Professional Documents
Culture Documents
Trame - Thèse MS
Trame - Thèse MS
Trame - Thèse MS
PROFESSIONAL THESIS
For the Advanced Master graduation
1
Titre Thèse professionnelle – Auteur - Date 2
AKNOWLEDGEMENT
Also, I would like to give a special thanks to my classmates for their contribution into making
our master a rich journey, full of experience, knowledge and fun sharing.
I would be remiss in not mentioning my family, especially my mom and brothers. Their
unconditional and unwavering support have been crucial to overcoming every challenge.
Finally, I could not have undertaken this journey without the help of the program Director and
my Thesis Director Mrs. Anne-Flore MAMAN, that was always available to guide me
through my interrogations throughout the whole program.
I would like to mention the name of some people who had a deep impact on me during this
journey:
I attest that this thesis does not incorporate without acknowledgement any material previously
submitted to any institution. To my knowledge does not contain any material previously
published or written by another except where due reference is made in the document.
Ali MASSOUDI
AKNOWLEDGEMENT.............................................................................................................3
DECLARATION OF ORIGINALITY.......................................................................................4
TABLE OF CONTENTS............................................................................................................5
Table des figures, tableaux et images.........................................................................................6
INTRODUCTION......................................................................................................................7
Partie 1.............................................................................................Erreur ! Signet non défini.
Chapitre 1.....................................................................................Erreur ! Signet non défini.
Chapitre 1.1..............................................................................Erreur ! Signet non défini.
Chapitre 1.2..............................................................................Erreur ! Signet non défini.
Chapitre 1.3..............................................................................Erreur ! Signet non défini.
Chapitre 2.....................................................................................Erreur ! Signet non défini.
Chapitre 3.....................................................................................Erreur ! Signet non défini.
Partie 2......................................................................................................................................10
Chapitre 1.....................................................................................Erreur ! Signet non défini.
Chapitre 1.1..............................................................................Erreur ! Signet non défini.
Chapitre 1.2..............................................................................Erreur ! Signet non défini.
Chapitre 1.3..............................................................................Erreur ! Signet non défini.
Chapitre 2.....................................................................................Erreur ! Signet non défini.
Partie 3......................................................................................................................................11
Chapitre 1..............................................................................................................................30
Chapitre 1.1.......................................................................................................................30
Chapitre 1.2.......................................................................................................................30
Chapitre 1.3.......................................................................................................................30
Chapitre 2..............................................................................................................................31
Conclusion................................................................................................................................32
Bibliographie.............................................................................................................................33
Annexes.....................................................................................................................................34
Le résumé managérial ou « Executive Summary.....................................................................35
Management consulting usually addresses C-suite level projects for international firms or
large public service institutions. They have therefore, by their very nature, contributed to
shaping not only the business outlook for several industries, but also various governmental
policies with a significant impact worldwide since the establishment of the first consulting
firms by the end of the 19 th century. The market size for consultancy services has reached an
impressive 860.13 billion dollars in 2023 with an annual growth of 4.1% over the past ten
years (Sources: Consultancy.org, IBIS World). This increasing demand for management
consulting services suggests a strong value proposition that managed to stay attractive
throughout the years and thrive despite several pessimistic warnings expressed by various
academics. Christensen et al. (2013), Friedrich von den Eichen and al (2019) Llewellyn,
(2017) and Nissen (2018) have all expressed the risks faced by traditional consulting business
models in an environment where digital disruption, amongst other disruptors, is expected to
change the competitive landscape.
Digital and technological innovations come with their fair share of “hype”. For instance, the
year 2021 has seen the rise of the Metaverse vision which led to various speculations by key
players in various industries. However, a year later the metaverse vision is facing considerable
doubt and a whole new technology is at the center of everybody’s attention: Generative AI.
