*1. To make these comparisons between the likely sales impact of different exposures (e.g. a 30 sec TV ad compared with a 30 sec of sound-only radio ad) we currently recommend pooling management judgement. If 30 second TV ads are the main sort of spot you buy then set this as your benchmark index of 100, then ask managers to independently rate other media spots you are thinking of including in your mix, for example 30 second radio spots might be rated at 15, website banner ads might then be rated as 10, and so on. The results of this survey can be used with (a) cost per reach and (b) % of OTS that result in actual exposure together to make media buying decisions.

The total time spent consuming commercial media has increased somewhat. However, there are far more media, so audiences have fragmented, which makes popular channels, programs, magazines even more valuable as advertising spots that can quickly reach vast audiences have become rarer.

Plan your main media carefully to gain maximum reach for every dollar you spend. Care must be taken even with large reach media such as TV. TV viewing levels follow a highly skewed pattern: there are many light viewers and few heavy viewers (i.e. most viewers watch less than the average). It is much easier to reach heavy viewers (people who spend a lot of time at home, e.g. retired, unemployed) simply because they watch so much TV. This presents a trap for unwary advertisers; it is easy to pay for many exposures that reach the same people (skewing to the heavier viewers). Take care to measure campaigns on reach, not GRP/TRP as they become inflated by high frequency to heavy viewers.

A media schedule must reach light viewers if it is to achieve high reach. Light viewers skew toward viewing TV during peak times in the evening and weekends. Consequently they make up a higher proportion of audiences of high rating programs. Scheduling during prime-time helps to reach these light viewers more cost effectively if reach has to be gained quickly. Fortunately, light viewers aren’t consistently light, i.e. the proportion of light viewers stays the same at an aggregate level, but the people within the group varies from week-to-week and even from day-to-day. This means that employing a continuous scheduling strategy over time will eventually reach most ‘light’ viewers.

Data: UK, USA, Australia, 1974, 2001, 2002, 2006 and Australia, 2010
See Reports: 15, 49, 55,
Further Reading: Planning for Synergy Report (written in conjunction with CNBC)

3. Then add complementary media

Use your first media channel to deliver your most cost effective reach, and most of it. If you have a small budget you may not need to look at other media. However, those with larger budgets will find that above a certain spend level reach becomes increasingly expensive, above this point other media gain in attractiveness. Other media included in the mix should reach those consumers who are difficult for your main media to reach: for example light users of TV; or people who are more easily found in a different environment (outdoors for example); or people who are accessible at different times of day (e.g. shift workers).

Use additional media that increase cumulative reach.

Young people are generally harder to reach with most media, this has been true for many decades. Young people are much lighter readers of print, as they were before the arrival of the internet. Young people also watch less TV, again, this is not new. They are heavier users of mobile devices but for many different uses, most which do not involve viewing advertising. The best way to reach them is often by advertising in the highest rating TV programs in the early evening. We recommend carefully researching how well the likes of social media, mobile at point-of-sale, and cookie targeted campaigns are at reaching and selling to young people before they take a major role in your media strategy.

If you have evidence that one media is more sales responsive that others (e.g. if advertising via radio or pre-rolls consistently nudges more sales than equivalent exposure to outdoor advertising), then this knowledge needs to be incorporated in planning your media mix.

Data: Australia, UK, USA, 2010
See Reports: 55, 56

4. Be wary of media that claims low wastage and tight targeting

Many media options look attractive because they claim ‘low wastage’; heavier buyers of the category (or even just buyers of the brand!). This can be much less attractive than it sounds because such buyers are often the easiest to reach with any media. Even though these heavier brand buyers are valuable they don’t warrant extra exposures as they are about twice as likely as light buyers to be able to ‘take-in’ and thus recall advertising for the brand. They are also typically the consumers who are fans on Facebook, seek out information on websites and so on, again making them easy to reach.

‘Earned media’ (e.g. a Facebook fan base, or loyalty club members) in particular skews to existing loyal buyers of a brand and is not suited to building substantial reach. Be careful if your focus is on earned media (it is occasional icing on the cake): instead manage reach and weigh up the costs versus reach return.

