Scale up strategy for SaaS-PaaS ventures

(1)

Scale up strategy for SaaS-PaaS ventures

A case applied to

Company project by:

Name: Rodolfo G. Gordillo

Student Id.: 11948035

E-mail: rodolfo.gordillo@student.uva.nl

Supervisor: Dr. Sebastian Kortmann

Date of submission: September 30th₂₀₁₈

(2)

I. Abstract / Executive Summary

Nowadays data analytics initiatives have gained relevance in top performing companies. The purpose of the present document is to assess current position of Volkswagen Data Lab towards a potential future growth.

Currently, SaaS (Software as a Service) has become a trendy buzzword among several industries. The stage of development varies depending on the industry under discussion, for example, tech giants (Facebook, Apple, etc.) embrace data management as part of their DNA. On the other side of the spectrum, capital intensive companies involved in consumer goods manufacturing were pushed into this field due to small margins and fierce competition for market share.

The automotive industry landscape has dramatically changed over the last decade. The biggest players are facing, more and more, the threat of technology giants. Tech giants like Google hold enough resources to broaden their portfolio towards traditional industries. This new paradigm has created fertile soil for new starters to provide solutions that in the past were only meant for big software developers. Volkswagen and other car manufacturers aware of the new business environment, started their own ventures in the IT field with the intention to deal with the newcomers and at the same time gain a competitive advantage in the long run.

Scale up strategies for two business unit (vertical axis) and a cross functional team (horizontal axis) will be described. These managerial recommendations and disclaimer aim to give Volkswagen Data Lab fresh insights in order to be prepared for further development in the data analytics field, and also on how to interact with its HQ maintaining its entrepreneurial mindset.

(3)

Table of Figures

Figure 1: Category Maturity Life Cycle (Escape Velocity, G. Moore, 2011) ... 8

Figure 2: The Competitive Positioning Compass (Crossing the Chasm, 2014) ... 8

Figure 3: The 5 Forces (Porter, 2008) ... 9

Figure 4: Hierarchy of Powers (Moore, 2011) ... 9

Figure 5: Revenue Generated from Big Data (BCG, 2014) ... 11

Figure 6: Captured Value from Data Analytics (McKinsey, 2017) ... 12

Figure 7: The Four Zones (Moore, 2015) ... 13

Figure 8: Collaboration Model and Startup's Maturity Stage (BCG, 2017) ... 14

Figure 9: Hierarchy of Power (Moore, 2011) ... 16

Figure 10: The Four Zones (Moore, 2015) ... 18

Figure 11: The Three Horizons (Moore, 2011) ... 18

Figure 12: Open Crowdsourcing Pillars (simplified from Kohler’s, 2018) ... 20

Figure 13: Hierarchy of Powers - Assessment Criteria ... 24

Figure 14: R&D Assessment ... 28

Figure 15: Production and Logistics Assessment ... 32

Figure 16: Deep Learning Assessment ... 36

Figure 17 – R&D (Motorsports) ... 40

Figure 18 - R&D - Connected Car ... 41

Figure 19: The Efficient Frontier (Zack and Brandon-Jones, 2018) ... 42

Figure 20 - Deep Learning Interactions ... 43

Figure 21 An Emerging Pathway to Digital Organization Maturity (BCG, 2017) ... 45

Figure 22 - Crossing the Chasm (Extended)... 47

Figure 23 - Diagnosing Organizational Context (Birkinshaw and Gibson, 2004) ... 48

(5)

II. Introduction

Societies around the world are facing dramatic changes, the globalization and means of communication have radically speed up the pace of changes. This new economic era finds us immerse in the transition from “traditional business economics” to “sharing economy” paradigm. The buzzword “shared economy” is currently spread everywhere, but in order to fully understand its meaning and implications, the reader must understand its roots. The first studies on gig economy dates back to 1978, when M. Felson and J. Spaeth made wrote “Community Structure and Collaborative Consumption” based on the US society. Basically, the authors applied A. Hawley’s frameworks (Human Ecology: A Theory of Community Structure, 1950) order to provide quantifiable examples of the benefits of applying this concept. They defined “acts of collaborative consumption” as “those events in which one or more persons consume economic goods or services in the process of engaging in joint activities with one or more others” (1, 1978). A simpler Oxford Dictionary definition states: “an economic system in which assets or services are shared between private individuals, either for free or for a fee, typically by means of the internet”.

Humankind and business history share some similarities, for example, those organisms (or enterprises) with the capability to adapt to new environments are the ones meant to survive. As a matter of fact, in early years of the 20th_{century more than two hundred} companies played a role in the automotive landscape. However, after the great depression (1929) less than fifty managed to overcome the inconveniences. Interestingly, Volkswagen started its operations in 1937 and made its path to become one of the worldwide biggest car manufacturers dethroning US based companies from the ruling position.

At a glance, the car as a good has also changed realms, from a luxurious good to a must have on every family. The already mentioned new economic paradigm plus younger generations have casted doubt into the later statement, in other terms, the concept

ownership is being challenged. The increasing interest and success of collaborative economy

initiatives added up to the current possibilities behind data management have made possible the disruption of what for decades were considered safe business environments. Some

(6)

examples, nevertheless more and more initiatives are out there to be picked up, e.g., Car2Go (backed up by Daimler) or IONIC (backed up by Hyundai).

Volkswagen Data:Lab Munich is a forward-looking branch from the mother company that has the purpose to secure company success and sustainability through the digital transformation era. Data Lab Munich might be categorized as a provider of advanced software services and solutions, however depending on the business unit analyzed it is possible to fall into two different categories: Software as a Service (SaaS) or Platform as a Service (PaaS). The NIST defines:

“Software as a Service (SaaS): The capability provided to the consumer is to use the provider’s applications running on a cloud infrastructure. The applications are accessible from various client devices through either a thin client interface, such as a web browser (e.g., web-based email), or a program interface. The consumer does not manage or control the underlying cloud infrastructure including network, servers, operating systems, storage, or even individual application capabilities, with the possible exception of limited user specific application configuration settings.

