Reflecting on Process, Voice and Leadership in the State Germany

Go-to-Market Project

Introduction

This reflective essay discusses my experience of participating in the State collaborative project.
The project requires the student team to formulate an international marketing promotion plan.
The project introduction positions State as a platform, which uses artificial intelligence-guided
interfaces to classify citizen issues and direct the problems to the corresponding elected
representatives. Our group chose the German market and responded to the challenges through
market analysis, blocker and risk analysis, stakeholder analysis, marketing strategy and business
model. I was mainly responsible for developing the business model canvas and customer journey
map, and also participated in the construction of SWOT analysis and value proposition.

I used the Gibbs Reflective Cycle because it required a transformation from description to
feelings, evaluation, analysis and action plan (Gibbs, 1988). I also drew on Schön's theory
(1983), because the most important learning experience happened when I had to decide whether
to speak, compromise or question the team process. The problem is that our group had not
established a legitimate and effective decision-making process to make choices. The team leader
initially assigned the task, but later made and presented another presentation without a clear
consensus. Some of us thought that this presentation was too general and lacked novelty; but at
the same time, it was more strategically comprehensive and visually more coherent than the
distributed version. Therefore, I avoided blaming the problem on someone's behaviour. Instead, I
reflected on how my silence, the unclear team process, and the contradiction between efficiency
and responsibility eventually led to such a result.

Project context and my role

At first, the work assignment seemed reasonable. Different members were assigned different
parts. This seemed to be in line with Belbin's view that efficient teams need complementary roles
rather than exactly the same contribution (Belbin, 2010). I was responsible for turning the
questions about trust and monetisation in the brief into a business model canvas. I suggested that
State should protect citizens' voice and generate revenue through municipal information
platforms, research subscriptions, media insight products and low-risk data cooperation. I also
drew a customer journey map from problem submission, artificial intelligence classification and
representative allocation to public progress tracking and institutional learning.

However, task assignment is not equivalent to teamwork. Katzenbach and Smith (1993) believe
that a real team needs common goals, performance goals, common methods and mutual
accountability. We have a common task, but we don't have a common approach. We have not
clearly agreed on the final ownership of editing rights, the method of evidence verification, the
method of combining slides, and whether to accept artificial intelligence-assisted writing.
Tuckman's model (1965) helped me understand this. We quickly transitioned from the "forming"
stage to the task generation stage, but we did not spend enough time on "norming". As a Chinese
student in the UK, my initial understanding of cooperation is that it is reliable, respectful of
others, and does not create unnecessary trouble. I expected leaders to be able to coordinate the
contributions of others instead of replacing others. But this expectation has never been clearly
put forward. Therefore, the first lesson makes me uneasy: I took fairness for granted, but in
group cooperation, fairness must be integrated into the design of the process.

Critical incident

The first critical event occurred after the team leader assigned the task, and a new integrated
presentation was created, which did not use the original materials of all members. I had invested
a lot of time in developing the business model canvas (BMC) and customer journey map, which I
understood thoroughly and could explain confidently. When another version emerged, I felt that
my contribution had been overlooked. This feeling was not only about the score, but also about
the value. I began to think about whether my ideas, words and efforts were regarded as an
indispensable part of the team's achievements.

Meanwhile, I was unsure whether to directly confront the team leader. This was partly due to my
personality and partly because of my status as an international student. In English group
assignments, I sometimes need more time to clearly express my different opinions. Research on
Chinese students shows that silence should not be regarded as a lack of participation; silence
may be a strategy to maintain respect, pragmatism and seek convincing views (Guo, 2024). This
helps me understand my own behaviour, but it does not justify it. My silence maintained short-
term harmony, but left process problems unresolved. Morrison and Milliken (2000) describe that
people may choose to hide their concerns when they think that speaking is risky or in vain.

The leader's behaviour also needs to be evaluated objectively. On the positive side, the integrated
presentation builds a more powerful strategic narrative. It covers PESTLE analysis, competitive
benchmark analysis, stakeholder mapping, Ansoff model, blue ocean strategy, phased
implementation, and trust/monetisation framework. On the negative side, this process
undermines psychological safety. Edmondson (1999) defines psychological safety as a shared
belief among team members that the team can safely take interpersonal risks. When members
believe that the results of their work can be replaced without discussion, they may stop proposing
ideas. In our case, although the team eventually completed a project, it now seems less and less
like the result of a team effort.

