Pre hi tec_innovationsforum_20111103-public

MANAGING INNOVATION AS A PROJECT COMPANYNOVEMBER 2011

HOLGER BLANK, SINNERSCHRADER TECHNIK@HBLANK

SINNERSCHRADERINNOVATION^3BONUS

SINNERSCHRADER AUF EINEN BLICK

Top-10 Digitalagentur in Europa (Forrester Digital Agency Review 2011)

Fokus auf E-Commerce – seit 1996

400 Talente mit Split Beratung/Kreation/Technik 25/25/50

Standorte in Hamburg, Berlin, Hannover und Frankfurt

Inhabergeführt, unabhängig, börsennotiert

400+ SPEZIALISTEN AN BORD

Technology(45%)

Hosting & Operating(5 %)

StrategicPlanning

(5%)

Web Analytics (5 %)

Account Management(15%)

Creative(25%)

SINNERSCHRADER GRUPPE

CRM

E-MAIL MARKETING

CONTENT PRODUCTION

E-COMMERCEFULL SERVICE

DIGITAL MARKETINGPROGRAMS

DIGITAL MARKETINGPLATFORMS

WEB DEVELOPMENT

TECHNOLOGY

DIGITAL STRATEGY

ANALYTICS

NEW SCREENS

ADVERTISING

PERFORMANCE MEDIA

mediaby

TARGETINGMEDIA TECHNOLOGY

DIGITAL MARKETINGCAMPAIGNS

GROUP

UMSATZ/ERGEBNIS-ENTWICKLUNG 1996-2011

96/97 97/98 98/99 99/00 00/01 01/02 02/03 03/04 04/05 05/06 06/07 07/08 08/09 09/10 10/11

30,9

23,9

20,9

18,3

14,213,2

12,210,811,3

13,0

15,7

11,4

4,0

1,50,3

EBITA in Mio. €

Nettoumsatz in Mio. €

AKTUELLE KUNDEN

AKTUELLE KUNDEN

Der Umsatz unsererKunden beträgt 2011 mehr als 2 Mrd EUR

im E-Commerce

BE LOCAL - ACT GLOBAL

1.300 Domains der Public Websites der Deutschen Bank mit 100+ verschiedenen Applikationen werden von uns 24/7 global betreut.

Konzeption und Entwicklung des globalen Marktauftritts der Allianz in 75+ Märkten.

Globaler Roll-out und Betreuung der E-Commerce-Aktivitäten aller Luxury-Brands von PPR.

SINNERSCHRADERINNOVATION^3BONUS

3 INNOVATIONSFELDER

Methoden und Arbeitsweisen

Eigene ProjekteRadical Innovation Lab

Kundenprojekte

3 INNOVATIONSFELDER

Methoden und Arbeitsweisen

Eigene ProjekteRadical Innovation Lab

Kundenprojekte

BOSCH

ARCHITECTURE FOR 10^X3d-scaling for real world problems

OUR TEAM

Markus Wolf, @KnisterPeterSenior Developer, SinnerSchrader

27

Meno Abels, @fastandfearlessDirector, SinnerSchrader

TALKING ABOUT ARCHITECTURE

Choosing an architecture has impact in several dimensions

Development

Deployment/Scalability

Maintaince

Maintenance

We want to sell our store-specific range of goods to our customers online

Customers should be able to collect these items within a appointed timeframe

In other words: “Click & Collect”

Sounds easy.

OUR BRIEFING FROM

FIRST APPROACH

WAIT: STORE SPECIFIC?

PERSONALIZED?

2*10^4 products4*10^2 shops (potential for up to 3*10^3)

... and adding:_unique prices and product-attributes per shop_subset catalogs for specific product lines

(e.g. Bio/Rewe-Feine-Welt/Angebote)_personalized catalog

We need to have as much webshops as real shops providing this service.

WHAT ELSE?