Platform as a Service (PaaS): The capability provided to the consumer is to deploy onto the cloud infrastructure consumer-created or acquired applications created using programming languages, libraries, services, and tools supported by the provider. The consumer does not manage or control the underlying cloud infrastructure including network, servers, operating systems, or storage, but has control over the deployed applications and possibly configuration settings for the application-hosting environment.” (The NIST Definition of Cloud Computing, P. Mell and T. Grance – 2011)

This report aims to provide advice to Volkswagen Data:Lab Munich for its scale-up strategy. Recommendations will be shared towards the end of this script, in order to create awareness on how to avoid the usual pitfalls that make 70% of startups fail (Forbes, 2017). In

(7)

III. Literature Review

In this section, different sources of literature will be presented, always focusing on the following pillars:

I. Data:Lab is an entrepreneurial entity within one of the largest MNE1_;

II. Data:Lab’s goal is to lead Volkswagen’s Group through the digital transformation towards a sustainable future. How? By thriving innovation through the usage of data analytics techniques;

III. Answering the research question: “What is the most suitable scaling up strategy for Volkswagen Data:Lab Munich?”

In order to tackle these three pillars with a 360 degrees spectrum, the background reading is divided in two different sections:

a) “As Is” Assessment b) Scale up alternatives

A. “As Is” Assessment

The “As Is” assessment is the cornerstone of this document, therefore the authors and articles reviewed on this section were carefully selected. For instance, Geoffrey Moore due to his background in software business combined with entrepreneurial experience made of him a must to have for this section. Two titles were selected as references “Crossing the chasm” (2014) and “Escape Velocity: Free Your Company's Future from the Pull of the Past” (2011). The selection was not random, while Crossing the Chasm is a book focused on pointing out the risks and death traps that entrepreneurs will face when trying to enter the mainstream of the software industry (or high tech). The second book complement these insights by sharing advice and tips on how successful companies managed to achieve sustainable growth in a highly competitive environment.

(8)

Figure 1: Category Maturity Life Cycle (Escape Velocity, G. Moore, 2011)

The precedent figure shows the technology life cycle of a product, since its early development until the end of life. Among the life span, the author defines for these cycles a type of customer, ranging from Innovators, Early Adopters, Early Majority, Late Majority and Laggards. Despite this might not sound relevant for the Data:Lab case, in further section parallelism will be drawn.

For the purposes of the present document we won’t focus on the marketing aspects embedded in this model. However, it is important to highlight that “the chasm” is defined in the book as the breach between the customers defined as “early adopters” and “early majority”. While the first are crucial during the first stages of

(9)

“headpin” to bridge the gap between customer segments; 2) their feedback is a must in order to develop 100% polished product that other customers might not be able to consume before it has been already tested by the market.

As it was previously indicated, the technological advancements added up to (old) new consumer behavior has casted doubt over traditional industries. Therefore, the traditional approach on how to deliver competitive and sustainable businesses is no longer valid.

VS

The figures 3 and 4 are placed in such way that the reader is able to contrast two different angles on forces acting within organizations. On the left side, the worldwide famous “Five Forces” by Michael Porter approach. On the left side, the so-called ‘framework of

frameworks’, “Hierarchy of Power” by Geoffrey Moore. In this particular work, the focus has

been made in the framework presented in the Figure 4. As it was mentioned in the beginning of this section, it will provide us with complementary angles to assess the current status of a company.

Also, on a personal note, I found extremely useful Michael Porter’s 5 forces analysis when the assessment is from an “inside out” perspective, this means, from the company perspective towards the surrounding environment. On the other hand, Geoffrey Moore’s Hierarchy of Power give scholars the ability to apply the concepts embedded in both ways, “inside out” and “outside in”. Each category of this framework will be explained in detail in

(10)

Up to now the literature described is mainly focused on how to strategize and deal with the environment from the outside in view. To close up the “as is” assessment, the inside out sight proposed by Julian Birkinshaw and Cristina Gibson (2004) on how a company should be able to steer its future while delivering todays result. This managerial review brings easy to apply charts allowing top managers to assess current resources and their “ambidextrous” mindset. These authors highlight that companies to succeed in the term needs to be both, easy to adapt and align. The hassle for the management is to find the right balance between responsiveness and long-term alignment.

This particular approach looks fundamentally useful for Volkswagen Data:Lab’s future, because it allows to keep track on today’s results (that company’s headquarters require) while giving the possibility to assess how future might change and how to act to sustain growth. In section VIII, Appendix B, a quick assessment format is shared. This simple but powerful tool could be used for further studies or by management to assess current Data:Lab’s ambidextrous performance.

B. Scale Up Alternatives

At this point, the reader might be wondering: why is important to scale up? The answer to this question is given by other MNE trends and the already mentioned pace of change in businesses. The research delivered by Boston Consulting Group and Hello Tomorrow (2017) showed that 82% of corporations find the interactions with startups ranging from “somewhat important” to “important”. Also, despite not having access to the official statistics, different consulting firms stated over the past few years the fact that corporations are establishing their own digital venture initiatives. Finally, the data management projects have also impact on companies’ revenues:

(11)

Figure 5: Revenue Generated from Big Data (BCG, 2014)

From the figure above we see that, on average, 12% of the revenue generated in leading companies comes from maximizing the use of data analytics techniques. The reader should note that the figures date back from 2014, therefore the reader should expect higher percentages these days.

On the same path, McKinsey Analytics 2016’s report stated that there are still 70 to 80% value to be captured by manufacturing companies from data analytics projects. However, the benefits are not spread all over the value chain and are mainly focused on early stage processes such as research and design. Very few companies, for instance GE, Siemens or Schneider Electric have managed to introduce data analytics into their manufacturing process. Not surprisingly, these enterprises have strong sales division in goods related with industrial automation and smart devices integration through proprietary communication protocols.

(12)

Figure 6: Captured Value from Data Analytics (McKinsey, 2017)

In terms of academic literature, the texts reviewed to provide advice to Volkswagen Data:Lab Munich on how to scale up covers a wide span. The underlying reason to have made this selection, is due to the fact that one expansion strategy does not fit the different business units.

The author that fits the best sketching how multinational enterprises should strategize their innovative ventures, and keep their competitive advantage over a longer period of time is, again, Geoffrey Moore in its book “Zone to Win” (2015). This book matches and gives this document a continuum to close any gaps that might have been open in the previous chapters. In “The Zone to Win” the author addresses with a pragmatic touch how to prepare organization to overcome this era of constant disruption. One of the main takeaways of the book is the “Four Zones” framework. This framework is intimately linked with the “Category Maturity Life Cycle” (see figure 1).

