Tag Archives: ROI

DRTV Response Rates

Published / by Simon Foster

We’re often asked to forecast or estimate campaign response rates, especially in DRTV. Here are some guidelines for those who want them:

Set 1 – DRTV Phone Response Rates (high to low range as a percentage of total impacts)

  • DRTV Type 1 – Hard Hitters – these DRTV hard hitters, with no nonsense creative, usually on a 60 second time length can achieve between 1% and 0.05%. But please note, exceeding 0.05% is a very rare achievement in DRTV. It’s usually delivered through a combination of an extremely powerful ad, very strong product, with a great offer transmitted on a low level but highly responsive audience. It is very difficult to exceed 0.05% at scale.
  • DRTV Type 2 – Lead Generators – these DRTV ads are usually seeking subscription trials, leads, quotes etc and run on time lengths between 30 seconds and 60 seconds.  Response rates tend to be around 0.05% and 0.005%.
  • DRTV Type 3 – Brand Response – these ‘BRTV’ soft sellers produce lower responses generally in the range of 0.005% to 0.0005%

Set 2 – DRTV Web Response Rates (high-low range as a percentage of total impacts)

  • DRTV Type 1 – Hard Hitters –  these are high response rate ads will generate 2-3 times their phone response equivalents so around  2% and 0.1%
  • DRTV Type 2 – Lead Generators – web response rates to these tend to generate around 0.5% and 0.05%.
  • DRTV Type 3 – Brand Response – these BRTV soft sellers produce lower responses generally in the range of 0.05% to 0.005%

How to get the best from DRTV

Published / by Simon Foster

Many advertisers are returning to Direct Response Television (DRTV). Whilst the goal today is to maximise web response as opposed to phone response, many of the rules of traditional DRTV remain constant. Here’s a summary of how to get the best from Direct Response TV:

  1. Remember all DRTV begins with the offer. Whilst issues around DRTV performance are often seen as “creative” or “media” we need to remember that the proposition to consumers is key to DRTV success. If you are offering free Ferraris you will not need to think in terms of creative or media optimisation. The offer will work. Equally, if you are offering a poorly differentiated product or service, you will find it difficult to sell. Your problems will be exacerbated further if you are in a mature market packed with established offers.  So ask yourself the “so-what” question against every line of copy. If you wouldn’t buy it, no-one else will.
  2. Develop compelling DRTV creative. DRTV seeks behavioural change, and consumers need to be given good reasons to stop what they’re doing and do something else. You need to talk in terms of meaningful benefits. There are certain category rules that are helpful. If you’re selling a financial product don’t use jokes. For most people, talking about their hard-earned money is not a funny business. Concentrate on explaining what the product is different, what it offers that is new and why your target audience should find out more.
  3. Be careful with emotional sales messages. Most mainstream advertising seeks to build emotional connections between people and brands. For many brands this is the right approach, but if you want to sell off the screen, stick to promoting the benefits that make you different and giving good reasons to buy.
  4. Make sure the creative identifies your target audience. Everyone watches broadcast media. The trick to making DRTV work is to create a sense of identification between you and your target audience. Show people and situations that your target audience will identify with. Create the impression that your target audience belongs in the ad.
  5. Understand the economics of broadcast media. TV companies use every possible device to maximise the yield on the audience they are selling. Yield is the revenue generated by advertising over the cost of attracting that audience i.e. producing or buying the programming.  High quality peak programming is expensive to produce or buy. More people are at home available to view during peak viewing times (5pm to 11pm) so audiences are higher. So TV companies put their highest cost, highest quality programming into the times of day when most people are available to view.  Moreover, ad agencies want high reach, so there is high demand for high quality, high audience programming. All this makes peak airtime expensive both in terms of unit cost and capital cost. The premiums embedded in peak mean it sells at rates that rarely work for DRTV advertisers. But off-peak airtime is an entirely different matter….
  6. Unlock the benefits of off-peak airtime. Everybody watches off-peak TV; young, middle-age, old, affluent, less well off, single, married, students, working people and those who’re retired. People take holidays and have days off. But most importantly, the homemaker watches daytime TV and the homemaker is often the person who researches and makes financial plans.  Daytime can be ideal for reaching family decision makers. Daytime can be ideal for reaching students and it can be ideal for reaching affluent grey markets. Off-peak is the place to test your product in DRTV.
  7. Test, measure and learn. DRTV is direct response marketing and the lifeblood of direct response is quantified planning and control. Using BARB data it is possible to match minute by minute audience data with your minute by minute click traffic. This allows advertisers to build a response database which matches TV audience with web response. Each spot can be defined by a number of planning variables that can be controlled: day of week, time of day, channel, proximity to previous spot, length of spot and creative execution. All these factors can be combined and used to optimise future DRTV media buys.

Advertising Evaluation Techniques

Published / by Simon Foster

Techniques for tracking advertising are often discussed by both advertisers and agencies as they seek to identify and maximise the ROI effect of media budgets. Deciding on which techniques to use can raise a number of issues depending on the data and budgets available for advertising evaluation.  All are data dependent which means that if you are not collecting response or sales data you will need to. Costs for implementation can vary but should always be viewed in the context of the potential savings that can be made from subsequent optimsation. For example, if a regression model costs £25k, but can optimise a £2.5m budget to save £500k, then the £25k is money very well spent. Here is a short summary of advertising evaluation techniques.