OpenAI has offered free access to its AI language model CHATGPT-3, which has caused
both excitement and apprehension regarding its potential impact on the business landscape
and specifically on information-based industries. Reactions were strong and immediate in the
public debate and amongst tech giants: Alphabet (Google) has, according to Forbes, declared
a code red over the threat of OpenAI and Elon Musk along with other experts published a
statement urging a halt to AI research for a period of 6 months. Half a year later, the debate is
still going strong, and all major consulting firms have released multiple articles and thought
leadership surrounding the potential and impact of generative AI on various industries as well
as the job market. As an information and expertise-based industry (Insert reference DD),
management consulting is likely to undergo significant changes to adapt to the disruptive
potential of this technology. This change can be inferred through several major steps taken by
major consultancies. BCG and Bain both announced their partnerships with OpenAI, with
Bain x OpenAI launching their first AI powered marketing campaign coined “Create your
In this exploratory study, we aim to understand how the traditional management consulting
business model is going to cope with the rise of generative AI, and what actions it should
address to thrive in a potentially disrupted environment. In order to understand how this
technology is likely to affect management consulting, it is essential to investigate
characteristics of this industry which allowed major players to thrive despite a century of
disruption and look more closely into the attributes of generative AI and its potential
disruptive power for the management consultancy industry.
1) MANAGEMENT CONSULTING
Due to its heterogeneity and lack of regulation, the consulting industry has received little
academic attention (Srinivasan, 2014). Its permeability and ever shifting nature also explains
the reason why it is difficult to find a consensual definition of this discipline. Nonetheless, in
this paragraph, we are going to identify converging defining traits of the management
consulting industry in academic literature.
Greiner and Metzger (1983) stipulate that management consulting firms are contracted
(external to the organization) to provide clients with advisory services that span problem
identification, analysis of potential solutions and implementation. These advisory services are
meant to yield organizational improvement through their cutting-edge experience and/or
expertise of their consultants.
Other definitions along these lines are focus on the management consultant rather than their
employer, Fincham and Clark (2002) stated that consultants are perceived as “professional
helpers remedying the illness in client organi(z)ations”. In this sense, consultants provide their
client organizations with value, through contracted projects, by leveraging expertise that is not
easily available within their structure in order to help them respond more efficiently to their
business environment (Momani, 2013). Consequently, in the scholar definitions (Greiner and
Metzger, 1983; Fincham, Mohe, and Seidl, 2013), it is possible to identify recurring traits that
characterize management consulting services:
(a) External to the organization.
(b) Time-bound: as their service is aimed at addressing a specific need limited in time.
(c) Information & knowledge based: the value of a consultant to the client lies within their
experience and expertise, which can be transferred to the client’s business.
(d) Compensation: consultants charge fees for their professional services
In the upcoming section, we are going to discuss the origins and evolution of management
consulting as a discipline in order to better understand how today’s industry was shaped.
This led to management consulting firms to shift their service areas from generalist to
functional (Christensen et al., 2013)
Globalization was yet another substantial challenge for the industry incumbents as one of their
main differentiators was their deep local knowledge (Von Nordenflycht, 2010). This has
contributed to their shift from a local to a global network of consultants and experts
(Christensen et al., 2013).
In the highly competitive landscape of the management consulting industry, companies face
the imperative of strategically positioning themselves close to their industry rivals. This
strategic maneuvering offers a range of significant benefits, including the amplification of
their perceived legitimacy, the reduction of uncertainties, and the heightened potential for
knowledge spillovers within the competitive landscape. However, this deliberate alignment
with competitors inevitably presents challenges, as it limits the firms' ability to differentiate
themselves distinctly from their closest rivals, thereby intensifying the direct competition
among them (Srinivasan, 2014).
These trends have naturally led to a segmentation of the consultancy industry by types of
firms. Although key service area segmentation in literature may slightly vary depending on
the source,
An article on management consulting from (XERFI, 2022) highlights that firms are positioned
across one or multiple service areas, they distinguish several types of firms:
- Strategy dominant firms: McKinsey, Bain & BCG
- Generalists: Kea & Partners, Sia Partners, Wavestone
- IT & Management consulting firms: Accenture, Capgemini Invent, DXC
Although the previous categorization is useful to analyze the industry and types of services
offered, it remains somewhat artificial and analytical when compared to the reality of the
industry. Due to the unregulated and evolutive nature of the market, there is significant
overlap amongst the consulting industry actors when it comes to their service areas and
typology of engagement. A quote from Christensen et al. (2013) illustrates this limitation:
At traditional strategy-consulting firms, the share of work that is classic strategy is
now about 20%—down from 60% to 70% some 30 years ago.