(13)

Figure 7: The Four Zones (Moore, 2015)

The zones detailed in figure above (Chapter 2, Zone to Win, 2015) are linked with “time horizons”. These marriage between Maturity Zones and Investment horizons, give investors the possibility to:

- Have a time frame when creating portfolios; - Understand the risks of innovative projects

As it was mentioned in previous passages, for Data:Lab Munich there is no unique expansion strategy, each business unit (or company axis) will fall into a different quadrant according to the complexity and length of the products developed. Each zone will be explained in the following section “Frameworks”.

As a closure, despite it is not the matter of the document, the analysis made up to this point is always wearing the startups “suit”. Further research that might be relevant for Data:Lab Munich, is to look into the “wishes” of what corporations are looking for when starting with their own high-tech branches or joint ventures. to consider are the corporate the counter part of startups ventures, this could be companies looking for leverage to introduce new products into the markets or simple venture capitalist. The following figure it is just the tip of the iceberg of what should be a totally separate research study:

(14)

(15)

IV. Frameworks

In the previous chapter the several sources from literature where summarized and explained to give context when the delivering the assessment analysis. The present section will be focus on highlighting those frameworks as they were conceived by the different authors before applying them to our company case study.

A. Hierarchy of Powers

As it could not be otherwise, the Hierarchy of Powers framework is the first one to be tackled. This “framework of frameworks”, as Moore’s defined it, provides tools and triggers uncomfortable questions to deliver a deep critical analysis of established organizations.

The reason why this framework is so robust are:

- Allows to play in both sides of the “outside-in” “inside-out” spectrum. This means that its application allows to pivot between customer-centric and company-centric initiatives;

- This framework not only focus on strategy creation, but also incorporates execution aspects that are highly valuable in these changing times. Each layer of the framework has embedded the concept of resources and the what’s and how’s to move forward;

- There’s a natural linkage between the hierarchies and products life cycle (figure 1). The possibility to draw this parallelism makes it flexible, this means that could be applied for enterprises in different stages, from incubators until industry leaders; - The framework establishes a hierarchy in the decision criteria. The pyramid is an

easy way to represent top down the importance of each component; - The investor’s point of view is included;

In one sentence, gives a full overview on how to “Free Your Company’s Future from the

(16)

The figure 9 illustrates five types of economic powers. The order of appearance is correlated with the specificity and/or the ability of the enterprise under assessment to shape it by its own means. According to Moore’s scale, the most general driver for growth is the “Category Power” while the most specific one is the “Execution Power”.

The different powers are not isolated one from each other, moreover the dynamic interactions between them creates totally different outcomes. For instance, let’s take as an example the California’s giant: Apple. Apple executes year over year a perfect marketing campaign to promote its products and shows its leadership. In terms of the framework it is possible to remark that for Apple’s case we found a company where the 5 powers are aligned. On the hand, going a few years before the iPhone was introduced to the market, Nokia enjoyed the same benefits, however the company fail to keep the lead on the offer to the market. The final result is well known, Nokia announcing bankruptcy while Apple being on the top innovative companies in the world.

The author created the following summary to understand each category:

“Category power: is a function of the demand for a given class of products or services relative to all other classes. It creates the generic growth dynamics within which your company operates.

Company power: power reflects the status and prospects of a specific vendor typically signaled by that company’s market share. Growth here is at the expense of your competitive

(17)

Offer power: is a function of the demand for a given product or service relative to its reference competitors. Growth here is signaled by net change in revenue.

Execution power: execution power is the ability to outperform your competitive set under conditions that favor no vendor in particular. Growth here is signaled by driving programs to tipping points, after which they become self-sustaining.” (Moore, 2011)

In a simpler way, and keeping in mind the Nokia example we will simply define each hierarchy as follows:

- Category power: the place where our product or service is placed in the Maturity Life Cycle chart (figure 1);

- Company power: bargaining power when dealing with participants of the supply chain;

- Market power: the relative position, in terms of market share, from the enterprise compared with competitors

- Offer power: the ability to meet customer’s demands, and at the same be ready to satisfy future ones;

- Execution power: ability to deliver initiatives and achieve results when execution an agreed strategy.

B. Four Zones

The “Four Zones” framework (Moore, 2015) is the second most relevant for the present research. This approach is the natural complement to the Hierarchy of Powers framework. The Framework is described in more detail by Geoffrey Moore in the second chapter of the book Zone to Win: Organizing to Compete in an Age of Disruption.

The zones analysis gives us a pragmatic interpretation on how to allocate resources in the era of disruption. Data:Lab is a business unit within Volkswagen Group, therefore from an investors point of view is a plus to understand in which time windows the returns should be expected.

(18)

The horizon’s segmentation is coherent with the aspects touched in the Hierarchy of Power framework (Figure 9) and the Maturity Life Cycle (Figure 1). The result of the combination of these two approaches, is a holistic view when assessing and giving recommendations to existing business.

Each zone has its investment horizon, KPI’s2_{, goals and pace of growth.} Hence, there is no single leadership style that could manage to deal with all these variables at the same time. MNE’s that understand this and organize themselves to overcome this constrain, are most likely to be successful in their disruptive ventures (or stay profitable in volatile business environments). In particular for large corporations the four zones have to managed at the same time in harmony, if the utmost goal is to achieve mid and long-term results. The horizons and its financial economical implications are sketched in the following figure.

(19)

disruptive wave, or might unleashed a hidden advantage against competitors. It is expected that teams performing in this area are focused on the “next big thing”, instead of dealing with daily processes. In the mid-term, the initiatives incubated should show a future benefit for the organization, in order to prepare the foundations to scale up into the Transformation Zone.

As it occurs with pharma developments, several initiatives proof not be profitable or even feasible at industrial scale. Therefore, the Transformation Zone is highly likely not to be crowded with projects, however once a project reaches this status, this means that has higher probabilities to become the next revenue source within an organization. To move from the Transformation Zone to the Performance Zone might take several quarters of careful performance analysis.

In the Performance Zone is where the major efforts are concentrated by organizations management. Initiatives located in this quadrant create the necessary revenue stream within an organization. In this area the goal is to have a fine-tuned organization, working with tangible metrics, and minimized risk.

The Productivity Zone, as it names stations, have as a main strategic goal to tackle inefficiencies and improve the gaining created in the Performance Zone. Moore describes the “primary goal is to extract resources from non-core work in order either to invest more in core tasks or take the savings to the bottom line”.