You will  notice there is no mention of coverage or frequency here. That’s because coverage and frequency are useful media planning metrics but they are not direct measures of ROI – coverage and frequency are measures of audience delivery not sales response. Reach and frequency may be linked to sales response but in my opinion spend levels or GRP weights and diminishing returns in specific time intervals are more robust ways of understanding sales response to advertising.

Does social media drive sales?

Published / by Simon Foster

The question of sales generation is a growing problem for social media. Despite all the hype, it’s almost impossible to find any conclusive cross-category evidence that social media drives sales.  Yes, there are some isolated examples of success; Dell’s Twitter pages announces some great deals and I’m sure ASOS can whip up a bit of extra demand by tweeting Axl Rose’s US flag shorts, but the reality for most brands is that they are going to struggle to make social media deliver measurable sales.  This view might not be flavour of the month, but the four experiences of social media listed below certainly give the “no sales” view a high degree of credibility.

  1. In 2010, Pepsi undertook a massive social media initiative called The Refresh Project which was designed to give $20m to good causes. According to Bob Hoffman, the AdContrarian, it delivered over 80 million votes, almost 3.5 million Facebook likes and nearly 60,000 Twitter followers. But there was just one big problem; it didn’t drive sales – despite the funding coming from Pepsi marketing budgets. Pepsi’s sales fell in the year the project ran and the brand lost 5% market share worth about $350m. To make matters worse, if that were possible, Pepsi slipped to third in brand share behind Coke and Diet Coke.
  2. In both 2012 and 2013 IBM used data from around 800 e-commerce sites to track social media’s contribution to sales. In 2012 it arrived at a figure of 0.34%. In 2013 it didn’t publish the number, but hinted that it was even less.
  3. In September 2012, one of the world’s leading digital research companies, Forrester Research reported that “Social tactics are not meaningful sales drivers. While the hype around social networks as a driver of influence in ecommerce continues to capture the attention of online executives, the truth is that social continues to struggle and registers as a barely negligible source of sales…”
  4. In March 2013, Mark Ritson, formerly a professor at London Business School observed in Marketing Week that “….marketers are finally beginning to apply some measures to assess the ROI of their [social media] efforts. Once they do that they can do the one thing the social media mavens have counselled against: compare the value of social media with other options, apples to apples. And, in many cases, they are discovering the hullaballoo drummed up by the marketing media and various industry events is not quite all it was cracked up to be.”

I think most people in social media are well aware of this “no sales” problem. And because social media can’t deliver sales, they’ve invented a snow-storm of flaky measures designed to obscure harsh commercial realities. These measures include: ‘likes’, ‘fans’, ‘followers’, ‘shares’, ‘retweets’, ‘pins’, ‘follows’, ‘friends’, ‘influence’, ‘amplification’, ‘forwards’, ‘mentions’, ‘tags’ and ‘reactions’. In a commercial context these are nothing more than diversionary measures. They might enable some positive looking PowerPoint charts but they don’t deliver positive looking sales. These are ROI potatoes, when everyone else is comparing apples.

Amazingly, when social media campaigns fail to deliver sales, social media experts almost always suggest that it was the company management who got it wrong rather admitting to any shortcoming of social media itself. Whilst this claim blames marketers and management, it also spawns a convenient stay of execution for social media’s “gurus”; failure brings an opportunity to “learn lessons”, to “revise approaches” and to “develop new strategies”. In other words social media failure provides a new opportunity for marketers to waste even more money on social media activity.

Marketers badly need a serious reality check on social media. Social media environments aren’t much more than an online version of a public waiting room. People drop in, take a seat, look around and leave. They may leave a bit of rubbish. They may take a bit of rubbish with them. But that, I’m afraid, is pretty much the long and short of it for most brands. Don’t spend too much time in there, nothing will come of it.

If this sounds old-fashioned, I make no apologies. Advertising exists to drive sales.  To have advertising that doesn’t drive sales is like going to a dentist who doesn’t look at your teeth, or a barber who doesn’t cut your hair, or a mechanic who won’t fix your car. If what you’re doing can’t be directly or indirectly linked to generating sales, you’re wasting precious budget.

What is programmatic advertising?

Published / by Simon Foster

If you are an advertiser you may have heard the expressions “programmatic buying”, “real time bidding” and “ad exchanges”. You may be wondering what all this is and what it means for advertisers, if you are then read on…

“Programmatic” advertising, is effectively automated online media buying – often at large scale and at very high speed (faster than lightening in some cases). Advertisers use computers to participate in real-time automated auctions for digital ad space across a large number of publisher web sites.

Bidding is supported by big data analytics; predictive algorithms are used to target bids using information about web users such as location, platform, device, browser, and where available, other forms of behavioural data relating to specific but anonymous users. There is some big data science behind this, much of which has its origins in high frequency algorithmic trading in financial markets where the principles are very similar to automated online advertising. For example, information about a user is matched to a bidding rule in a minute fraction of a second – enabling a bid to be made and a relevant ad to be served by the time the page being visited by the prospect fully loads. This high speed automated decision making is not dissimilar to rule-based or algorithmic trading in financial markets.

