Gartner Business Intelligence Summit February 2012

Gartner Business Intelligence Summit February 2012
http://www.gartner.com/technology/summits/emea/business-intelligence/

I had the pleasure of attending this intense and slick two day event on Business Intelligence and Analytics. My personal goals were to get up to speed quickly by immersing in the world of the BI Suit and for that I take my hat off to Gartner. Colleagues and I have developed a Business Intelligence InfoKit http://www.jiscinfonet.ac.uk/bi based on solid national research. Field tests conclude 31 August 2012 and we’ll refine and improve the resource soon after. It highlights Predictive Modelling as a feature of the highest maturity level for BI capability and so Analytics is an area of interest to JISC. Indeed we’re about to commission a resource to explode this area with the aim of moving the Higher Education Sector on in this area. Why? Well, analytics undertaken on structured, semi or unstructured and Big Data is already allowing accurate forecasting to help effectiveness, agility and efficiency. Who could say no to that.

Below are my bulleted notes on the sessions I attended. Rather than unpack these as a story of analytics I’ll leave them raw and link through to that new piece of work I allude to as a developing rich resource to really set the scene on what Analytics can do for Higher Education.

1. Excellence Awards
a. Spanish Company
i. What
Multi million turnover, 40 FTE
ii. How
Multiple Datamarts are faster to implement than a data warehouse
A full set of BI dashboards via IBM Cognos 10 demonstrated without detail.
BI Competency Centre
iii. End Game
Performance measurement

b. Medway Youth Trust
i. What
Demographics data mining and analytics for youth unemployment
ii. How
IBM SPSS Modeller allows extraction and combination of structured and unstructured data prior to modeling and visualization
iii. End Game
Informs resource planning and service prioritization
BI for Social Benefit

c. TFL
i. What
Smoothing traffic flow on roads using BI
ii. How
Strategic pressure to Innovate
Cameras, roadwork permits, buses, traffic sensors provides big data, data rich
Data Service via Oracle BI Enterprise with ESRI Arc GIS Server
No manual data entry
Delivery via mobile devices
iii. End Game
Data derived intelligence at the right time
Performance measurement
Target resources to achieve targets via KPIs
Culture shift; Data as burden becomes Data as asset
Single source of truth
Customers self serve
VFM – automated production of BI vs Manual

2. Premier Sponsor Panel – nice format
a. The key requests from customers
i. Desktop efficiency
ii. Speed wrt can I do more modeling in the same timeframe
iii. Making analytics more pervasive (undergrads with machine learning backgrounds)

b. Failures in analytics investment
i. Define the problem space to avoid a random walk into failure
ii. Make sure you’re going to cause a business action from your deliverable. Make a difference to working practice, step beyond ‘that’s interesting’

c. Problems in supplier offers
i. Cover the entire analytics maturity spectrum
ii. Tie BI and advanced analytics to achieve
iii. Focus less on report production (they don’t result in actions resulting in change)
iv. Achieve the right ‘services partner’ as well as technical solution
v. Service should offer big data benefits to the consumer

d. What’s the next big problem
i. Identify relevance of data to avoid drowning in it for internal purposes as well as customer base

e. Horizon queries
i. Will in-memory become dominant technology for BI?
ii. Will prevalence of embedded analytics in business applications mean that standalone BI capabilities will become less important?
iii. Will data warehouses that fail to include content analysis support be retired?
iv. Support for mobile delivery will be standard
v. Open source BI will become a significant force in the market

f. Workshop on communication
Excellent practical session on communicating and presenting

g. Social Media and data mining
i. Why analyse social media?
1. Reduce support costs
2. Better engage with customers
3. Influence outside customer base
4. Competitive intelligence
5. ID underused expertise
6. Brand protection
7. Market research
8. Public relations
9. Etcs

3. Technical Insights: Develop a Big Data Behavioral Analytics System for Business Value
a. What
How does behavioral analytics deliver value to the organization
What are components of big data system
Which data processing designs deliver best
What role for Hadoop and noSQL databases

b. How
Using activity data to create predictions on big data samples
Big data is semi structured and non-uniform, great in volume, see Gartner diagram
Big data cannot be databased

c. Why
Use behavioral and statistical analysis models to run predictions such as;
Did products meet client expectations?
What products will customers buy next year?|