C. Crowdsourcing Pillars

The last framework to be reviewed is a simplification from “Paths and Strategies to Scale Crowdsourcing Platforms” (Kohler, 2018). Crowdsourcing platforms are gaining relevance when it comes to scale up business. Private (or hybrid) alternatives open alternatives that leads to companies with hyper-growth figures3_{, for instance Facebook or} Netflix.

(20)

Crowdsourcing unleashes enterprises possibilities in terms of resources. According to McKinsey one of the limitations that companies will face in near future is the scarcity of highly trained data analytics professionals, in particular in the fields of machine and deep learning. This is the main reason behind the selection of this brand-new analysis.

The full version of Kohler’s scheme is included in Section VIII Appendix C, in this section it is shared an adapted version that will be applied in the preceding chapter of this work.

Figure 12: Open Crowdsourcing Pillars (simplified from Kohler’s, 2018) From the simplified scheme it is possible to highlight three pillars:

• Stakeholders: "an individual, group, or organization, who may affect, be affected by, or perceive itself to be affected by a decision, activity, or outcome of a project" (Project Management Institute, 2013)

o Internal: in terms of the present analysis we conglomerate all those actors that within the organization might affect or be affected by means

(21)

stakeholders (or customer following Kohler’s definition) would be those subjects that actively look for the product.

• Value creation / Monetization: according to the author during the scale up phase the crowdsourcing venture should adjust their initial value co-capture mechanism. The value created and the way it is shared could be express with monetary and non-monetary incentives.

• Interactions:

o Create: in order to foster interactions crowds are in demand of technically complex problems, no organizational burden. As it is happening with many Open Innovation (OI) initiatives, the members of these spaces require clear guidelines in order to deliver value through their “winning” intervention. Another interesting point rise by Kohler is that “complexity could also lead to diseconomies of scale if the platform needs to provide additional unscalable customer service”.

o Curate: the final goal is to deliver high quality product. Of the course sometimes, iteration will lead to a final POC (proof of concept), but it is also necessary to understand which is the reasonable threshold for trial-error phases. Also, on the customer shoes, it is mandatory to understand their limitations, this means that an overload of solutions and patched might end in excellent products never used due to their impossibility to process them.

o Consumption: basically, a crowdsourcing platform exists If there is consumption, “no consume → no crowd → no platform”. This is the reason behind it is equally important to manage customers and interactions, in other words supporters of the crowd.

In one sentence, if the interactions created inside the crowd create value, and simultaneously stakeholders are managed, therefore it is possible to confirm that the conditions for scaling up are suitable.

(22)

V. Applied Frameworks

The frameworks and literature reviewed in chapters 3 and 4 will now be applied to the Volkswagen Data:Lab Munich. The input for the present analysis is based in a series of interviews held during the last week of May 2018 at Data:Lab’s premises in Munich. Despite underlying subjectivities, this assessment can be used a compass when running future decisions or creating new strategies.

A. Research and Data Gathering

The complete research process counted with different stages:

• Management interview: held with Volkswagen Data:Lab Munich representatives, The Amsterdam MBA Program Director, and MBA Candidates at Amsterdam Business School premises – April the 20th_2018;

• Questionnaire review interview: held with Volkswagen Data:Lab Munich representatives, and MBA Candidates via Skype – May 11th_{of 2018;}

• Research interviews: held with Volkswagen Data:Lab Munich representatives, and MBA Candidates at Volkswagen Data:Lab premises in Munich – from May 28th_until May 30th_{of 2018;}

• Complementary interviews: held with Volkswagen Data:Lab Munich representatives, and MBA Candidates via Skype – several meetings between June and July to complement the information gathered during the visit to Data:Lab’s offices in Munich.

(23)

• What are the large companies would you prefer us to benchmark with? (E.g.: Heavy industry (Shell, GE), FMCG (Unilever, P&G), Tech (Booking, FB), Consulting?

• What are the scale-up companies that you would like to consider (Y Group - Partnership consulting model)/ any other you have in mind?

• What are the kind of projects that you have done for each of the divisions? What has been the success rate of implementation of these projects (for some divisions it might be 50%, for some divisions it might be 0%)? Can a successful implementation be applied to other Brands? What were the reasons for successful implementation? Can/ How have you leverage(d) your interactions to do more projects for the same brand (same division or could be other divisions)?

• What are the Brands for which Data: Labs has the strongest/ weakest partnerships? What are the reasons for the same? How is the Business Development/ Sales team for Data: Labs currently organized? Is it organized according to divisions or Brands? How big are the size of the projects that the division currently carry on (EUR 50K-250K / EUR 50K-250K- 1 million)?

• Who are the external partners (could be out-sourcing) that you have worked with? Where do you have strong relationships? Was this partnering Brand led or Data: Labs lead? Are you leveraging your knowledge from external partners? (E.g. Schneider Electric and its 4.0 industry vision)

• What is your experience with open-innovation? What processes are you currently outsourcing? What are your initial ideas when it comes to an implementation model? Increase the team-size of Data: Labs, bring external partners, Do JVs with brands, dedicate senior resources to ensure implementation?

• How easy or difficult it would be Data: Labs to get top-management onboard and shift focus towards implementation? What is the management expectation from Data: Labs? Be self-sustaining, on-demand service or become strategic, thought leaders? What are the current KPIs? How are you measuring success? Have you ever overlapped teams from the lab with Brands?

The data gathered from the interviews was recorded in electronic format for its analysis, none of them were digitally recorded.

(24)

B. Data:Lab’s Hierarchy of Powers Assessment

As it was already mentioned in the section IV “A” of the present document, when applying the Hierarchy of Power framework, the outcome will provide us a 360 degrees assessment of Volkswagen Data:Lab Munich creating the foundations for further analysis.

In this case, each layer of the pyramid has been assigned with a specific weigh. Each value was assigned from an “outside-in” priority perspective, this means that those layers mainly Data:Lab dependent have less weightage than those environments linked. The decision, far from being randomly taken, has grounds on the idea that a corporation itself has stronger possibilities to steer its own policies than to do it with the market as a whole.

Figure 13: Hierarchy of Powers - Assessment Criteria

(25)

or late are we in the market, or even if the project belongs to a blue ocean4_{. The} third pillar is to know what size has the market and how big is the company proposal into it. The author also refers to the terms TAM5_{and SAM}6_{that might be} helpful to determine in a later stage in order to come up with bigger investments. The last beam of the category power has no direct correlation with financial figures, but definitely it has an impact. The non-financial incentives refer to any decision driver that has a positive impact to the organization, for example, currently it would be impossible to think in companies that wants to lead market but has no proven track record in CSR7_matters.

