Title: Worldwide quality measurement systems TransQ Monitoring System for Deutsche Post Global Mail
1Worldwide quality measurement systemsTransQ
Monitoring System for Deutsche Post Global Mail
What do we see?
Nobodys Unpredictable
2Contents
- About Ipsos
- TransQ at a glance
- Cycle of the survey
- Sender / receiver tasks
- Validation
- Transponder management
- Background of panel management for worldwide
measurements - Recruitment of panellists Process Training
- Control of panellists
- Quality management of panel
- Expected risks and technical options
3 4Ipsos Worldwide
- Ipsos Worldwide
- Founded 1975
- Head Office Paris
- Rank among global survey-based research
companies 3 - First listed on Paris Stock Exchange July 1,
1999 - Total Revenues in 2006 857.3 million euros
- Number of countries with Ipsos offices 50
- Total number of countries we do research in 100
- Number of full-time Ipsos employees 6,800
- Number of clients around the world 5,000
- Interviews conducted each year 6 millions
5Quality Research at Ipsos
- Competence Centre Quality of Service Measurement
Hamburg, Germany
Since 1990 more than 20 specialized employees of
the Quality Research Department measure based
on approved methods of market research the
quality of services in postal business
6Quality Research - Our Business
- Quality of service measurements in the postal
business - National and cross-border transit time
measurements - Quality of delivery (e.g. damage, deposition not
in mailbox) - Reach of Direct Marketing Mailings
- Electronical capturing with transponder
technology - Process Measurement in Logistics
- Online-Monitoring
- Reference Clients
- Deutsche Post World Net Deutsche Post Global
Mail - IPC International Post Corporation
- Swiss Post International
- La Poste, C.N.
- Posta Romana
- Other third party operators
7 8Worldwide measurements - Trans Q
- What is measured?
- Coverage of real end-to-end stretch, i.e. from
customer to delivery - Characteristics of the study
- Test Mail volumes up to 30.000 test letters per
year (including 25.000 transponders) - Mail Characteristics Formats P/G, Weight up to
150g (depending on country) - Number of Links currently 130 links
- Panellists more than 250 independent receiver
panellists worldwide in 35 countries - Senders Real customers professional senders
- Weighting mechanism 3 weighting steps number
of test letters per linkdepends on importance of
link - Central project and panel management from Ipsos
Hamburg, Germany
9(No Transcript)
10 11Study cycle of Trans Q
12Sender tasks
- Sender/ employee
- Print out the daily letter documents which are
available from the internet - Insert the letter documents in window envelopes
- Layout of the envelopes according to their real
mail - Insert a transponder if foreseen
- Induct the test letters according to requirements
- Confirm the daily sending including the
transponder ID via Internet to Ipsos
13Webpage sender portal /1
Personal password secured homepage
14Webpage sender portal /2
First daily step for the senders Download of
test letters
15Webpage sender portal /3
Further steps for the senders Production of test
letters and daily confirmation of correct sending
16Receiver tasks
- Receiver/Participant of Ipsos WMM
- Empty their letter box daily
- Confirm the receipt and transponder ID via
Internet - Give notice on their absence and answer promptly
to emails - assess the state of the mail (bent, wet, dirty,
etc.) - report not received items, in case a mailing has
not been received by them until JX (exact timing
to be agreed upon) - Sending the transponder back to Ipsos (return
envelopes provided by Ipsos) - Test letter will be disposed by panellist
17Webpage receiver portal /1
First step for the receiver after
login Selection of the received test letter
18Webpage receiver portal /2
Next step for the receiver Confirmation of the
data.