Data: Australia, UK, 2006, 2010 and Australia, USA, 2012
See Reports: 55, 57, 60

5. Avoid ‘niche’ media – they’re usually just small and take time to cumulate reach

Low reach, niche media vehicles can present an illusion of cost effectiveness. For example, a targeted pet magazine may have a reasonable CPM for reaching pet owners, it will claim low wastage, and a highly engaged readership.

Specialist magazines are different in that more of their readers are genuinely interested in that genre/topic and may be from a particular demographic (e.g. magazines on fly-fishing are rarely bought by people who don’t fish, whereas all sorts of people end up watching a bit of a TV program on fly fishing (if it is on while they are scanning for something to watch)). However, in spite of this targeting, specialist magazines usually deliver low reach, even into their target.

Another consideration is how well the magazine cumulates reach over time, some media will do better than others. Specialist magazines that largely sell to subscribers are unlikely to cumulate reach at a reasonable rate, so buying advertising space in this magazine will end up talking to many of the same consumers (a fraction of the market) over and over again; a high frequency strategy that neglects the majority of category consumers.

The more niched a medium is, the worse it will be at cumulating reach. It will reach the same audience over and over.

TV is a bit more niched that it was, but less than most people realize. Single genre (or niche) TV channels can attract audience profiles somewhat different from mainstream channels, however not to the same degree as specialist magazines because TV channels are typically sold to viewers in subscription bundles. Preferences for the same genre do cluster together but audiences are rarely strongly segmented. For example, 45% of the viewers of children’s channels are actually adults. When planning media spend, genre specific channels should usually be judged on their reach metrics not on having special or more engaged audiences.

We see similar patterns when we look at viewers of programs. We might expect that people who like drama series or sports watch mostly these sorts of programs but, while there is some evidence of clustering, overall viewers still seek variety in their viewing. There are some differences between demographic groups in terms of preferences to program genres, for example men spend more time watching sport than women however, the majority of any demographic usually still watch the big programs and big channels. This suggests that scheduling for only sport or only drama is unlikely to generate high reach. Scheduling for high reach is often best done by ignoring program genre.

Data: UK, USA, 1984, 2003 and USA, UK, Australia, 2003, 2010
See Reports: 15, 18, 49, 57

6. Avoid program sponsorship unless a substantial discount is offered

Program sponsorship, because it usually involves placing substantial amounts of advertising in the same place, is a poor way to cumulate reach. However, program sponsorship should not be entirely dismissed as week-to-week repeat viewing levels are low (e.g. 30%). Half or more of next week’s audience will be people who missed this week’s episode. Therefore, advertising (or sponsorship) in the same program each week will inevitably reach new viewers: cumulating reach.

The main problem with program sponsorship is that many exposures occur within the same 30-60 minutes of that evening. It’s not known ( probably unlikely) that these less effective exposures are compensated for by promises of audience engagement – research is needed.

Data: UK, USA 1979, 1984, 1985, 2002, 2003
See Report: 37

7. Should you pay more for less clutter?

When exposed to advertising in less clutter, viewers typically remember a greater proportion of the ads. However, paying a premium to air an ad in a very low clutter pod may not be worthwhile, as halving clutter (which is likely to mean paying double) does not produce double the effectiveness.

While it’s ideal to minimise clutter around your advertising, in practice the best option is to focus mostly on cost effective reach and ensure you have well branded ads that people want to watch in order to achieve impact despite the clutter.

However, this recommendation may not apply for very new brands, with no existing memory structures. Such brands are likely to be particularly vulnerable to advertising clutter in which case it may be worthwhile to pay a premium for a low clutter environment.

Data: Australia, 2002, 2007
See Report: 50

WHEN TO ADVERTISE

8. Spread out your advertising – don’t burst

Spending all the budget in a short period leaves long gaps with no or little advertising. Instead spread advertising out over time. For example, at its simplest if your budget allows for 12 TV spots, then these spots should be spread out over the year, one aired per month. This continuous scheduling strategy minimises the gap between a consumer making a category purchase and them last seeing your advertising. As a consequence you are likely to be fresher in more people’s minds.

A simple rule is that every week of the year you should spend about 2% (1/52) of your advertising budget, with due allowance for seasonal changes in category demand and media costs – i.e. spend a bit more in weeks when category sales are higher and spend less in weeks when media costs are higher.