For the assessment’s purpose Category Power has been entitled with 10 points of weightage.

B. Company power: this layer is referred to the possibility of the company to create a sustainable competitive advantage in an existing market. The term “competitive advantage” has embedded the customers’ positive valuation. The basis for the assessment of this power are:

▪ Technological complexity: of the product offered. Moore states that the is highly relevant to determine if the market a company is intending to play is volume or complex driven.

▪ Customer type: Who is our customer? This question refers not only to a specific entity but also in a broaden analysis, what characteristics in general have our customers, e.g., tech savvy, early adopter, laggard, etc.

▪ Relationships – Quantity / quality: That type of relations relationships do we intend to create? Do we prefer to have more and superficial or just a few but deep?

▪ Solution re-cycling potential: this pillar refers to the availability of an existing product or service to be implemented or partially re-utilized in a different market. For our assessment purpose, it also includes the possibility to be integrated by another vertical inside Volkswagen Data:Lab.

(26)

For the assessment’s purpose Company Power has been entitled with 8 points of weightage.

C. Market power: The Market Power assessment is based in “As is” and “To Be” scenarios taking into account how is internal and external competition. The analysis includes what other alternatives current Volkswagen Data:Lab customers currently have, and what possible options will arise in the future. As this assessment is dynamic, it is smart choice to map competitors quarterly and understand how their offers evolve over time.

For the assessment’s purpose Market Power has been entitled with 7 points weightages.

D. Offer power: Geoffrey Moore established Offer Power as the distance or separation between what’s being offered by a company and its competitors within the same market. It is possible to simplify this concept as “the value proposition and how our customers value it”. Also, it is possible to measure the distance on proposals following the Treacy and Wiersema criteria (1994) Product Leadership, Customer Intimacy and Operational Excellence. In this particular case, the Volkswagen Data:Lab offer power has been assessed taking into account the following criteria:

▪ Offer: is the customer receiving a complete integrated solution, or a POC8_? Is the customer satisfied with the offer and appreciate the differentiation aspects included on it?

▪ Pricing model: the pricing model should vary according to the type of solution offered. For instance, disruptive solutions should be monetized in such way that “level of disruption” created is being valued, and not simply as a cost-plus commodity.

(27)

▪ Is it possible for the vertical under assessment to make use of available external resources or partners? If so, would it be possible to develop or target a new segment with a more functional product or service?

For the assessment’s purpose Offer Power has been entitled with 6 points weightages.

E. Execution power: the author defines Execution Power as a positively correlated function of the speed, scale and focus during the implementation of a Go-to-Market Strategy. In our particular analysis it will be related with the capability of Volkswagen Data:Lab’s departments to implement or go live with projects. The pillars considered to perform the study are:

▪ Length of solution cycle: it refers to the time period between a customer raises a problem and a solution is provided;

▪ Customer capabilities: this term comprehend all the technical and non-technical skills that a customer has before a project is performed;

▪ Path dependency: these are the boundary conditions or constrains for short term implementation or growth of a particular Volkswagen Data:Lab vertical or horizontal team;

▪ Future resource requirement: this are the requirements that effectively will limit the scale-up of a team or the might jeopardized the possibility to deploy a successful project in different markets.

For the assessment’s purpose Execution Power has been entitled with 5 points of weightage.

After having explained all the assessment criteria, the following section will include the assessments conducted for:

- Research and Development (Vertical team) - Production and Logistics (Vertical team)

(28)

A. Research and Development Assessment

Figure 14: R&D Assessment

The first Volkswagen Data:Lab’s branch assessed is Research and Development. As it was explained in the previous section, the “As Is” analysis will cover the five layers presented on Moore’s framework.

Category power

Overall, in terms of Category Power, the Research and Development team is very well positioned. On the financial side, based on the information gathered during the interviews the current level of intercompany transaction ensure the self-sustainability of the team, and also the projects pipeline allows to aim for growing figures.

From a non-financial point of view, the current activities held by the team have high visibility due to their linkage with the motorsport’s world. The high exposure of the projects delivered is a great asset to consider, that cannot be easily translated into a KPI.

(29)

The only downside present is the entry point, competitors such as Toyota or Ford have been increasing their budgets for data analytics ventures over the past few years. A recent example is the partnership signed between Toyota Motor Corp. with Albert Inc. early this year to speed up the autonomous driving initiatives, adding up to its current subsidiary in Silicon Valley to foster AI initiatives (Toyota AI Ventures).

Based on the previous description, the result of Category Power for Volkswagen Data:Lab Research and Development vertical is 9 points of weightage.

Company Power

It is undeniable Volkswagen’s Company Power as whole, therefore it was expected that the same legacy has been inherited by the new ventures such as Data:Lab. In particular, the R&D team is able to deliver highly complex projects, and thanks to its current focus on the motorsports division, the outcome is tangible. Actually, there is a shared knowledge between customer and developer, that promotes relationships among teams and increase the probability of having positive results.

Also, the conscious strategy of deliver stand-alone solutions on this niche market, have created a proper terrain to think in future platforms that might leverage from previous projects. It is possible to say that this is a good example of “think big – start small”.

Based on the previous description, the result of Company Power for Volkswagen Data:Lab Research and Development vertical is 8 points of weightage.

Market Power

If we analyze the R&D’s team competition, it is possible to split it between internal and external. In the short run, it is not possible to identify internal competitors, however in the long term if the corporate strategy turns to speed up the digitization process it might create fierce internal competition, jeopardizing future growth perspectives. Internal competition could be neutralized by creating partnerships and/or center or expertise, under this scenario

(30)

On the other hand, considering potential external competitors, currently the competitive landscape count with plenty of small and medium enterprises with expertise in delivering software packages based in data analytics. However, it seems unlikely that there would be room for external companies on the motorsports side due to the current good results. Hence, current R&D market position for the motorsports division looks secure.

It is interesting to remark that those future projects linked with brands, e.g. connected car initiatives, should be studied and delivered isolated from the positive results previously mentioned. The fierce competition to gain competitive advantage and enlarge the customer base will require a different set of skills and pace of implementation that might allow external competitors to step in and develop these projects.

Based on the previous description, the result of Market Power for Volkswagen Data:Lab R&D vertical is 4 points of weightage.

