This document discusses data products and how they differ from traditional products. It defines three types of data products: data as a service, data-enhanced products, and data as insights. The document highlights that data product management may require different processes and methods than traditional product management. It provides examples from interviews with companies about how they approach aspects like defining value propositions, positioning, and identifying data for data products. The conclusion is that data products are different and require categorizing them to help with business modeling and product definition.

1. What are data products and
why are they different from other products?
Christoph Tempich, Thomas Leitermann
www.inovex.de
Many thanks to Aaron Humm and Florian Tanten
for carrying out the interviews 18.5.2017
#Datenprodukte @ctempich @thomasleiterman

2. 2
What do these people have in common?
#Datenprodukte @ctempich @thomasleiterman

3. 3
What do these people have in common?
They build their businesses around “data products”.
#Datenprodukte @ctempich @thomasleiterman

4. 4#Datenprodukte @ctempich @thomasleiterman
Growing market for „data products“

5. 5#Datenprodukte @ctempich @thomasleiterman
Hypothesis
Data product <> traditional product
› Should the product management process be adapted to the special needs of data
products?
› Is there a need for new product management methods or should product managers
apply existing methods in a different way?
Research foundations
› Interviews with employees from Otto, Xing, mobile.de, Eventim, … (n ~ 10)
› Roles of the interviewees: Product Owner, Product Manager, Data Scientist
› Bachelor thesis written by Aaron Humm at Technische Hochschule Stuttgart
› ... Experiences from our projects ;-)

6. 6
Definition of data products

7. 7
Data products: types
Data as a Service
Data-enhanced
Products
Data as Insights
Type 1 Type 2 Type 3
› Autonomous driving› Weather data › Marketing planning
#Datenprodukte @ctempich @thomasleiterman

8. › Minor differences in the interpretation of insights
› Distinction in general is seen as very helpful
8
Interview results
Differentiation between types of data products
Recommendation Definitely useful, as the distinction supports the explanation
of different revenue sources. Furthermore it supports the
creative process when searching for new ideas for data
products (product discovery)
!
#Datenprodukte @ctempich @thomasleiterman

9. 9
Data products
Processes and methods

10. 10* See http://www.pro-produktmanagement.de/open-product-management-workflow.html
Data product –management process
3 phases* Strategic
data product management
› Identify
› Analyce
› Check
› Strategy
› Consolidate
Technical
data product management
› Build Team
› Delivery
› Control
Go-to-Market
› Build Team
› Plan
› Maintenance
Current
research
focus

11. 11
Identify problem
Problems of
any kind
are in scope
Restricted to value drivers with
›rational utility
›social utility
Classical product management Data product management

12. 12Quelle: Laura Dorfer: Datenzentrische Geschäftsmodelle als neuer Geschäftsmodelltypus …, 2016.
Value proposition and readiness to pay
Readiness to pay grows with value of supported decision
User Buyer
Social interaction
Entertainment
Curiosity
Decision support
Transparency
Information
Readiness to payValue proposition

13. › All interviewees have data products in place which build on the decision
support value proposition
› The social aspects of data are less systematically analysed
› As a motivation to trigger a feedback loop social aspects are currently
not systematically deployed
13
Interview results
Value proposition
Recommendation Both, rational and social, value propositions should be
considered. The later is particularly helpful to foster user
interaction with the service.
!
#Datenprodukte @ctempich @thomasleiterman

14. 14
Positioning
Customer Value Proposition
oriented
Add-on: Service provider becomes
customer of the user (Inversion
of the value proposition
perspective)
Classical product management Data product management

15. 15* Osterwalder A.: https://strategyzer.com/books/value-proposition-design
Part of the Business Model Canvas
Find a USP: Value Proposition Design
Value Proposition Design*1

16. 16
Master data
Transaction data: basis for the USP
Transaction data
Bsp.: Google Maps

17. 17

18. 18* Osterwalder A.: https://strategyzer.com/books/value-proposition-design
Method proposal
Inversion of the Value Proposition Canvas
Value Proposition Design*12

19. › User interaction data with a service are currently not systematically used in
order to define the value proposition of the product
› A/B-tests are a first step
› The inversion of the Value Proposition Canvas was discussed controversially
› No common agreement on separation of master data and transactions data
19
Interview result
Feedback Loop, USP and positioning
19Recommendation It is sensible to reuse the user interaction data in order to
enhance the value proposition of the product. The inverted
Value Propostion Canvas can support this process.
!
#Datenprodukte @ctempich @thomasleiterman

20. 20
Data Security
›Personally identifiable
information,
›Country specific laws,
›Ensuring availability, etc.
Classical product management Data product management

21. 21
Identify Data
›Identify relevant data objects
›Evaluation of internal available
data sources and identification of
gaps
›Definiton of the data creation
strategy (Buy, Build, Partner)
Classical product management Data product management

22. 22
Further processes with a need of adoption
› Pricing Strategy
› Portfolio Strategy
› Market Strategy
› Identify Persona

23. › Categorisation of data products suppots the business
model and product definition
› Distinction between different value drivers for data
products supports defining the best customer approach
› Product managers currently to not actively design
feedback loops. Methods are still under construction
23#Datenprodukte @ctempich @thomasleiterman
Conclusion
Data products are different

24. Vielen Dank
inovex GmbH
Ludwig-Erhard-Allee 6
76131 Karlsruhe
Weitere Standorte: Hamburg, Köln,
München, Stuttgart
Dr. Christoph Tempich
@ctempich
Thomas Leitermann
@thomasleiterman
www.datenprodukte.de
blog.inovex.de
www.inovex.de

25. ^
Dr. Christoph Tempich
Mobile
Development
Dominik Helleberg
Portal
Development
Peter Dimitri
Project
Management &
Quality Assurance
Max Wippert
Application
Development
Tobias Joch
IT Engineering
& Operations
Matthias Albert
Data Management
& Analytics
Patrick Thoma Collin Rogowski
Operations
Daniel Bäurer
DevOps
Engineering
Alexander Pacnik
Hosting
Nils Domrose
Big Data Solutions
Dr. Stefan Igel
BI Solutions
Stefan Kirner
Data Science
Dr. Lars Perchalla
Search
& Text Analytics
Product Discovery
Product Ownership
Datenprodukte
Agile Audits,
Trainings
und Coachings
Technologie-
Trainings
Lean Product
Development
Product Discovery
and Ownership
inovex Academy
25