19Online Validation
- The following checks are done automatically by
the system during data entry - Send date is in the future (fatal error)
- Send date is on a Saturday, Sunday or national
holiday (soft error) - Send date is before the desired send date (soft
error) - Send time is not desired send time (soft error)
- Date of receipt is in the future (fatal error)
- Data is not complete (fatal error)
- Date of receipt entered seems to be too long ago
(soft error) - Date of receipt on a Sunday or a national holiday
(soft error) - Date of receipt on Saturday (soft error)
- Postmark date is in the future (fatal error)
- Postmark date seems to be very long ago (soft
error) - Test number wrong/missing (fatal error)
20Offline Validation
- After the online data entry all sample data is
checked for completeness, consistency and
plausibility -
- Invalid date (sent, received, postmark)
- Postmark before / beyond send date
- Induction date on a non-induction day, divided
between private and business (Saturday, Sunday,
national holiday) - Induction time not the desired time
- Delivery date on a non-delivery day (Saturday,
Sunday, national holiday) - Delivery date not after induction date
- Induction time after last collection time
- Transit time exceeding a predefined value
21Transponder management
Ipsos is familiar with transponder management
since 2000 which mainly consists of the following
aspects
- Transponder cycle
- Insertion at the production unit (wherever)
- Sender confirms the induction of test letter
transponder - Receiver confirms the receipt of test letter
transponder - Receiver sends transponder back to production
unit - Transponder handling
- Each transponder has a fixed baptised number
which has to be matched with the test letter id
via scanning - Transponder monitoring
- Each transponder has assigned the status of
condition (ok, defect, lost) and the status of
location (stock, out to sender, out to receiver,
received). - Transponder loss
- Only panellists with proven activity should
receive transponders
22Transponder readingsOverview
Technical tracing of test letters if diagnostic
gates are available.
23- Background of
- panel management
- for worldwide measurements
24Panel Recruitment - Process
- Panel recruitment is mainly done by phone / Ipsos
call centre - Each respondent answers two consecutive
questionnaires on general suitability - 1st via phone to verify their command of
English - 2nd on paper/email
- after receiving detailed information on their
tasks confirmation of profile and participation
agreement - Further possibilities
- ? Ipsos network
- ? Participants recommend participants
- ? Internet
25Panel Recruitment - Training
- Respondents receive detailed information on their
tasks online and in hardcopy - Training material emphasizes reliability by
rewarding performance to plan - Afterwards future participants undergo a 2-step
introduction into the study -
- 1st 4 weeks handling of dummy test letters
- Monitoring of data entry, posting to plan,
data recency -
- 2nd after successful training phase panellists
handle transponder samples - Respondents are contacted and receive additional
teaching information in case of shortcomings
26Panellist Control
- Continuous review of each panellists performance
by automated checks for ... - Individual response rates such as availability
percentages - Data recency (sending, receiving)
- Posting-to-plan performance
- Transponder loss rate
- Information on delayed response/on-time
- Individual reliability indicators
- recency in responding to mails and entering data,
- rate of unannounced absences,
- clusters in reception dates etc.
Panellists receive a 3-step treatment
27Quality management of panel
- The three step concept to ensure the quality of
the participants - Motivation
- Proven incentive-system In every project the
participants collect points which they can later
exchange for attractive gifts. - Intensive support Regular contact from the panel
management as well as helpdesk email address with
reply times within one day. - Training
- Training of the participants regarding special
products, assessment of the postal services and
the portal used for the logging of information. - Training on quality aspects Why it is important
to always deliver exact and dependable
information. - Controlling
- Controlling of the actual information regarding
regularity, completeness and timing. - Participants, that repeatedly sent useless
information or no information at all, are
replaced straight away.
28Motivation bonus page
29Motivation bonus page - selection
30- Expected risks and technical options
31Survey design
- Simple design is easier to handle than a complex
system - Simple statistical design and geographical
coverage (i.e. mainly city links between capitals
of major business cities) - Simple network and panel requirements (i.e. only
private panellists) - Simple induction criteria (i.e. all induction in
mailboxes or at Post Offices) - Simple mail characteristics (i.e. stamped
letters, machine addressed 1 size, 1 weight, 1
format) - Location of the production unit
- Centralized or local
- Possibilities of production at the sender
- Transponder management
- Start of transponder integration in the test
letters - Archiving of data
- Necessity of sending back test letters and
envelopes
32Thank you for your attention!
Contact Klaus Riedel 49 40 800 96 257
klaus.riedel_at_ipsos.de Ipsos
GmbH Heidenkampsweg 100 20097 Hamburg