– HBR, Christensen et al., Oct 2013
Arthur N.Turner (HBR Review, 1982) in his hierarchy of consulting purposes, had seemingly
identified the trend for overlaps early on. The author insists on the importance of
implementation support, which was not necessarily addressed at the time by top strategy firms
and which has become much more common nowadays.
Understanding the role of a consultant is important to grasp the motivation driving consulting
clients as well as the expected value from these services. There have been various references
to the role of a management consultant in academic literature. The review of a sizeable
portion of these sources allows us to identify converging roles that can be grouped in major
categories by decreasing order from most to least explicit/tangible.
CATEGO
RY ROLE DESCRIPTION IN LITERATURE REFERENCE
Consultants provide expertise and Schein (1988),
Expertise and problem solving
The “political influence and engagement” value of the consultant are contingent to their more
primary roles of “problem solvers” which ensures the required level of legitimacy.
Human Capital: is the primary resource for a management consulting firm. Although
consulting is one of the most coveted career options for business school graduates for a
variety of reasons (Kakabadse et al., 2006) such as an attractive pay, a steep learning curve,
social prestige and attractive exit opportunities, recruiting and retaining qualified talent is a
challenge due to the extremely selective nature of the consulting hiring process and the high
turnover rates which can be explained by the intense work culture. In their last report on the
industry, XERFI (2022) highlighted talent attraction and retention as one of the most crucial
challenges for the year for the management consulting industry in France.
Brand value and reputation: As Canbäck (1999) and Nordenflycht (2010) state, the
market value of many companies is closely linked to intangible elements such as brand image
and “social proof” such as educational background and posture. These factors serve as
substitutes for tangible outcomes, providing incumbents with an advantage. Brand and
reputation are fundamental factors in the management consulting industry, serving as vital
assets. In the face of growing competition, consulting firms must distinguish themselves to
attract clients. A robust brand and reputation act as compelling indicators of a firm's expertise,
credibility, and reliability (Amonini et al., 2010). Clients are more inclined to engage with
consulting firms that possess an established and respected brand, as it assures them of quality
and value. Furthermore, a positive brand image can foster client loyalty and referrals,
bolstering the firm's market position (Furusten, 2013). Moreover, a strong brand enables
consulting firms to command premium fees and draw in top-tier talent, thereby fostering
sustainable business growth. Hence, it is imperative for management consulting firms to make
strategic investments in constructing and nurturing a reputable brand to thrive in the ever-
evolving marketplace.
2) GENERATIVE AI
a. WHAT IS AI?
AI has transformed healthcare, finance, and entertainment (Topol, 2019). It aids in medical
diagnosis, recommendation systems, and autonomous vehicles (Topol, 2019). Ethical
challenges like bias, fairness, and interpretability must be addressed for responsible AI
(Russell & Norvig, 2016). In conclusion, modern AI offers tremendous opportunities for
innovation and societal impact (Russell & Norvig, 2016).
AI can be traced back to the mid-20th century when Alan Turing proposed the Turing Test, a
machine exhibiting intelligent behavior (Turing, 1950). The Dartmouth Workshop in 1956
marked the formal beginning of AI research, with notable participants like John McCarthy,
Marvin Minsky, and Nathaniel Rochester (McCarthy et al., 1956).
During the 1950s and 1960s, researchers pursued symbolic AI, creating programs for
reasoning and problem-solving using symbolic representations (Newell & Simon, 1958).
Progress in symbolic AI included computer programs playing games like chess and checkers
(Shannon, 1950).