Offer Power

As per the interviews result, Research and Development team is able to deliver whole solutions to the different customer within the motorsports divisions. The main reasons behind this are the stakeholders involved in the process and their capabilities. If the customers are analyzed, it is possible to identify a high level of competence and openness to implement breakthrough technologies. Also, on the suppliers (or partners) side, similar characteristics have been appreciated. The deep learning horizontal team has been identified as one of the preferred teams to partner with when it comes to deliver projects. If scaling-up turns to be a priority, enhance relationship and create agreements with partners would be a must.

From a monetary point of view, the current pricing models are mainly “consulting type” or “cost plus”, these types might seem appropriate in early stages but should migrate

(31)

Execution Power

In line with the Category Power and Company Power, the assessment of the Execution Power for R&D team is positive.

The cornerstone of the favorable outcome is the alignment between the strategy and the current execution of projects. The acceptance criteria before allocating resources to a venture plus the agile work mindset are the identified success factors. Also, as it was addressed in the Company Power section, the possibility to count with highly skilled customers (when thinking in the motorsports customers) significantly reduce the friction and conflict points when working and looking for consensus. As outcome, positive and tangible results have guaranteed a continuum request for new developments.

On the downside, that is transversal to all the data analytics ventures, the scarcity of highly skilled resources ends up being the bottleneck to speed up the implementation of initiatives.

Based on the previous description, the result of Execution Power for Volkswagen Data:Lab Research and Development vertical is 4 points of weightage. In total, the global assessment of the R&D team resulted in a score of 29 out of 36.

(32)

B. Production and Logistics Assessment

Figure 15: Production and Logistics Assessment

In organizations with more than 100 years history of manufacturing, like Volkswagen, the roles of departments such as Production and Logistics play a strategic role. Therefore, the second Volkswagen Data:Lab’s branch assessed is Production and Logistics.

Category power

When thinking about Volkswagen and its brands, an immediate link with manufacturing process is triggered. In particular Data:Lab’s Production and Logistic team has the possibility to develop projects intimately related with Volkswagen roots and history. There are uncountable opportunities to innovate into these areas, with the advantage of almost immediate correlation between a project and its financial implications. For example, all the initiatives aiming to increase processes performance will have a correlate in cost figures.

(33)

Also, in terms of market size the production and logistics ventures for its nature are constrained to manufacturing site boundaries, due to the diversity of supporting platforms found in each facility. The bright side of these drawback, is the endless potential to develop project across facilities, and the possibility to become a game changer team within the organization.

Based on the previous description, the result of Category Power for Volkswagen Data:Lab Production and Logistics vertical team is 8 points of weightage.

Company Power

The Production and Logistics team counts with the ability to deliver projects to the different manufacturing sites across brands, yet it is facing the difficulties to implement as many initiatives as it might do, not only due to the resources constrains, but also because of the inherent “fear to change” from traditionally minded managers.

According to the results from the interviews, it was remarked as a customer’s strength the processes’ knowledge, but also a not so deep understanding on data analytics ventures. Nevertheless, this gap might be bridged by enhancing communication bonds between company’s headquarters (or facilities) and Data:Lab’s team.

Another constrain is the possibility to effectively implement projects, there is a natural friction between existing available time to move forward with these ventures and the time allocated to perform everyday activities. Again, finding the balance between regular production and taking the risk to implement new projects is another matter that management continuously face.

Based on the previous description, the result of Company Power for Volkswagen Data:Lab Production and Logistics vertical is 5 points of weightage.

(34)

Market Power

The market power of Data:Lab’s Production and Logistics team it is extremely difficult to analyze without a complete understanding on how the masterplan of projects is created. Hence, for the purposes of the present study a simplified analysis has been made.

In terms of current internal competition when it comes to perform projects, it is expected that there are overlapped initiatives in both ends, VW and VW Data Lab, hence in short term perspective the competition can be categorized as medium. If the long term is under analysis, the internal competition might increase if the creation of brand-oriented data labs is multiplied.

On the external side of competition, the situation is slightly different. On the short run, the constant creation of startups focus on data analytics trends creates a fertile place for competition. However, on the long run due to the high rate of failure of startups, plus a saturated market and cannibalization process, the negative effects will be compensated and stabilized.

Based on the previous description, the result of Market Power for Volkswagen Data:Lab Production and Logistics vertical is 4 points of weightage.

Offer Power

As per the interviews result, Production and Logistics team is able only to deliver proof of concept projects, with limited capacity to roll out complete solutions. This difficulty is in aligned with the knowledge gap in data analytics between customer and supplier. The diverse background and different goals make hard for traditional units to allocate exclusive resources on Data:Lab’s initiatives.

(35)

a challenge from the customer point of view. The Production and Logistics Data:Lab’s team have the challenge to deal with conservative customers, that are reluctant to new schemes such as revenue sharing or value-based models.

The result of Offer Power for Volkswagen Data:Lab Production and Logistics vertical is 3 points of weightage.

Execution Power

Currently, Data Lab’s Production and Logistics teams project execution is highly dependent on the problem to tackle, also on the savings that any initiative might generate to the corporation on a whole. These two factors are essentially slowing down the agile implementation of projects Data:Lab’s organization as whole fosters to create.

The fluency and ways of working between a Data Lab team and a Facilities team is definitely a matter to be addressed. On one side of the table there’s a group of highly trained professionals in the data field, while on the other side the counterpart has vast field experience and proven tangible results. The group dynamics are expected not be fluent, but with the right soft skills it is possible to steer and align the group.

The future resources that solutions related with the operations and logistics areas will be closely related with the decision on who is going to lead DevOps teams. If the outcome is to have it inside Data Lab’s team, then the human capital required will experience a steeped increase project over project. As long as this is not clear for everyone involved in these initiatives, there is a high risk for Data Lab resources to remain captive from operative areas for longer periods of time than originally agreed.

Based on the previous description, the result of Execution Power for Volkswagen Data:Lab’s Production and Logistics vertical is 3 points of weightage. In total, the global assessment of the Production and Logistics team resulted in a score of 23 out of 36.

(36)

C. Deep Learning Assessment

Figure 16: Deep Learning Assessment

The Volkswagen Data:Lab’s Deep Learning team is the last branch assessed under the present scheme. This team has the particularity to be considered a “horizontal” instead of vertical, this means that them interact with all the other areas within Data:Lab.