Try to avoid consumers receiving multiple ad exposures within a short time period (e.g. a single evening). Traditional claims that consumers need to see an ad three times for it to be effective are wrong, the biggest incremental change in likelihood to purchase occurs when exposure increases from 0 to 1 within a particular time period (e.g. an evening, or a week, or month). So, even with different executions, avoid bunching advertising exposures together.

Another reason for this strategy is that when exposures are bunched together closely in time viewers tend to screen them out more (as they avoid the tedium of repetitive stimuli); resulting in consumers being less attentive, which reduces mental processing of ads, resulting in fewer memories.

A further key metric to guide your media buy is the length of the gap between exposures: the aim is to minimise the gap between exposures for even the lightest viewers.

Spreading advertising over time also lowers costs. A media brief that demands high reach within a short time period will pay a high price per reach point. By far the cheapest way to gain reach is to spread advertising dollars over a longer time period – this will deliver more reach for less frequency (and hence a greater ROI on the budget).

Data: 1100+ real world and lab studies over 20 years (see meta analysis Simon and Arndt 1980 and Taylor, Kennedy and Sharp 2009), Jones 1995, Ephron & Heath 2001

See Report: 57
Further Reading: Planning for Synergy Report; Jones (1997)

9. Look for unique reach

Unique reach is valuable, though not easy to find. You have to look carefully.

Media is usually priced in terms of exposures (e.g. CPM). These exposures are made up of reach (how many people were exposed) and frequency (how many times they were potentially exposed). Pure reach means everyone receives only one exposure. All media find it hard to deliver additional reach without a lot more frequency, for example the Monday issue of a newspaper delivers pure reach that week, but the Tuesday edition delivers little extra reach with most readers being the same people as on Monday.

Look first for media that delivers the highest reach per dollar/pound/Euro/Yen – with due allowance for quality of the medium (e.g. video vs plain text). Then add media that reach as many additional category buyers as possible; increasing cumulative reach cost efficiently. The total media campaign should aim to maximise unique reach (i.e. reaching a higher proportion of all category buyers once) and minimise frequency. It should not be assumed that adding a media vehicle will achieve extra reach, adding a media can provide more duplicated audience than new audience (see Romaniuk, et al 2013).

There are varying media scheduling strategies that can be implemented to enhance reach for a set budget. In the first instance, if the advertising campaign is not affected by seasonality or time, lengthening the campaign with lower advertising weight will deliver greater unique reach. Having longer campaigns means there is a greater probability that light users of the medium will also see the advertising. In TV scheduling experiments in Australia and the UK this lengthening strategy consistently delivers 3% greater cumulative reach for the campaign, compared to the original campaign, with equivalent spots. For seasonal campaigns or for advertising that is time sensitive, further unique reach can be delivered by applying the following media scheduling tactics:

1. Spreading the advertising across different times of the day throughout the campaign – this delivered an average 9% further cumulative reach in the UK, a television market that is heavily fragmented. This also means that if the advertising is screened in selected television channel(s), further reach can simply be gained by adjusting the ad placements in different times across the day.

2. Spreading advertising across commercial networks and channels – this delivered up to 8% higher cumulative even for longer campaigns.

3. Road-blocking (placing a spot on each commercial TV channel at roughly the same time) – this delivered at least a further 2% cumulative reach and can achieve high unique reach, fast.

These strategies can be implemented for comparable budgets, which debunks the belief that further unique reach can only be gained by spending more money.

Data: USA, Australia, 2006, 2007, 2009, 2012 UK 2008, 2009, 2010 Greece, Japan, Russia, South Africa, South Korea, Spain, Taiwan 2009, 2012
See Reports: 55, 57
Further Reading: Planning for Synergy Report; Ephron, 1997 (Admap); Romaniuk, Beal and Uncles, 2013

10. Use 15-second ads

Research suggests that 15-second TV ads generate 80% of the recall of one 30-second ad. Given the cost advantages of 15-second spots it is better value to schedule two 15-sec advertisements at different times rather than one 30-sec ad. Being seen more often, by more people is better than being seen for a longer time by fewer people.