Expert systems emerged in the 1970s and 1980s, using knowledge-based approaches to solve
complex problems in specific domains (Feigenbaum & McCorduck, 1983). These systems
aimed to capture human expertise for intelligent decision-making.
These advancements collectively laid the foundation for AI-powered systems and applications
that are prevalent today.
Machine learning, a branch of artificial intelligence (AI), enables computers to learn from
data and make predictions without explicit programming (Mitchell, 1997). It has made
significant advancements in various domains, such as natural language processing and
computer vision (Goodfellow et al., 2016). Deep learning, a subset of machine learning,
utilizes neural networks with multiple layers to automatically learn complex features from
raw data (LeCun et al., 2015). Deep learning tends to require larger volumes of data, but also
less human intervention.
In their article “Getting to know-and manage-your biggest AI risks”, Kevin Buehler et al.
(2021) identified and categorized 6 types of risks associated with the use of AI in
organizations:
Privacy risk: Privacy risks associated with the utilization of artificial intelligence (AI)
have become a significant concern in contemporary society (Smith, 2022). As AI systems
increasingly handle vast amounts of personal data, there is a pressing need to safeguard
individuals' privacy rights and prevent potential breaches. These risks encompass various
aspects, including data collection, storage, sharing, and processing. Improper handling of
sensitive information can lead to unauthorized access, identity theft, or discriminatory
profiling. To address privacy risks, researchers have proposed numerous approaches such as
privacy-preserving AI techniques, encryption methods, and data anonymization. Additionally,
legal frameworks like the European Union's General Data Protection Regulation (GDPR)
(EU, 2016) provide a foundation for protecting individuals' privacy. By incorporating
privacy-enhancing mechanisms and adhering to robust regulations, AI developers and
organizations can ensure the responsible and ethical use of AI while upholding privacy rights.
Bias: Ensuring fairness and mitigating bias in AI models is a critical concern for
organizations deploying artificial intelligence. Biases can inadvertently permeate AI models,
either through the encoding process or by inheriting biases present in the training data
(Buolamwini & Gebru, 2018). Such biases can have detrimental effects on specific classes or
groups, leading to fairness risks and potential legal liabilities for companies. Detecting and
addressing biases in AI systems requires robust data preprocessing techniques, diverse and
representative training datasets (Caliskan, Bryson, & Narayanan, 2017), and ongoing
monitoring and evaluation. Employing techniques like debiasing algorithms and fairness
metrics can aid in reducing bias and promoting equitable outcomes. Organizations must also
prioritize transparency and accountability in their AI development processes to establish trust
and mitigate the potential impacts of bias. By proactively addressing fairness risks,
organizations can foster inclusive and responsible AI deployments.
According to BCG’s description, generative AI has emerged as a rapidly advancing field that
encompasses a diverse array of algorithms and techniques aimed at producing new authentic
content. These algorithms are designed to learn from extensive sets of training data and
generate outputs that demonstrate creativity and realism. At the core of generative AI are
foundation models, which serve as the backbone for these algorithms. These models undergo
As defined above, generative AI has the potential to create new unstructured content
following instructions. We have covered LLMs and their ability to generate text output using
text input. However, the neural networks that generate those texts have the potential to
generate other forms of unstructured context. McKinsey has compiled the current technology
panorama of generative AI in a matrix crossing editors with the type of generative AI (Text vs
Image vs Audio etc.)
Being a fairly recent advancement (November 2022), there is little consistency in publications
about the impact of IA on the business landscape. Most of the work in this literature review is
future oriented and exploratory. However, in the short lifespan of generative AI, major key
In this paragraph, we are going to structure and highlight relevant publications on this expect
impact.
Impact by industry: In its report (Global economics: The potentially large effects of
artificial intelligence, 2023), Goldman Sachs has published the following data on the
percentage of workforce exposed to generative AI automation by industry. It estimates that
28% of the workforce across the US and Europe is exposed to automation. With the industries
ranked in the following descending order:
What most stands out in this classification is the fact that the most impacted industries are
expected to be mostly white-collar industries with Legal and Business and Finance being in
the top 5 while more manual / blue collar work is least impacted with Transportation,
Production and Maintenance being in the bottom 5.
a. WHAT IS DISRUPTION ?