Category Power

Due to increasing relevance and scarcity of human resources, for corporations like Volkswagen to invest in deep learning initiatives is a must. All artificial intelligence ventures have the potential to disrupt a market, or optimize assets utilization. However, despite Data:Lab’s Deep Learning team is not self-sustainable financially speaking, in the long run this paradigm is for sure to be changed. The limits of AI9_{are yet not define, therefore the market} opportunities, added to the cross functional strategy of Data:Lab’s Deep Learning team, creates a proper environment to this unit to create bonds beyond the current landscape.

(37)

is not leading nor behind other competitors on the same field. The history is different is the comparison is drawn with FMCG organization, where AI projects has been on top of the management almost simultaneously with the entertainment industry.

Based on the previous description, the result of Category Power for Volkswagen Data:Lab’s Deep Learning horizontal is 8 points of weightage.

Company Power

Undoubtedly, due to the fact that Artificial Intelligence is a new topic that is gaining space within the organization, the nature itself of these initiatives are appointed as complex. On the positive note, the strategic decision to be defined as a horizontal team, makes the interaction customer-supplier easy, assuming that there is a common shared language within Volkswagen Data:Lab.

Another positive factor for this team, is the capability to create a knowledge or intra-center of expertise, allowing the re-utilization of previously developed solutions to be used in different projects.

On the downside, as the organization under analysis is a car manufacturer, the initiatives where Deep Learning teams has been involved are those that are not car-development related. This could be related to the strong bonds with the Data:Lab vertical units, hence missing the opportunity to share the potential benefit of the implementation within other environments.

Based on the previous description, the result of Company Power for Volkswagen Data:Lab’s Deep Learning horizontal is 5 points of weightage.

Market Power

In terms of market competition, there are two scenarios rigorously different. As per the research, the current internal competition is low, the main reason behind this output is the type of projects that are handed by the Deep Learning team. Also, if a forecast is done,

(38)

that even with the foundation of other Data:Lab’s within Volkswagen, fierce competition wouldn’t be expected with increasing number of initiatives and scarcity of human capital.

From the external point of view, as AI is leading most of entrepreneurial ventures it would be impossible for Data:Lab’s Deep Learning team to be ahead or cover more areas than competition. Again, if a future perspective is drawn, the contemporaneity of AI leads the thoughts towards a hostile competition for resources and knowledge. However, the founding initiatives for startups could be the starting point for a broader center of expertise (CoE) in this field.

Based on the previous description, the result of Market Power for Volkswagen Data:Lab’s Deep Learning horizontal is 4 points of weightage.

Offer Power

According to the interview held, Data:Lab’s Deep Learning team has capability only to deliver proof of concept products. The cross functional nature of the team does not promote the ability to provide full solutions nor to lead end-to-end projects currently.

Similarly, to what happens with Production and Logistics team, the pricing model chosen to perform transaction is “cost plus”. As it was mentioned at the beginning of this assessment, the current non-self-sustainable condition of the team do not create the best scenario to implement more developed pricing models.

In terms of internal partnering, the team is closely interacting with different Data:Lab’s verticals simultaneously. This practice requires alignment when agreeing on deliverables and functionalities. The use of techniques such as design thinking has proven to be useful to set up common grounds since early stages of projects. Also, it is interesting to highlight the 3-months financial support program for start-ups, that allows to gain knowledge from third

(39)

Execution Power

To finalize the assessment, the Execution Power of the Deep Learning team will be studied. Deep Learning and R&D share similarities on how the initiatives are implemented. Both teams work having an agile mindset, mainly with a pipeline of projects up to 3 months. The main difference is that within Deep Learning, still exceptional ventures have a longer duration, between 6 months and 1 year.

As it was mentioned in the introduction, Deep Learning is a cross functional team, one of the biggest advantages of this approach is the reduced dependency to work with a single business unit or brand. However, there might be risk of lack of focus and consistency on the results if projects are not properly analyzed before being introduced in the pipeline.

On the customer side, when it comes to intra Data:Lab’s the common grounds are pre-existing, therefore communication and projects are easier to handle than when it comes to deal straight forwards from other company divisions.

Based on the previous description, the result of Execution Power for Volkswagen Data:Lab Deep Learning horizontal is 4 points of weightage. In total, the global assessment of the Deep Learning team resulted in a score of 27 out of 36.

(40)

VI. Recommendations

The final goal of this research was to assess and provide a pragmatic approach to the studied areas on how to scale up its operations in a smooth and sustainable way. In order to keep the consistency, the reader at this point is familiar with the document structure, therefore the recommendations will be tailored for each team previously mentioned. Also, the reader should recall section IV.B and IV.C from the present document, where the scale up frameworks were theoretically described.

A. Research and Development Team

During the interviews process it was possible to identify the Research and Development team as the more aligned and mature in terms of strategy and execution. Currently R&D team has a very clear project’s pipeline, supported by robust results. This is the reason why the recommendation on how to scale up would be split in two sub-teams. This division far away from being random, it is based on future challenges that Volkswagen Data:Lab’s R&D team will face in the near future.

Motorsports

The first set of recommendations are based on the sub-team within R&D that might have focus on the motorsport’s initiatives.

From the total score of the Figure 17: R&D (Motorsports)

(41)

visibility across the organization gaining stakeholders support, and hence resources. This has been a virtue of the team, the understanding the non-financial drivers that might steer decisions.

As the current results are promising, and the strategy is clear and properly executed, the next step for R&D Motorsports is to move into the “Productivity Zone” of the framework. Moving into this direction mean to work on the cost structure of the initiatives, this might be contradictory with the existing entrepreneurial mindset. However, if the expansion boundaries are clearly defined, it is possible for this area to become a highly productive business unit. Some remarks that need to be consider to succeed in the long rung are:

a) The first one is technology related, the team should ensure the utilization of a common platform across projects.

b) The second one is human related. Talent in the data analytics field is scarce resource. For a young organization like Volkswagen Data:Lab the focus has to be made on talent retention and development.

Connected Car

For the purposes of the present study, the second sub-group identified within R&D will be called “connected car” team.

According to the interviews, it was several times mentioned that Connected Car initiatives are Volkswagen Data:Lab Research and Development future initiatives in terms of revenue generation. As these initiatives will be more integrated with the car, it is expected to have longer Figure 18: R&D - Connected Car

(42)

Therefore, in order to reduce any possible friction, it would advisable to start small and agile to provide quick responses to customers’ demands, and have a customer centric approach.