So is programmatic advertising important? Yes; it’s important to the long-term health of digital display as a medium and it’s important to advertisers in terms of increased advertising efficiencies. Let’s look at each of these.

Firstly, automated buying is boosting the fortunes of digital display advertising by creating renewed interest in the medium. Online display has struggled to demonstrate efficiency in the face of PPC which is based on pay per click (PPC) trading. For many years display has been traded on a CPM basis, that’s simply the cost of reaching people in their thousands, with no accounting for click or sales performance. That’s why Google has commanded such as large share of digital budgets over the last decade. But programmatic buying allows advertisers to place data-driven bids to ensure campaigns deliver the most responsive target audience at the right rate. This will significantly boost the ROI delivered by digital display and make it much more competitive with PPC. This in turn should enable it to take larger share of digital advertising budgets.

For advertisers automated buying offers a real opportunity to increase ROI from digital display. This opportunity comes from three sources: ROI-based trading mechanics, better ROI based audience targeting and clear performance transparency. All this offers advertisers a chance to make digital display much more cost effective. Moreover, the increased efficiencies delivered by automated trading will make digital display more competitive against PPC. Long term, automated display buying could have the effect of diffusing spend out of PPC alone and across the two platforms – theoretically this reduction in demand could reduce bid prices in PPC.

Are there any down sides?

It remains to be seen whether automated buying – which by its nature can reduce ad revenue – will deliver consistent long-term growth to digital display. It’s also worth noting that not all media owners will sign up to ad exchanges; those who feel they can realise the value of a web site more holistically than the lowest CPC denominator may well be resistant to signing too much inventory over to automated trading platforms – leaving them to fight over the lowest value inventory.  There are also  issues around the quality of the traffic delivered through high volumes of remnant inventory – remnant inventory is by its nature ad space that can’t be sold by normal means because it’s not demanded by media buyers. Buying remnant inventory through ad exchanges can mean you are buying into some low quality sites which may not be right for your brand’s image.

On a clear day: Measuring ROI in Social Media

Published / by Simon Foster

Measuring ROI in social media is a big concern for marketers as they consider moving budget away from traditional media channels and into social media activity.  But before they can invest in social media, marketers need to get an idea of what it can contribute to their brand.  This has driven a debate about measurement in social media but unfortunately much of the discussion is focused on measuring social media for social media’s sake. What we should be asking is how do we measure the delivery of marketing objectives when we run activity across the social media platform. When we look at it this way we focus on measuring marketing outcomes versus marketing objectives and the answers become much clearer.

As a start point, everyone needs to recognise that social media is a media channel. It is not a marketing discipline. It is not a marketing objective. It is not a marketing strategy. So we might use the social media channel to raise brand awareness (objective) by targeting affluent new car buyers in social media (strategy), we might use social media to increase consideration (objective) by informing new car buyers about the unique benefits of the car we are selling (strategy) or we may use it to increase sales (objective) by communicating a last minute ‘walk-in’ trade-in deal (strategy). The metrics we use to measure social media should therefore relate directly to the objectives and strategies that we managing through the social media channel.

So, before we can measure social media we need to understand what we want social media to deliver from a marketing perspective. Only then can we select the right types of measurement and metrics to get the job done properly. Here are three examples of how we might measure social media activity against the delivery of three different marketing objectives:

  1. Objective: Raise Awareness: There are a number of good tools for measuring online brand awareness, ad awareness, product awareness and salience. Ad Index from Dynamic Logic allows you to play ‘spot the attitude difference’ between web users who have been exposed to your messaging and those who have not. You can ask exposed and non-exposed groups bespoke questions about your brand and campaign activity which allows you to contrast and compare the differences between the two groups. Brand sentiment can be measured using sentiment trackers like Sentiment Metrics; through without bespoke surveys these may include a range of external references to your brand, not just your own social media activity.
  2. Objective: New Customer Acquisition: If we want to use social media as a new customer acquisition tool then we should be using customer acquisition metrics. Microsoft’s Atlas can be used to track the online behaviour of your social media users across all touch points in the sales funnel. Bespoke tracking URLs in your social media pages can be used to identify visitors to your site originating in your social media pages. This type of tracking means you can ultimately relate customer value back to your social media activity.
  3. Objective: Increase Retention / Loyalty: Here we can combine online tracking, data collection and customer data analysis to understand the contribution of social media. We can collect prospect and customer data in social media pages or in pages that link directly to social media. Fusing data collection with online tracking means we can find the data source of known named customers and measure their progress and value in the sales funnel and through cross sell and up-sell. The results from this type of activity may not be instant; customer value from market source can take a year or more to establish, but once it’s in place you will be able to see how social media is building sales revenue for your business.

The message is that we can’t measure social media for social media’s sake. We should always be measuring how social media performs against a given marketing objective. If we are clear about this, the techniques and metrics for measuring and evaluating social media ROI become much easier to identify, select and implement.