What do we know about Direct Response television advertising

In the first of an occasional series on best practice in direct response advertising, Beverly Barker looks at television

Beverley Barker

DIRECT RESPONSE advertising is not only the preserve of marketers without retail distribution. It can be an integral part of communication strategies where the objective is to achieve accountability, customer information or sales leads, or to reach selected lifestyle cluster groups with a relevant targeted message.

Direct response advertising, whether 'above-the-line' or 'below-the-line', provides the researcher or planner with additional quantitative information and feedback upon which to base future decisions.

Direct Response Television (DRTV) planning requires the use of traditional television research, in combination with quantitative and qualitative analysis of response information.

HISTORY

Direct response television was started by Al Eicoff, a copywriter in Chicago, in the 1940s. At around the same time the technique was taken on by Lou Wunderman who wanted to produce accountable advertising to market products and services directly to the consumer. The growth through the '60s and '70s was further enhanced with the introduction of cable television in the '80s. In 1986 the home shopping networks were launched. For the '90s US DRTV is exploiting advances in interactive technology, allowing the viewer to complete the transaction at the time of broadcast.

According to Teledirect Magazine in the US, 54 per cent of all US commercials carry a telephone number, and after 10 pm the figure rises to 85 per cent.

Within the UK, the advances in telephone technology for marketing facilitated a rapid growth in the use of this technique between 1988 and 1992 (Exhibits 1 and 2), particularly from the charitable and financial sectors. Freefone 0800 was introduced in 1985 (freephones have been available in the US since 1967).

EXHIBIT 1. UK (0800) & 0345 (BT) MARKET GROWTH

Calls per capita
19860
19871
19881.5
19892
19903
19914.5
19928.25
19938.5
Source: BT

EXHIBIT 2. TOTAL TV EXPENDITURE BY MAIL ORDER ADVERTISERS

TV adspend (000)
19872300
19882116
19893223
19907137
199111449
199215090
199311252
199414406
Source: Register MEAL

Creative testing has been notable, with commercial lengths increased from the average 30-second executions to 60-second and 90-second lengths. In 1992 120-second executions were being tested. However, these have reduced in usage over 1993 and 1994.

INDUSTRY RESEARCH

At the end of 1993, AGB carried out the first DRTV research study (sponsored by Channel 4 in conjunction with British Telecom). The study merged two sets of data; the audience to DRTV commercials and telephone response data. These were cross-analysed against various criteria to establish what factors produce the best response rates.

Broadcast Monitoring Services monitored the airtime to detect the inclusion of a telephone response mechanism. The research period consisted of three waves, each of two weeks, ending 03/10/93, 31/10/93 and 05/12/93. Commercials meeting these initial criteria were recorded for date, channel, time, programming environment, commercial length, brand/product advertised and duration of exposure of the telephone number on screen. This detail was merged with BARB audience data, with each commercial separately identified. These data were in turn merged with the response information collated by British Telecom. The report is available from many sources, including Channel 4.

The research produced a series of industry averages that provide a useful starting point for new advertisers.

SUMMARY OF RESEARCH FINDINGS

A sample base of 91 campaigns delivering 8457 separate DRTV commercials was used. Seventy separate telephone numbers were noted, indicating that some advertisers use the same number for different brands.
  • Highest recorded response rate was 0.23 per cent: for every 10,000 viewers, 23 people made a call.
  • Average response rate was 0.038 per cent.
  • Day of the week to advertise: Monday to Friday indexed higher than Saturday or Sunday.
  • Time of day to advertise: 12am-4pm indexed higher than average for any one day.
  • Commercial length: 60 and 90 seconds proved the most effective, even when weighted for cost factors. Ten-second commercials supporting longer time lengths also produced good results.
These, and other results (Exhibit 3-6), indicate that viewers may well not wish to spoil their viewing, and prefer to phone between programmes.

EXHIBIT 3. RESPONSE TO BREAK PATTERNS

INDEX
Centre break77
End break136
All commercials index = 100

EXHIBIT 4. RESPONSE TO LENGTH OF TIME TELEPHONE NUMBER APPEARS ON SCREEN

INDEX
1-5 seconds50
6-10 seconds83
11+402
All commercials index = 100

EXHIBIT 5. RESPONSE TO TELEPHONE NUMBER WITH AND WITHOUT VOICE-OVER

INDEX
With voice-over142
Without voice-over44
All commercials index = 100

EXHIBIT 6. RESPONSE TO FREE-CALL vs PAY-CALL

INDEX
0800 Freephone105
0345 Local rate78
All commercials index = 100

The research was repeated, and extended, across the first quarter of 1995, to establish this ongoing trend.

KEY ISSUES FOR PLANNING DRTV

The planning of a successful DRTV campaign involves a ten-point plan to address the key issues:
  1. Set response objectives (eg cost per response/sales, brand vs response weighting).
  2. Define the target audience: identify the channels and times of day that reach the target group (use TGI/BARB, hold focus groups).
  3. Plan well ahead (creative, production, fulfilment).
  4. Fix a test budget.
  5. Agree creative/production values.
  6. Select a memorable telephone number.
  7. Establish a telemarketing facility.
  8. Calculate maximum response patterns, ie audience capacity.
  9. Test the system.
  10. Set benchmarks for response analysis and forecasting.

RESPONSE TARGETING: TESTING

Test areas for new projects can be selected in the usual way: TGI profiling (or other relevant data source) of product consumption by region, and average impact delivery by time band, as provided by BARB. Test area selection criteria may also be affected by the need to minimise the cost (by choosing a small area), or to deliver a cost-per-response no greater than a predetermined maximum (by choosing from available cost data). For example:
  • Target cost per sale: £50.
  • Target conversion rate: 20 per cent.
  • Target media cost-per-response £10.
Assuming a response rate of 0.05 per cent, the advertising would generate 0.5 calls per thousand viewers. The media price per 1000 viewers would therefore need to be no more than £5.

In general, areas with a relatively small universe, and without extreme user bias, are best selected to provide useful information for prediction of response generation. The test campaign must be constructed to ensure that the maximum information is derived prior to 'roll-out'. Criteria should include:

  • Creative execution;
  • Commercial length testing;
  • Position in break, top and tail, consecutive breaks, etc;
  • Day of week;
  • Day part;
  • Programme environment;

AUDIENCE CAPACITY

Successful DRTV campaign planning demands an understanding of the nature of the call-handling operation, the number of telephone operators, and the length of time it takes an operator to handle an effective call through to completion. This will establish maximum generation targets by day part. Calculation of the individual station impact delivery will then identify which airtime can and cannot be used.

Television response arrives in peaks, with around 70 per cent of all calls being generated within 15 minutes of airtime transmission (source: Channel 4/BT research study). If the telephone marketing team consists of 50 operators, the airtime on Central will be borderline, whilst London will cause call overload and calls will be lost. Defining accurate response rates from the test activity is a priority for effective future planning (Exhibit 7).

CONCLUSION

Planning DRTV advertising requires the overlaying of quantitative research parameters to traditional TV response techniques. It necessitates an understanding of the client's capability to process response generated, and must be based on thorough campaign testing. Having said that, it will enable the planner to produce an accurate campaign forecast, and to control the advertising campaign in a manner that is unavailable to the traditional TV advertiser.

NOTES & EXHIBITS


Beverley Barker

Beverley Barker

Beverly Barker is a director of TMD Carat, and is responsible for the direct marketing division, TMD Direct. She joined Carat in 1992, following an agency career which included work at Davidson Pearce and Ogilvy & Mather.
What do we know about Direct Response television advertising

Exhibit 7. Average response delivery by quarter hour