My response was not entirely flawless either. I attempted to restore the original materials of the
team because I believed that my own content could better reflect the true contributions of the
team and be more persuasive. However, at that time, I took the approach of unilateral revision to
counter unilateral control. Now it seems that a better solution would be to hold a brief decision-
making meeting, allowing team members to compare the two versions based on the established
criteria.

Comparing the two slide versions

The existence of the two versions not only creates conflicts, but also constitutes a useful
diagnostic tool. The distributed version containing the ninth part written by me and SWOT
analysis/value proposition has many advantages. It is closer to the actual work done by each
member, so each speaker can explain their own logic. In addition, it also conveys a clear
message of "trust first": State should not operate like a protest application or monitoring
platform, but should play the role of a privacy-safe routing and accountability layer. For
example, OECD data shows that only 36% of Germans express high or moderate trust in the
federal government (OECD, 2024), while the German Bundestag reported that a total of 9,260
petitions were received in 2024 (Deutscher Bundestag, 2025). The guidance of the German Data
Protection Conference (DSK) on artificial intelligence and data protection also confirms our
concern that interpretability, legal basis and manual supervision are crucial in the field of
artificial intelligence for German citizens (DSK, 2024).
Its shortcoming is that it may be too one-sided. My business model canvas (BMC) and customer
journey are rigorous in logic, but they do not automatically solve the broader problem of how to
prioritise entry into the German market. The presentation needed clearer geographical division
and more reasonable order. In this sense, the leader's version adds value by putting forward a
more ambitious "Hamburg-first" argument and a phased 24-month roadmap.

The leader-led version is more like a strategic consulting presentation. It uses recognised tools,
such as Ansoff's market development logic (Ansoff, 1957), Osterwalder and Pigneur's Business
Model Canvas (Osterwalder and Pigneur, 2010) and Blue Ocean Strategy (Kim and Mauborgne,
2005). It also more clearly clarifies the trade-off between trust and monetisation by positioning
the source of income in low-intimacy, aggregated insight products, which coincides with
Zuboff's criticism of the surveillance-based business model (Zuboff, 2019).

However, there are also risks in the leader-led version. Some assertions seem to be too rigid and
lack reliable evidence. Although a large number of references seem convincing, they are not
necessarily reliable. For core political assertions, official data, OECD reports, parliamentary
publications and peer-reviewed research are more convincing than Wikipedia or extensive online
resources. The best solution is to combine the two versions: not only to retain the strategic
structure of the integrated presentation, but also to take into account the authenticity, chapter
ownership and evidence prudence of the decentralised presentation. Our problem is not how to
choose between quality and fairness. Our problem is that we have failed to design a process that
can promote quality and fairness.

Presentation incident and team dynamics

The second key event occurred when the leader used his own version of the presentation without
explicitly notifying the team. Presentations require rehearsal, transition and team trust. If the
final version was changed without communication, other members would not be able to be fully
prepared, and the audience may interpret hesitation as lack of ability, not lack of information.

This incident transformed the conflict from a task conflict to a relationship conflict. The meta-
analysis of De Dreu and Weingart (2003) shows that relationship conflicts often damage team
performance and satisfaction. In our case, if the initial disagreement can be focussed on the task
itself, it may be more productive: which slide is more accurate? Which version is more in line
with the project requirements? Which structure is more favourable to the audience? However,
due to the lack of transparency in decision-making, the problem evolved into a personalised
conflict: whose work is more important? Who has the authority? Can we still trust the leader?

I also found a contradiction between our project theme and the team process. At that time, we
were providing advice to State on how to build a democratic infrastructure, representative
accountability and trust. However, within our own team, the right to speak and accountability are
not fully guaranteed. This made me realise the problem facing State: when people can't see the
decision-making process or how their opinions affect the result, they will lose trust. The lesson I
learned from it is that professional differences must be expressed before the deadline, not after
causing damage. I should have written a clear and calm email before the presentation: "Can we
confirm which version is the final version, and whether the people responsible for each section
can approve the modification of their respective parts by tonight?" In this way, the problem can
be transformed into a process problem.

