Letterbox Consumer Survey

METHODOLOGY

The initiative

The Letterbox Consumer Survey is an initiative born of the need to collect solid, current and objective data for the Belgian market (agencies and advertisers) about customer attitude and behaviour toward their letterbox and, more specifically, toward advertising mail (direct mail and mass mailings). These data, and more precisely, impact measurements, will enable them to build better campaigns.

The Belgian Direct Marketing Association (BDMA) has set up a technical commission of media and direct marketing experts to write the specifications, select the market study institute and develop the most relevant methodology.

TNS Media was selected as the market study institute following a two-round process.

The study is a first. It provides mail box media with a very wide spectrum of media data collected using a very reliable methodology. This site is another innovation. It provides direct access to the significant data collected about direct mail.

Objectives and methodology

The goal of this study is to:

  • enable advertisers and agencies to assess the impact of direct mail campaigns;
  • provide a reliable reference and a recognised measurement system for the letterbox, and more specifically, for direct mail.

To achieve this, the technical commission chose to use a representative "panel" approach with two complementary steps:

  • Step 1: The "Establishment Survey" or framework survey. This step uses one-on-one interviews to understand and describe general behaviour, attitudes and preferences toward "letterbox" media and the profile of respondents. A customer panel will be created based on the respondents to this study.
  • Step 2: "Behaviour recording" During this step, the participants recruited during the first step will fill in a notebook identifying each piece of advertising mail they receive and indicate their response to the mail.

Below are the main methodological details of the two steps:

Establishment survey:

Sample
  • N=1245
  • Representative sample of the Belgian population over 18
Method One-on-one interviews (CAPI) at home
Survey period May-June 2013
CIM Golden Standard Weighting Gender, age, education, profession, social group, region

Behaviour recording: "notebook"

Sample
  • 3,000 notebooks were gathered during the five successive measurement waves.
  • The notebooks come from 1,600 different respondents (additional panelist recruiting was needed during the fifth wave to offset the natural erosion of the panel) meeting the quotas required to create a representative panel of the population over 18. On average, each participating respondent had to fill out three notebooks.
  • The recruitment quota was based on gender, age, social class and internet access.
Method
  • Self-administered questionnaire (paper notebook) covering one week (seven days) of mail receipt and two weeks of behaviour
  • Recording of:
    • the sender's identity (sector/brand) and the type of relationship (client/non-client)
    • behaviour: reading and response type
Survey period
  • The survey was split over several periods in order to cover a greater variety of sectors:
    • Wave 1: May – June 2013
    • Wave 2: September – October 2013
    • Wave 3: November – December 2013
    • Wave 4: January 2014
    • Wave 5: February 2014
CIM Golden Standard Weighting
  • Gender, age, education, profession, social group, region

Access to the data and definitions

Access to data:

This tool enables access to direct mail impact data in two or three steps, depending on the user’s need:

  1. Choice of the sector: the entire sector or a specific sector
  2. Choice of the universe: for some sectors in which the number of findings was sufficient (supermarkets, financial institutions, mail order), it is possible to select a reference universe (Fr-Nl, Men-Women, Purchasing decision maker) that will enable the selection of other variables in step three
  3. Choice of specific criteria: access to other socio-demographic criteria (social classes, etc.).

Note that the level of socio-demographic detail available for each sector can vary based on the number of findings available. For example, we suggest grouping age classes 18-34 when the number of findings for the 18-24 and 25-34 classes is not sufficient.

Definitions

Reading and action rate:

  • The reading rate is the total of three reading levels: fully read, partially read or skimmed/glanced at.
  • The action rate measures the action or intent to respond at three levels: search for information (online or off-line), visit to a point-of-sale (to test a product) or purchase.
  • All of the percentages are expressed based on the total gross number of mail pieces received.

Direct Mail:

All addressed commercial mail with an advertising purpose including letters, commercial catalogues and magazines and any other printed commercial materials that include a name. This, therefore, does not include business or personal mail, administrative mail, bills, press (newspapers, magazines) or packages. Unaddressed advertising flyers are not included in the tool either.

Relationship to the sender:

  • Client: the respondent knows the sender and is their customer.
  • Non-client: The client knows the sender but isn't their customer or the client doesn't know the sender.

Social groups:

Social groups are defined using a ratio calculated based on the profession and education level of the main income earner. The population is classified by order of the value obtained by combining these two variables and divided into eight groups which are roughly the same (roughly because categories calculated this way can't be split up). By convention, "Group 1" is the group that corresponds to the highest values (and therefore to the highest professional and/or education level) and "Group 8" to the lowest values.

Nielsen region

Nielsen 1Provinces of West and East Flanders
Nielsen 2Provinces of Antwerp and Limburg; province of Flemish Brabant with the exception of metropolitan Brussels (30 municipalities)
Nielsen 3Metropolitan Brussels (19 municipalities) 19 municipalities + Drogenbos, Grimbergen, Kraainem, Linkebeek, Machelen, Rhode-St-Genese, Tervuren, Vilvorde, Wemmel, Wezembeek-Oppem and Zaventem
Nielsen 4Provinces of Hainaut and Walloon Brabant (arrondissement of Nivelles)
Nielsen 5Provinces of Liège, Namur and Luxemburg

Family size

Child under 15: people surveyed belonging to a household with one or more children under 15