First coined by Christensen (1995), and more widely formalized by Christensen et al. (2015)
in HBR, the theory of disruptive innovation has gained significant recognition as a valuable
framework for understanding innovation-driven growth. At its core, disruptive innovation
refers to the process by which smaller companies challenge established incumbents by
targeting overlooked market segments with more suitable offerings at lower prices.
Incumbents often overlook these segments as they focus on their most profitable customers,
enabling entrants to gain a foothold and eventually disrupt the market when mainstream
customers embrace their offerings.
This theory evolved from being a descriptive framework to a more predictive one, in which
the incumbent is displaced to a high-end market by sustaining incremental improvements in
their services or products as the disruptor enters from-below in terms of product performance
serving a low-end market while improving product performance until they capture the
mainstream market.
It's a process: Disruption begins with small-scale experiments, focusing on refining the
business model rather than just the product. Over time, disrupters erode incumbents' market
share and profitability by targeting overlooked market segments.
Business models are often very different from incumbents: Disruptors often build
business models that are distinct from incumbents. They introduce new approaches, such as
standardized protocols in the healthcare industry, to diagnose and treat specific disorders.
Some succeed and some fail: Not all disruptive innovations succeed. Success is not
inherent in the definition of disruption, and not every triumphant newcomer follows a
disruptive path. Failures serve as boundaries for the theory's application, and winning in the
foothold market requires playing the odds and avoiding direct competition with better-
resourced incumbents.
While it allows for a more rigorous definition of disruption, the limitative nature of the theory
excludes many interesting use cases that offer a broader understanding of the notion.
To deepen the understanding behind this customer behavior, Christensen et al. (2015) seems
to make 2 major assumptions that have been challenged repeatedly in recent literature,
particularly for the B2C market. The first assumption is that the customer is completely
rational in their purchasing decision, they wouldn't buy a higher price product with
"superfluous" features, the second reason is that all the dimensions of a product/service
performance are tangible and measurable. With the iPhone example, one might argue that the
ease of use and brand value cannot be accounted for in Christensen's disruptive innovation
framework.
The gig-economy: The gig economy is rapidly expanding and has emerged as a
potential factor of disruption in the business landscape. According to Cambridge University,
the gig economy is characterized by temporary jobs and independent work arrangements
(Cambridge Dictionary, n.d.). The rise of digitized platforms has fueled the popularity of this
new way of working, allowing individuals to have greater control over when, where, and how
they work (Kässi and Lehdonvirta, 2018). Companies like Uber and Airbnb have become
synonymous with gig work, but the gig economy is not limited to these examples (Lau, 2020).
In fact, highly experienced professionals, including management consultants, are increasingly
being hired as gig workers to provide specialized expertise and advice (Lau, 2022). The gig
economy offers flexibility and freedom for workers, which has become highly valued in
today's business landscape (Tadros, 2021). COVID-19 further accelerated the importance of
remote work, making it possible for jobs to be performed regardless of physical location
(McKinsey, 2020). However, it is important to note that the gig economy also presents
challenges, such as limited customer base and potential restrictions imposed by on-demand
Academic literature provides several definitions and frameworks to describe what a business
model is. At its core, a business model covers three main areas structured around the notion of
value: Creating, proposing and capturing value (European Journal of Innovation Management,
March 2021). It serves as a structural framework that outlines how an organization operates,
taking on various interpretations. Some view it as a cognitive and linguistic schema,
resembling a narrative that describes how the firm functions (Magretta, 2002; Casadesus-
Masanell and Ricart, 2010). On the other hand, some consider Business models as formal
conceptual representations, explicitly documented in mathematical models or visual
representations (Teece, 2010; Massa et al., 2017). In this context, we can take interest in the
interpretation that views Business models as real attributes (Casadesus-Masanell and Ricart,
2010). This view comprises an array of resources and activities that are leveraged with or
without partners (Zott and Amit, 2010) to deliver value to the client. Therefore, the business
model concept comprises the solutions offered by the firm, the activities and processes
employed to deliver the promised value, and the underlying earning logic that covers costs
and generates profits (Bouncken and Aslam, 2019; Clauss, 2017).