Evaluation of our proposed solution

The low trust data of the OECD (OECD, 2024), the continued use of Bundestag petitions
(Deutscher Bundestag, 2025) and the sensitivity of political data under GDPR all support a
cautious, trust-first entry strategy. State should enter the market as citizen infrastructure: as a
neutral platform, it is responsible for conveying issues, building representative signals and
providing aggregated insights into the public interest.

The most powerful part of my contribution is business logic. According to the project brief
(Loughborough University London, 2026), advertising has been excluded, but this is not a
weakness. This compelled us to design a model that relies on trust rather than attention as the
source of income. My business model proposal treats citizens as free users and institutions as
paying customers. They purchase aggregated dashboards, subscription services, and reports. This
maintains the ethical boundary that political behavior of individuals should not be treated as a
commodity. From the perspective of data protection, this also reduces risks. Because according
to Article 9 of the General Data Protection Regulation (GDPR), political views fall into a special
category of data, and high-risk civic artificial intelligence may need to undergo a data protection
impact assessment in accordance with Article 35 (European Parliament and the Council of the
European Union, 2016).
The weak point in our plan lies in the verification process. We proposed a seemingly reasonable
strategic plan, but we did not always clarify how State could conduct tests at a low cost and in a
safe manner. The lean startup method proposed by Ries (2011) helped me realize that the launch
of the minimum viable solution should first verify the assumption with the highest risk. For State
in Germany, the assumption with the highest risk not only lies in the accuracy of the technical
routing, but also in trust, neutrality, local data quality, the willingness of civic institutions to
cooperate, and the public's acceptance of artificial intelligence assisting citizens' voices.
Therefore, a more complete final plan should start with one city, two thematic areas, clear
transparency rules, manual review, a published data protection plan, and a small number of
institutional partners.

Personal learning and action plan

I thought that being a good team member means completing one's own tasks, being polite, and
avoiding unnecessary conflicts. Now I realise that this is just a part of professionalism.
Responsible members will also question the process, maintain common ownership, and make
differences meaningful.

First of all, in future projects, I will propose a team charter at the first meeting. If someone is the
editor, I would suggest adopting the principle of "the editor is responsible for integration, and the
person in charge of the chapter is responsible for the review".

Secondly, I will distinguish between task conflicts and interpersonal conflicts. If I have an
objection to a slide, I will focus on evidence and the brief: "This argument needs a stronger
source" or "This part is very convincing, but I can only express it confidently if I understand the
logic in it". I will avoid waiting for dissatisfaction to turn into a personal attack before taking
action. As an international student, this is especially important to me. I want to respect others,
but respect should not mean keeping silent when project risks arise.

Third, I will improve my speaking strategy. Since I sometimes need time to express my
disagreement in English, I will prepare a short written script before the meeting. For example: "I
agree with the necessity of maintaining consistency, but I am afraid that I will lose control of a
certain part. Can we compare the two versions and agree on a final version?"
Fourth, I will regard citation management as teamwork. Personal analysis does not need to be
cited, but factual arguments, frameworks and data need to be cited. Finally, I will keep in mind
the deeper contradictions in this project. We think that State should help citizens speak out, but I
don't always make sure that my voice is heard. My Chinese background has taught me to value
harmony, patience and respect. My study experience in the UK made me realise that professional
respect also requires open participation, questioning and evidence-based objections. My future
goal is not to raise my voice, but to improve the clarity and timing of my expression.

Conclusion

The main lesson I learned from it is that high-quality collaboration requires three elements at the
same time: strategic consistency, evidence rigour and procedural legitimacy. The two versions of
the slide deck have their own advantages, but our team lacks a mature process to integrate these
advantages. In future projects, I will participate in the construction of processes as soon as
possible, speak with evidence, and ensure that the final results are not only excellent, but also
truly shared.

