Since 1995, Marketing Week and the Direct Mail Information Service (DMIS) have been gathering information on direct mail response rates through a magazine insert. The findings represent one of the key benchmarks within the direct marketing industry for the performance of this core medium. Although the findings are averages for all the campaigns reported, they provide a means of comparison which clients and agencies cannot find anywhere else.
This year, due to a change in methodology, the time series data from previous surveys has been dropped. The figures quoted are based on the reported volume, cost, response and other information provided on 276 campaigns.
The average direct mail campaign, based on the Marketing Week/ DMIS response rate survey, can therefore expect to achieve a 4.4 per cent response rate – from an average of 102,000 targets. If an average of 46p is spent to produce each item, this gives a cost per response (CPR) of 10.47. As all direct marketers will know, it is this figure which measures a campaign’s success.
It is also the reason why planning a direct mail campaign around financial return on investment is so difficult. The marketing budget is based on an available margin within the product or service. A certain cost of sale is acceptable which may make a customer immediately profitable.
If the client has a tendency to convert an enquiry into a sale, this acceptable cost per sale can be worked backwards to yield the cost per response permissible. With this figure in hand, planning can be used to identify the most cost-effective media, based on historical data about CPR performance. This will generate a target for response rate for each campaign.
The majority of campaigns work from the target volumes first, assuming a certain number of hits are necessary to yield a good response. As an intermediary step, response rate analysis might be used to calculate the budget required to produce the necessary sales. Even this approach is relatively uncommon.
Mick Byrne, planning director at OgilvyOne, admits: “You have to do some maths to see what is really worthwhile. You need to look at what you might expect in response to determine the amount of money you might put behind a campaign.”
The average quantities mailed by traditional direct mail users appear to reflect this methodology. The charity sector was the highest average volume mailer in this survey, at 485,000 items per campaign, but still achieved a reasonable response rate and the third lowest CPR in consumer direct mail of 7. This is due to the low cost per item (just 28p), which is itself a function of economies of scale.
Food and drink mailers sent the third highest volumes on average, at 329,000 per campaign, but still pulled a reasonable response and low CPR. But mail order mailers, who sent an average 400,000 items per campaign, must be worried about a CPR of 23.
Retailers must also be wondering about the wisdom of their rush into direct mail. Although response rates were below average at four per cent, overall volumes were quite high at 210,000 items. Worst of all, CPR was 28 per person, equal to industrial supplies in the business sector, where the value of a sale is likely to be higher, and second only to financial investment.
The question retail direct marketers should be asking their agencies is whether it is acceptable to spend four times as much per item to reap the same level of response. .
DM practitioners should feel confident they are leading through example – their campaigns pulled the third highest response in the business sector and had a well-below average CPR. They are high volume mailers in comparison, at 78,000 items per campaign on average, second only to industrial supplies (157,000). If they do get a business-to-business account as a result, they will have to learn to think leaner – most business mailings were numbered in the low tens of thousands.
The biggest problem for those involved in direct mail is the lack of information to guide them. Response rates are closely guarded, so getting into the detailed issues of CPR and conversion rates is unlikely. But there are signs of an easing in this attitude. In the financial services sector, a joint initiative called Tank has been launched by 40 financial direct marketing companies.
The project is aimed at information sharing, education and best practice. Within Tank, a database will be set up of campaign histories, to be run by Market Movements. This will allow clients to benchmark campaigns against the measures provided by their peer group, including response and conversion rates, spend and acquisition costs, volume and prospect type.
“In other industries, such as alcohol, automotive and packaged goods you know down to postcode level how you are performing in terms of market share. In financial services, we keep our cards close to our chest. There is no reason why that should be. We can learn at the same time as respecting commercial confidences,” says Sean Larrangton-White, head of direct marketing at the Bank of Scotland.
He points out two different banks might carry out direct mail campaigns for the same product and be happy with their response. But if one achieved one per cent conversion and the other ten per cent, there is an important difference in performance. Without benchmarking data, that discrepancy might never emerge.
One of the few clients to have systematically developed historical data on direct mail is the Royal Mail, which runs a Direct Mail Centre of Excellence. According to OgilvyOne’s Byrne, whose agency works for Royal Mail, “It goes through a pre-planning process to get a tight brief. It then reports on whether the brief is possible.” That includes assessing predictions on response rates through comparing campaigns by volume, target and spend.
The issue of target setting for direct mail is worth considering in light of new questions on this year’s survey.
Respondents were asked to indicate whether their campaigns exceeded target or not and whether they were satisfied with the results. Overall, 59 per cent of campaigns did beat target, and 74 per cent of respondents were satisfied.
Table 3 shows the level of response rate achieved against the proportion of campaigns which exceeded their target or did not exceed it. Not surprisingly, 12.8 per cent of those campaigns which did beat target had acheived over 15 per cent response – who would not be satisfied with that? Successful campaigns are grouped fairly evenly across all levels of response, which can be taken to reflect careful and cautious forecasting.
Looking at the proportions which did not beat target tells another story. Of all the campaigns which failed to meet their target, nearly one quarter produced less than one per cent response. Few agencies or clients are likely to set themselves such a low threshhold for success. But 74.6 per cent of all campaigns which failed to meet their targets had yielded a response of under four per cent.
In other words, three quarters of all direct mail campaigns that subsequently failed to meet expectations had been set targets of below this survey’s average (4.4 per cent). Given that four out of ten campaigns were a disappointment, this means of all activity, 30.6 per cent will not meet the target.
This appears to be at odds with the level of response rate at which survey respondents said they would be satisfied – 95 per cent said campaigns that pulled four per cent or less were satisfactory. Unless 65 per cent of people in this survey are lying about the targets which they are satisfied with, there is a real disparity between actual target setting and the real performance of mailings.
On the positive side, 23 per cent of all direct mail will be set a target which is below this survey’s average and will beat it. Comparing performance with satisfaction, 90 per cent of those carrying out industrial supplies mailings were satisfied, even though just 40 per cent of campaigns beat targets. This suggests realistic expectations and planning. Conversely, in retail direct mail, just 50 per cent of those doing such campaigns were satisfied and only 50 per cent beat target. Half of all direct marketers working on retail accounts clearly expect too much.