No standard shop software will meet these requirements

THE ESTABLISHMENT WILL FAIL

Accept the challenge and think yourself

Use memory extensively=> around 10^6 faster than disks

SCALING 1DHardware

Use stateless components=> less error prone multithreading

SCALING 1DHardware

Use parallel algorithms=> leverage multicore cpu

SCALING 1DHardware

Use view optimized Business Objects=> no transformation for view rendering

SCALING 1DHardware

=> more requests per Server Instance (approx. 100 per second)

SCALING 1DHardware

SCALING 2DServer Instances

“Stateless Statefullness”_No server-side state_All state on client

"Kind of" SOA_Simple Components for fine grained logic_Messaging between Components and Services_Multithreaded Asynchronous Service Calls

Each server instance can answer all questions without synchronizing any server state between instances.

This means easy load-balancing=> we’re cloud-ready!

SCALING 2DServer Instances

SCALING 2DClient Computer

Scale to the Client_Use Client-CPUs_Calculate display logic

and maybe some business logic

Less server cycles_Gaining performance idependent from

server side_Green-IT

SCALING 2DClient Computer

Use client storage _Persist private data into browser_Caching of product data

=> Less requests, faster user experience, happy users!

SCALING 2DClient Computer

REALITY CHECK - SIMPLE CASE

ENTERPRISE TOOLCHAIN

FUTURE PROOF

Server-Software lines of code <2*10^4 (common > 5*10^4)

Client-Software lines of code < 10^4(no one should need more)

Development Effort approx. (5*10^1) % (compared to traditional approach)

Deployment < 5 Systems**Cost of operations per year is C.H.E.A.P.

Cost per user/year is roughly about:

RESULTS

KLEINE FEATURES

3 INNOVATIONSFELDER

Methoden und Arbeitsweisen

Eigene ProjekteRadical Innovation Lab

Kundenprojekte

DIGITAL FOOSBALL

Slide 60

Slide 61

Slide 62

Slide 63

DIE SOFTWARE

Slide 64

Slide 65

Slide 66

Slide 67

Slide 68

START

Slide 69

CHOOSE YOUR GAME

Slide 70

NAMES PLEASE!

Slide 71

END OF GAME

Slide 72

LIGA IST WICHTIG

Slide 74

SPREAD THE WORD: OPEN SOURCE

Slide 76

BLOGOSPHERE UND TITELBLÄTTER EROBERN!

Slide 78

3 INNOVATIONSFELDER

Methoden und Arbeitsweisen

Eigene ProjekteRadical Innovation Lab

Kundenprojekte

Video iPad Tag?

Slide

Me

_Software Developer at SinnerSchrader_JavaScript Victim Extraordinair_http://nonblocking.io_@cramforce

85

UNERWARTET KOMMT OFT

1996POP3 (Mail only)

2000Lotus Notes2002IMAP + Scripte2004Exchange 2003

2010 ???

Mail Any Device, Sharing, Spam, Viren, Push, Archiv, Backup, Sign&Crypt

Kalender Sharing/Delegation, Attachments, Ressourcen, Recurring, Free/Busy

Kontakte Sharing, Sync

Requirements

NEU ORIENTIERUNG

Shortlist

NEUE WEGE GEHEN

SINNERSCHRADERINNOVATION^3BONUS

MAKE WIFI SOCIAL

Take Aways

IN EINEM SLIDE

Innovation als Dienstleister ist eine Herausforderung(selbst in unserer Branche)

Projekte sind eine gute Basis

Innovation ist eine Frage von Haltung und Kultur

Keine Innovation ist Stillstand

SINNERSCHRADER OPEN DOORS

Open DoorsWann: 11.11.11

Was: _Agenturführung _Gemeinsames Mittagessen _Vortrag und Diskussion mit

Manuel Stolte, Strategic Planning

Noch mehr Infos bekommt ihr bei:[email protected]

oder:radicalmonday.detwitter.com/sinnerschrader

VIELEN DANK.