References

Ansoff, H.I. (1957) 'Strategies for diversification', Harvard Business Review, 35(5), pp. 113-124.
Available at: [Link]
10001836084 (Accessed: 8 June 2026).
Belbin, R.M. (2010) Team Roles at Work. 2nd edn. Oxford: Butterworth-Heinemann. Available
at: [Link]
meredith-belbin (Accessed: 8 June 2026).
De Dreu, C.K.W. and Weingart, L.R. (2003) 'Task versus relationship conflict, team
performance, and team member satisfaction: a meta-analysis', Journal of Applied
Psychology, 88(4), pp. 741-749. doi: [Link]
Deutscher Bundestag (2025) '9.260 Petitionen erreichten den Ausschuss im Jahr 2024'. Available
at: [Link] (Accessed: 8 June 2026).
Edmondson, A. (1999) 'Psychological safety and learning behavior in work teams',
Administrative Science Quarterly, 44(2), pp. 350-383. doi: [Link]
European Parliament and Council of the European Union (2016) Regulation (EU) 2016/679:
General Data Protection Regulation. Available at:
[Link] (Accessed: 8 June 2026).
Gibbs, G. (1988) Learning by Doing: A Guide to Teaching and Learning Methods. Oxford:
Oxford Polytechnic Further Education Unit. Available at:
 [Link]
[Link] (Accessed: 8 June 2026).
Guo, Y. (2024) 'Strategy for negotiating engagement: Chinese international students' silence in
multicultural group work at a UK university', SRHE Annual Research Conference 2024.
Available at: [Link] (Accessed: 8 June 2026).
Katzenbach, J.R. and Smith, D.K. (1993) 'The discipline of teams', Harvard Business Review,
71(2), pp. 111-120. Available at: [Link]
(Accessed: 8 June 2026).
Kim, W.C. and Mauborgne, R. (2005) Blue Ocean Strategy: How to Create Uncontested Market
Space and Make the Competition Irrelevant. Boston: Harvard Business School Press.
Available at: [Link]
(Accessed: 8 June 2026).
Konferenz der unabhängigen Datenschutzaufsichtsbehörden des Bundes und der Länder (DSK)
(2024) Künstliche Intelligenz und Datenschutz: Orientierungshilfe. Available at:
[Link]
KI_und_Datenschutz.pdf (Accessed: 8 June 2026).
Loughborough University London (2026) The Collaborative Project - Brief: State - building the
infrastructure layer for democracy. Unpublished project brief. State website available at:
[Link] (Accessed: 8 June 2026).
Morrison, E.W. and Milliken, F.J. (2000) 'Organizational silence: a barrier to change and
development in a pluralistic world', Academy of Management Review, 25(4), pp. 706-725.
doi: [Link]
OECD (2024) OECD Survey on Drivers of Trust in Public Institutions 2024 Results: Country
Notes - Germany. Available at: [Link]
on-drivers-of-trust-in-public-institutions-2024-results-country-notes_33192204/
germany_1b23ffcd.html (Accessed: 8 June 2026).
Osterwalder, A. and Pigneur, Y. (2010) Business Model Generation: A Handbook for
Visionaries, Game Changers, and Challengers. Hoboken: Wiley. Available at:
[Link]
%2Bfor%2BVisionaries%2C%2BGame%2BChangers%2C%2Band%2BChallengers-p-
9780470876411 (Accessed: 8 June 2026).
Ries, E. (2011) The Lean Startup: How Constant Innovation Creates Radically Successful
Businesses. London: Portfolio Penguin. Available at: [Link]
(Accessed: 8 June 2026).
Schön, D.A. (1983) The Reflective Practitioner: How Professionals Think in Action. New York:
Basic Books. Available at:
[Link]
donald-schon (Accessed: 8 June 2026).
Tuckman, B.W. (1965) 'Developmental sequence in small groups', Psychological Bulletin, 63(6),
pp. 384-399. doi: [Link]
Zuboff, S. (2019) The Age of Surveillance Capitalism: The Fight for a Human Future at the New
Frontier of Power. London: Profile Books. Available at: [Link]
age-of-surveillance-capitalism/ (Accessed: 8 June 2026).