On the long run, as it happened with the motorsport’s ventures, once connected car projects have reached breakeven point, then it would be time to scale up into the transformation zone.

On short run, the recommendation is to build up an ambidextrous team, this means that those who integrate connected car initiatives will be resources that might be understand and pivot when it is necessary, therefore a high sense of urgency combined with financial acumen would be a must have in the profile.

There are embedded risks on moving forward with these initiatives, such us misplacement of resources, morale reduction, or strategical misalignment. However, if clear and timely communication is made, the friction will be significantly reduced during the transition periods.

(43)

In order to overcome the issues, the deliverables strategy should focus on developing major number of proof of concepts, or minimum viable products, while applying early management techniques to involve business units across the whole development process.

Also, it would be wise to leverage from external resources, because for the particular case of areas such as production and logistics, the power does not reside in the resources themselves, but on the effectiveness to deploy several projects at the same time aiming to optimize the internal value chain.

The model proposes to shape a new efficiency frontier is the hyper-growth scale up, by means of successful implementation of PoC’s, having in projects the sensibility to understand and adapt products to each different region, and create cross cultural teams supported by an internal crowdsourcing platform where best practices are shared.

There are inherent risks related to the production and logistics projects that worthwhile mentioning again, the first one is the extension. Projects or PoC’s that might take longer than 6 months should be revised, and if possible breakdown into sub-units to stay up with the idea of having an agile and responsive way of working.

C. Deep Learning Team

(44)

After the interviews and the assessment being performed, the Deep Learning team represented the biggest challenge when it comes to recommend a scale up path. Based on the information gathered, and market trends, the model that fits the best for Data:Lab’s Deep Learning horizontal is the “Open Crowdsourcing Platform”. There main reason to opt for this choice is the scarcity of resources, and the fact that demand-supply is imbalanced, therefore the capacity to find or retain talent is low. Here follows and abstract from 2017 Boston Consulting Group Report:

“Graduates from data science programs could increase by a robust 7 percent per year, our high-case scenario projects even greater (12 percent) annual growth in demand, which would lead to a shortfall of some 250,000 data scientists”

As per application of the framework proposed in section IV.C, the success factors identified for this particular analysis are:

a) Create value and share, all successful initiatives have to be shared in order to create a green field where best practices and knowledge is shared;

b) Foster interactions and enhance bonds, for instance, intra company networking; or externally to increase the budget to support start-ups. These two activities that help to gain visibility among experts’ community must remain active in order to keep it this way;

c) While it sounds difficult, results from activities performed have to be monetized. The starting point to fulfill the projects pipeline could be to work with a “cost-plus” model. However, as it was already mentioned in previous sections, the value it is given by the solution or “pain” solved to the customer. This shift towards more develop pricing models will also help to reach the financial self-sustainability.

(45)

VII. Conclusion

The outcome of this research aimed to provide support for future strategical decision within Volkswagen Data:Lab Munich. Volkswagen has defined sixteen group initiatives10_{, from} that sample the relevant that requires special Data:Lab attention are:

10. Build mobility solutions;

12. Improve operational excellence 14. Drive digital transformation; 16. Create an organization 4.0;

This means that 4 out of 16 (25%) of the strategic initiatives are data analytics’ driven. However, despite the fact that the digital path is a strategic priority, there is still room to improve for Volkswagen. As per this study reflects, currently the organization can be placed in the boundaries between “Hybrid” and “Decentralized” (see figure 21).

Figure 21: An Emerging Pathway to Digital Organization Maturity (BCG, 2017)

Also, currently companies across the globe report that finding the right talent is the biggest hassle they are facing in trying to integrate data and analytics into their existing operations. In the 2016 McKinsey survey, approximately half of executives across geographies

(46)

and industries reported greater difficulty recruiting analytical talent than filling any other kind of role. Forty percent say retention is also an issue.

In conclusion, Volkswagen Data:Lab Munich has to focus on:

Long-Term Vision: it is undeniable the relevance of the ventures developed within

Data:Lab. A common long-term vision has to be developed to give understanding to all the layers of the organization on what is the “holy grail” to live for. The highly proficient organization in the data analytics fields know what they are trying to achieve with their big data programs, and have organized in such way to accomplish their vision

Execution: brilliant ideas that are never put into practice have zero value for the

customers. The execution power will play a determinant factor when it comes to decide what organization will lead the path of changes. Data:Lab has the advantage to play a fundamental role in 4 out of 16 strategic company initiatives, hence, to excel in projects implementation will be a must.

Information Management: the organization has access to indefinite amount of data,

instead of focusing on the missing points, it should rely on the internal ability to provide solutions to different issues. It is impossible to have a perfect set of information, risks have to be taken.

Platform and skills: execution and data management strongly depend on the

compatibility of the existing structures. A combination of the right technology to be implemented with the adequate human resources has to be considered to steer the technical choices.

(47)

VIII. Appendixes

(48)

(49)

C. Path and Strategies to Scale Crowdsourcing Platforms

(50)

IX. References

A. Literature

a. Community Structure and Collaborative Consumption: "A Routine Activity Approach"; Melson, Marcus; Spaeth, Joe L.; The American Behavioral Scientist; Mar 1, 1978; 21, 4; ProQuest pg. 614.

b. The NIST Definition of Cloud Computing: “Recommendations of the National Institute of Standards and Technology”; Mell, Peter; Grance, Timothy; NIST Special Publication 800-145; September 2011.

c. “Crossing the Chasm”; Geoffrey A. Moore; Harper Business Publication; 2014 revision. d. Escape Velocity: Free Your Company’s Future from the Pull of the Past Moore

e. “To Succeed in the Long Term, Focus on the Middle Term”; Geoffrey A. Moore; Harvard Business Review; July-August 2017 issue.

f. “A framework for deep tech collaboration”; N. Harle, Phillipe Soussan, and Arnoud de la Tour; Boston Consulting Group and Hello Tomorrow article; 2017.

g. “What deep tech startups want from corporate partners”; N. Harle, Phillipe Soussan, and Arnoud de la Tour; Boston Consulting Group Henderson Institute; April 2017. h. “The Age of Analytics: Competing in A Data-Driven World”; McKinsey Global Institute;

December 2016.

i. “The Top Reasons Startups Fail”; McCarthy, Niall; www.forbes.com; November 3 2017. j. “From the Beetle to a Global Player. Volkswagen Chronicle “; Manfred Grieger, Markus

Scale up strategy for SaaS-PaaS ventures