The elements of this definition are fairly similar to the visual representation by (Osterwalder
& Pigneur, 2010) in their now famous business model canvas.
Otswalder &
Description
Pigneur, 2010
The unique service or product offered by the company and how it
Value Proposition delivers value. It encompasses performance, design, brand, price,
accessibility, and usability.
The essential tasks required to achieve and deliver the value
Key Activities proposition. They encompass manufacturing, sales, advertising,
VALUE
As one can see in the table compiled above, the business model canvas is transposable to the
European journal of management and innovation’s description of a value creation, value
delivery and value capture. In literature, all aspects of the business model are not always
treated equally in terms of depth. We will focus on the descriptions provided in the following
chapters of this thesis.
In management consulting, Christensen et al. (2013) in his HBR article has identified 3
traditional business models for management consulting and highlighted the main differences:
Solution Shop: Which aims to investigate and solve an issue with an undefined perimeter.
The main value driver is the consultant’s judgement process is not repeatable. This type of
business model aligns well with strategy dominant firms and the fees tend to be high.
Value-Added Process: The scope of the engagement is defined, and the intervention of
the consulting firm usually follows a repeatable process. This aligns well generalist type of
firms, and the fees are paid for a specific output.
Facilitated Network: which is essentially a network that enables product and service
transactions. Fees are usually paid to the network by the client.
Christensen et al. (2013) identifies a variety of trends indicating disruption in the consulting
industry, he argues that the management consulting business model has remained virtually
untouched ever since its creation at the turn of the 20 th century. Christensen suggests that the
opacity of the value creation process as well as the agility of major players in the industry are
the reasons why the industry has been spared by disruption. However he identified back in
2013 trends that are gaining momentum and affecting the consulting industry:
Putting these factors in perspective with the French market, we have identified 3 emerging
business models that align with these predictions.
In conclusion, this comprehensive literature review provides valuable insights into the
fundamental characteristics and dynamics of the management consulting industry as well as
the emerging potential of generative AI within the business landscape. While the existing
literature offers limited direct connections between generative AI and management
consulting, the examination of both fields offers a unique opportunity to explore the resilience
of consulting business models amidst technological and societal disruptions, as well as the
implications of generative AI for the future of consulting.
On the other hand, generative AI holds significant transformative potential that warrants
attention. The reviewed literature emphasizes the need for further exploration into how
generative AI might disrupt and reshape the business landscape. The easy access and use of
LLMs has the potential to commoditize data analysis, deliverable formalization, and the
production of personalized solutions showcasing a capacity to redefine the consulting
landscape. It is imperative to understand how any industry can effectively leverage generative
AI to redefine its value proposition, identity and “raison d’être”.
As the consulting industry explores the implications of generative AI, it becomes evident that
a thoughtful approach is crucial. The adoption of generative AI raises essential questions
regarding the evolving role and identity of consultants in delivering value to clients and
harnessing AI technologies to optimize solution efficiency and customization.
This research offers a valuable opportunity to further explore the resilience of consulting
business models in the face of disruptions and the potential impact of generative AI on the
future of consulting. By embracing technological advancements, responsibly leveraging
generative AI, and fostering a culture of innovation, management consulting firms can
position themselves at the forefront of industry transformation, consistently delivering
exceptional value to clients within a rapidly evolving business environment. Future research
should delve deeper into the practical implications of generative AI in management
consulting, with a specific focus on strategic decision-making, successful integration, ethical
considerations, and the evolving role of consultants in an AI-driven era.