Data: USA, Australia, 2007
See Report: 48

11. Spend more when more buyers are buying

Advertising is more effective closer in time to the point of purchase. So if a category sells more intensely for a period of time (e.g. Thanksgiving), then we should be spending more to reach all those buyers in the period right before they purchase. This said, the memories that advertising refreshes do linger, so if media spots in the lead up to peak period(s) are more affordable it is advisable to spend more during these times. Maintaining some continuity
across the year in most categories is likely to still be needed because different buyers come into the market at different times. Experimentation is strongly encouraged to determine the right balance to maximise sales.

Once again, a simple rule is that every week of the year you should spend about 2% (1/52) of your advertising budget, with due allowance for seasonal changes in category demand and media costs – i.e. spend a bit more in weeks when category sales are higher and spend less in weeks when media costs are higher.

In highly seasonal categories, reach of category users preceded by a recent exposure is also key to maximise.

HOW MUCH TO SPEND ON WORKING MEDIA

Budgeting is a forecasting exercise – forecasting how much spend is needed to maintain or grow sales by a particular amount. Two key principles of good forecasting are to: 1) be conservative and 2) use multiple data sources, ideally from different perspectives. We recommend averaging the 3 approaches outlined below to arrive at a final media budget figure. Drawing on knowledge from media experiments will also be useful (ideally linked to sales response).

(A) To maintain sales brands need a share-of-voice appropriate for their market share

There is a simple, but non-linear, relationship between share-of-voice and market share. This suggests that for maintenance of share, smaller brands need to spend slightly more than expected given their size, while large brands can survive with a share-of-voice slightly less than their market share. This is because larger brands have greater physical and mental availability, so while they have to spend more to maintain their larger sales, each dollar they spend is a little more effective.

Most of the evidence regarding this pattern comes from established consumer brands in developed markets. The evidence suggests that brands with less than 13% market share should spend to achieve a share of voice at least equal to their market share; and it is even dangerous for larger brands (above 13% share) to underspend by 4 share points, i.e. risking sales declines (Danenberg et al. 2016, Jones, 1990, Hansen & Christensen, 2005).

(B) Spend about 10% of gross profits on advertising

Another simple budgeting guide is to relate advertising spend to what will generate most profit within the time period. Here the brand should set its advertising budget as its gross profit multiplied by their advertising elasticity, this is the budget optimization theorem as determined by Wright (2009). Research suggests that a typical advertising elasticity is about 0.10, but they do vary e.g. for new vs. established brands; and by market (Europe generally higher and America lower).

Gross profits are the contribution to overheads after direct costs are deducted from the sale price. Advertising expenditure, as a % of gross profit, should be at a level where net profit can no longer be improved by changing expenditure up or down; that is, the marginal return to change in advertising expenditure is zero. This means that a brand with a negative gross profit should not advertise – as every unit they sell will increase their losses. There will be situations however when this is a deliberate business strategy (e.g. to build scale and thereby improve profit margins), which brings us to the third approach to setting an advertising budget.

Data: Wright 2009; advertising elasticity from 50+ studies from 1996-2008 (see meta analysis Sethuraman, Tellis and Breisch 2011); Danaher et al., 2008

(C) Cost from the bottom up for the task required of the advertising (Task Objective)

This bottom up approach involves costing different media plans that could be bought (e.g. TV, online and print vs TV, outdoor and cinema) using the principles discussed above to achieve a particular task, for example “exposing all buyers to at least one advertisement a maximum of 14 days before they make a category purchase”. We encourage marketers in partnership with their media agency to cost a range of approaches to maximise reach metrics (for each time period: year, quarter and month) to determine how much should be spent on the brand.

Task Objective has the default objective of “reaching all category purchasers”, followed by the second tier objective of reaching them with continuity. How much continuity is needed (and when) will vary with the likes of level of competitive clutter and seasonality of sales in the category.

In costing scenarios, start with the highest reach media feasible (typically TV) then add media that bring in the most unduplicated reach (e.g. outdoor, cinema). Where there is in-market knowledge of the sales response from one media over another that information should be incorporated in media selection decisions.

For an independent audit of your media plans or to work out your ad budget contact us: info@marketingscience.info