You are here:

Writing your thesis at DVJ Insights

DVJ Insights
Graduation Internship
Required language
Dutch, English
Commences at
06 April 2024
Finishes at
06 April 2025



We provide you with the opportunity to write your thesis with our data. You have the opportunity
to work at the DVJ office in Utrecht either for some days a week or also for a whole week and ‘work
along’. There is the option to work from home, which is the default if you only write your thesis
based on our data. In case of working along, you would work partly on your thesis and partly on
other projects, unrelated to your thesis. You will also follow different courses from our onboarding
program for new researchers that teaches you among others how we work at DVJ Insights, how to
develop a questionnaire, and how to program the survey in our survey software.
The internships allowance is €400 per month when you work 40 hours a week.


DVJ Insights is an ambitious, innovative and fast-growing marketing research agency located in the
Netherlands, the UK, Germany and Sweden with the ambition to be the best global agency for brand
growth. We add value by helping our clients to better understand consumer behaviour, improve
their brand positioning, realise more successful product introductions, and increase the effectiveness
of their media campaigns. At DVJ Insights, almost 100 passionate colleagues from 5 different
locations work for clients such as Philips, Samsung, Beiersdorf, Heineken, Cloetta, Domino’s, Nestlé,
ING, Rituals, and the Dutch Central Government. We are a dynamic and informal place to work with
people from different cultures and backgrounds.


The opportunity to write your thesis at DVJ and be able to work with our data. We are conducting
internal meta-analyses on a regular basis but want to learn even more from all the data we gather.
By conducting research on our databases, we can gather a lot of general learnings on why some
innovations are successful, how advertising becomes more effective, why brands grow and why they
don’t, and the way media deployment can be optimised.

Therefore, you get the opportunity to work with data that is interesting, very rich and practically
relevant. Furthermore, you will be able to use sophisticated econometric modelling to find
interesting insights, both from an academic as well as a practical perspective.
For a thesis topic, we can develop a topic together, and you would be able to use our panel for data
collection. Or, since we conduct so many different studies and we have a lot of data, you have the
option to work on one of the following topics:

1. Shopper missions (Marketing Analytics topic). For our clients we sometimes conduct
customer journey or shopper mission studies. In those studies we examine what needs are
triggers to go to the store and buy. Based on those studies we can identify different shopper
missions, such as the shopper on the go, or doing the large groceries. Additionally, we have
purchase data from Danish supermarkets available at the individual level for about two
years, which was registered via an app. Thus, it includes all purchase receipts including
product information for different supermarkets in different regions. There is also information
on the users, namely age and region where they live.
This enables you to examine whether the shopper missions as identified, can also be
distinguished in the data. Combining this again with background information of the users,
leads to rich insights on shopper missions and consumers. More specifically, one can think of
the following research questions: Are certain categories bought more/less often for certain
shopper missions? Are certain types of products (i.e., A-brands, private label, different pack
sizes) more/less popular for certain shopper missions? Are certain consumers more/less
price sensitive for certain shopper missions? Do we observe brand loyalty versus variety
seeking? Etc. etc.

2. Packaging (Marketing Analytics & Management topic). In today’s market, brands whose
products are sold in retail face several challenges. As shelf space increasingly comes at a
premium, each product will have to compete with many others for the consumer’s attention.
And, especially for fast-moving consumer goods, where consumer involvement is relatively
low and decision-making is primarily based on heuristic cues, it becomes vital that, within a
fraction of seconds, a product’s package is able to capture the consumer’s attention and is
identified as belonging to the brand (so that the product’s main features will also be linked to
the brand). Therefore, DVJ’s vision is that a proper pack test should not only force consumers
to pay attention to a package in evaluating it, but should also measure whether the package
is able to immediately attract (positive) attention and link its key features to the brand. This
database contains evaluation, associations and attention metrics for approximately 150
packs from about 10 countries and different categories. A few example research questions
can be: what determines the success of packaging? Which elements are crucial for standing
out on the shelf and what determines brand recall?

3. AI based eye-tracking validation (Marketing Analytics & Management topic). Recently we
have been partnering up with a party that is able to predict with an AI-based tool, where
people will look at an ad. They trained their AI tool on real eye-tracking data. We would liketo know more about how the AI eye-tracking metrics relate to other elements from our pre-
test. One can think of the following questions (but do not feel limited to those): How doesthe online eye-tracking correspond to our own mouse-tracking? How does the online eye-
tracking relate to evaluation of an ad, brand recall, and engagement? What are elements in ads that generate much attention? And which less?

4. Cross-country differences in advertising response (Marketing Analytics & Management
Many brands nowadays operate at least at a somewhat global scale, i.e. they sell their
products, and advertise them, across multiple countries. However, we still quite often
observe the exact same advertisements being used across different countries, either because
of the desire to save costs, or because one assumes that the advertisement will be perceived
in a similar way across different markets. However, is this actually the case? Do typical "style
elements" in ads invoke the same response across countries, or are there profound
differences? And if so, can these differences be conceptually linked to the attributes of these
countries (e.g. their national culture, socio-economic position, et cetera)? This project would
require the coding of different ads in terms of several "style elements" they may or may not
be using. We have different datasets available with different types of ads, namely for TV,
TikTok, Online video, or OOH. We also have one specific dataset containing only Christmas TV

5. Valuable open ended feedback (Marketing Analytics & Management topic). In marketing
research, we aim to collect respondents opinions and attitudes around a plethora of topics.
Much market research relies only on quantitative measures such as pre-defined statements
and questions. However, the danger is that important things are missed. And therefore, it is
important to start with letting respondents share their stories and associations around a
certain topic, without limiting them to pre-defined statements. However, the added value of
this approach all lies within the richness of the open answers. We already did some internal
tests to see how we can enhance richness of the open answers, for example by using a social
nudge and through AI-SmartProbing technology (using AI to ask follow-up questions). What
are more ways to ensure qualitatively rich open answers in online survey research? Are there any new ways known in academic research and if so, how effective are these? And related to
this: what are cost and time efficient ways to filter out AI-generated answers?


A master student (M/F/O) with a marketing (research) or psychology/consumer behaviour
Do you recognise yourself in these points? We are looking for someone who:
• Is studying econometrics, data science, computer science, marketing management,
marketing research, psychology, (technical) business studies, or anything related
• Is interested in consumer research and the why behind the what
• Loves (marketing) data and modelling
• Knows how to handle big data from multiple sources
• Can identify interesting research topics based on (patterns in) the available data
• Has excellent knowledge of different statistical software packages such as SPSS or R
• Has excellent knowledge of MS Office
• Is fluent in English
• Is willing to go the extra mile
• Has high quality standards

Interested? Apply now at