II – RESEARCH METHODOLOGY
1) INTRODUCTION
2) RESEARCH APPROACH
In line with this exploration, quantitative and qualitative data will be collected through one-
on-one semi-structured interviews. This method facilitates a direct correlation between the
theoretical concepts outlined in the preceding chapters and the practical realities observed
within consulting firms (Aubin-Auger I. & al., 2008). Notably, these interviews will include
not only factual information but also subjective opinions from the participants, underscoring
the need for careful analysis and interpretation (Jacquemin A., 2019; Aubin-Auger I. & al.,
2008). The richness of qualitative research enables a more nuanced understanding of the
impact of research questions, while acknowledging potential bias in the responses provided by
the interviewees.
The qualitative nature of this research methodology allows for flexibility and openness to
unexpected findings (Jacquemin A., 2019). The interview data will be meticulously examined,
seeking emergent themes and patterns that contribute to a comprehensive understanding of
how AI influences consulting firms. This analytical process necessitates continuous
interpretation and an open-mindedness towards unforeseen insights that may arise.
These interviews will inform the practical analysis conducted in this study. By delving into
the collected information, a deeper appreciation of the practical implications of AI in the
context of consulting firms will be garnered. Through qualitative exploration, the diverse
perspectives and experiences of the participants will be considered, allowing for a holistic
exploration of the thesis subject.
4) DATA COLLECTION
a. SAMPLING CRITERIA
The sampling criteria for this study will encompass individuals from established management
consulting firms who meet specific requirements. Participants will be required to have a
minimum of three years of experience within the consulting industry to ensure a sufficient
level of professional expertise and familiarity with the industry dynamics.
This study will incorporate interviews with both middle managers and top-level executives
within management consulting firms. Middle managers will be included in the study as they
are actively involved in the operations of consulting engagement, possessing hands-on
experience and insights into the practical application of generative AI. Their perspectives can
provide valuable insights into the day-to-day implications associated with integrating
generative AI into consulting projects. On the other hand, interviews with top-level executives
will be conducted to gain insights into the strategic vision and high level decision-making
regarding the adoption and implementation of generative AI technologies within their firms.
This dual perspective from middle managers and top-level executives will provide a
b. DATA ANALYSIS
Content analysis is a widely utilized qualitative research methodology that entails the
identification of themes, patterns, words, and concepts within qualitative data. In the context
of our research on the impact of generative AI on management consulting business models,
content analysis provides a valuable framework to glean insights from industry stakeholders.
By quantifying and examining the meanings and relationships inherent in the data, our goal is
to obtain a comprehensive understanding of how professionals perceive and navigate the
evolving landscape of management consulting. Our data sources encompass a range of
interviews, industry reports, articles, and pertinent publications, enabling us to employ a
relational analysis approach that explores interconnections between various concepts. For
example, we may explore how the integration of generative AI influences service offerings,
client engagements, or overarching business strategies. Through the implementation of
content analysis, we seek to unveil valuable insights into the implications and opportunities
that arise from the incorporation of generative AI within the management consulting domain.
5) RESEARCH ETHICS
The study adhered to ethical principles of anonymity and confidentiality to safeguard the
privacy of human participants throughout the research process, including data collection,
analysis, and reporting (Allen, 2017). Measures were taken to remove personally identifying
information provided by the interviewees, ensuring the preservation of confidentiality.
This research was conducted with confidentiality (Allen, 2017). Both instances involved the
collection, transformation, and examination of information from participants. Rather than
exposing individuals' experiences and behaviors, the focus of this study centered on
understanding their actions and experiences. As responsible researchers, it is essential to
respect the respondents' wishes, particularly when they express a desire for anonymity.
1) MAIN FINDINGS
2) QUANTITATIVE PORTION
Our semi-structured interview comprised a set of quantitative questions. The objective of this
portion is to gauge, beyond the verbatim and semantics, the magnitude of the expected impact
of generative AI on the management consulting industry. We have asked our panel of
participants 3 quantitative questions that we will, at a later stage, put in perspective with
elements such as their background, their roles and their level of experience.
1) ADOPTION
“Q1 - Do you occasionally use generative AI (more than once) in a professional context?”
