refresh salience. To keep buyers thinking about them.

And even though luxuries are expensive they still need to reach a broad audience. Wealthy people may individually buy more, but there aren't many wealthy people so (personal jets and luxury yachts aside) a huge volume of sales comes from everyday people occasionally buying your brand.

So media that only reaches wealthy people is of little advantage. And that's why we all see a lot of advertising for cosmetics, perfume, scotch whisky, and fine wine.

Fortunately for us consumers these ads are a lot nicer to look at than the average ad.