Although all the participants included in this survey have already tested a generative AI
model at least once, mainly ChatGPT, it was interesting to collect data on the current use of
generative AI by these participants and correlate it other elements such as the expect impact
on the industry, their roles and their level of experience. The results show that 5 out of 12
participants use generative AI regularly in a professional context, which is impressive in
terms of adoption given that the technology has been available for roughly 6 months.
# ROLE USE
1 Manager YES
2 Senior Manager YES
3 Partner YES YES 5/12 participants
4 Director NO NO 7/12 participants
5 Senior consultant NO
6 Senior Manager NO
7 Manager YES Top-level Execs 25%
8 Partner NO Middle Management 50%
9 Principal NO
10 Senior Manager NO
11 Manager YES
12 Senior Manager NO
Figure 7 - Results 'Do you occasionally use generative AI in a professional environnement?"
When asked about the purpose of use of ChatGPT, the participants identify 3 main use cases:
- Formalization – 17% of participants: Under time constraints and when the stakes
are relatively mild such as internal communications. The participants use ChatGPT to
create drafts of paragraphs that they intend to write by communicating their ideas
using bullet-points. Although the final version is thoroughly reviewed and revised in
all circumstances, having an advanced draft allows for substantial time savings. It is
important to mention that some participants were extremely reluctant at the idea of
using ChatGPT to formalize any sort of communication even in draft form as they
identified this task as very personal and close to their identity as consultants. At least 2
of the participants argued that nuance and judgement are quintessential to the
consultant’s identity and cannot be delegated, even to more junior consultants.
Last but not least, as stated per our sampling, we have interviewed both top-level execs and
middle managers to get both a strategic and operational view of the topic at hand. The usage
of ChatGPT by middle managers in our participant sample is higher (50%) than by top-level
executives (25%), which could be explained by the fact that top level executives are more
involved in client facing roles and acquiring business, and less involved in the operations that
would require the use of an AI-powered assistant.
2) SCALE OF IMPACT
“Q2 - On a scale from 1 to 10, how much would you rate the impact of generative AI on the
consulting industry?”
Although quantitative in nature, this question is very appreciative since it doesn’t provide a
framework for assessing each level of impact in the scale. The nature of a semi-structured
Figure 8 - Results " From 1 to 10, expected impact of generative AI on the management consulting industry"
Globally, results indicate that generative AI is expected to have a low-mild impact on the
management consulting industry with values ranging from 1 to 6 with a 3.67/10 average. The
justification provided was example-based in most cases, stating that management consulting
has existed and survived before excel, cloud computing and big data and has continued to
thrive and adapt to technological disruptions throughout its history. Furthermore, they argue
that despite being a knowledge-based industry, it is primarily human conveyed which explains
why it has remained successful despite a commoditization of knowledge. However, the
participants on the upper side of the scale justify that generative AI is different in the sense
that it does not provide just knowledge, but also reasoning, customization and know-how.
3) TIME HORIZON
“Q3 - On what time horizon do you expect the impact of generative AI on the management
consulting industry to materialize?”
After assessing the magnitude of the potential disruption, the purpose of this question is to
gauge the urgency of disruption in the management consulting industry. The participants were
given a time-frame for each mode: Short-term (1 to 3 years), mid-term (4 to 7 years) and
long-term (above 8 years).
The results in this case seem quite divisive. Roughly half of the respondents (5/12) estimate
that the change is impending while the same portion estimate that the impact won’t
materialize anytime soon. Only 2/12 respondents provided a moderate answer with a mid-
term horizon. Solid arguments were presented on both sides to justify opposing positions.
Advocates of a short-term disruption argue that the adoption curve of generative AI, even
within their firms is unprecedented. Several examples have been provided such as company
wide emails sent to clarify the rules of generative AI use in a professional context to avoid
any client data leaks (interviewee #2), or spontaneous consultant initiatives to animate
workshops on the best practices of AI usage (interviewee #5). On the other hand, supporters
of the long-term impact vision argue that the technology is far from being mature enough to
justify company wide transformations on the short-term, without minimizing the importance
of local and contained experimentation with the introduction of generative AI both in
consulting engagements and