4. Extreme Data — Large-Scale Data Warehousing, BI and Analytics
a. Unpacking headings within the Gartner Big Data diagram; Velocity, Volume, Variety, Complexity

b. Data QA headings; Perishability, Fidelity, Linking, Validation

c. Extreme Information Management and New Applications Drive New Opportunities; Predicting furture market trends ahead of the competition

d. The top business opportunities are around social software for sentiment analysis, combining operational tech monitoring and metrics with logistics and planning and targeting a high fidelity data strategy

e. Interesting slide as schema of extreme data system topology as could be

f. Discussion around the failures in existing data warehousing to address the former as Velocity, Volume, Variety, Complexity

g. Extreme Information Issues wil demand a different approach to Information Management Capabilities

5. Financial Analytics

a. Consider what a finance department does; Transaction Processing (50%), Stewardship (25%) and Strategy (25%)

b. Financial data is viewed hierarchically and highly structured so fits well with relational databases

c. They need real time ‘aged trial’ (telescoped so a bit like Time Machine) data built from transaction data accurate to a half day so no need for real time view

d. Analytics capability requires more than Excel can provide

e. ERP applications have been the major initiative in Finance over last few years

f. ERP has no standard definition. It has a number of strengths but is complex data model, reporting tools are adequate at best and are hard core, the information is locked in thus external data is excluded.

g. An ERP system cannot do budgeting or forecasting, for that you need a CPM suite

h. Vocabularies are a major issue as finance crosses business domains

i. Need an Enterprise Performance Management Centre (better marketing than a BI Competency centre, more likely to get executive involvement such as CFO and CIO) doing Business Analytics (see Gartner Business Analytics Framework)

j. Emerging trends – business modellling, predictive analytics and simulation will be massive in 5 years, less so for mobile, less so for Big Data

k. Gartner survey 430 CFOs annually about top technology interests – top is BI and Analytics

l. Recommendations
Understand different types of financial analytics tah finance needs to fulfil its different roles
Need multiple analytic capabilities
Educate finance in trends in analytics
Build a road map including mobile, predictive analytics and simulation
Find the spreadsheet jockeys – these are the pwer uses to get involved
Draft a business analytics strategy and take to the CFO
Build a joint IT/Finance working group and prepare a joing nalytics road map
Implement at least one analytics to handle a financial need

6. IBM Lunch
a. Good coverage here. Lunched with an Information Architect for a bank and a DPA expert. The guy from IBM discussed Analytics as undertaking user needs analysis, then a matrix of desirability vs difficulty, prioritizing based on those, then identifying the data streams needed, then considering information management issues, revisiting the matrix to determine whether the candidate is still viable and only then undertaking analysis and modeling. IBM don’t buy into the Gartner Big Data slide. Big is in the eye of the beholder. By following the IBM steps only at the point of analysis do we encounter issue of data size. They will check to see whether they’re in touch with JISC.

7. Magic Quadrant Power Session (February 2012)
a. BI Platforms
It’s a mature Market.
Leaders; MS, SAS, QiKTech, SAP, IBM, Oracle, Microstrategy
Niche Players; LogiXML, Actuate, Prognoz, Panorama Software, Salient Management, Board International, Arcplan, Targit, Alteryx, Pentaho, Jaspersoft
Challengers; Tableau, Tibco Spotfire
There’s been a lot of talk about analytics capabilities being built into best of breed services

b. Data Warehouse Database Management Systems
Leaders; Teradata, Oracle, IBM, EMC/Greenplum, Sybase, MS
Visionaries; Vertica
Challengers; 1010data
Niche players; ParAccel, Kognito, SAND Technology, Infobright, Actian, Exasol
Over next 3 years a rise from 3% – 15% in logical data warehouse
There was a depression in terms of big data (volume, velocity, variety)

c. Corporate Performance Management
Leaders; Oracle, SAP, IBM (bought the technologies, didn’t develop them)
Visionaries; Exact-Longview, Tagetik
Niche Players; Board International, Prophix software, Host analytics, Bitam, Winterheller, KCI Computing
Challengers; Infor, SAS

d. Master Data Management of Product Data Solutions (supports information governance apparently)
Leaders; IBM, Oracle
Visionaries; SAP, Tibco, Riversand
Niche players; Stibo Systems, Hybris, Heiler, Enterworks, Informatica, Orchestra Networks
Challengers; None

e. Data integration tools
Leaders; Informatica, IBM, SAP, Oracel, SAS/Datafox
Visionaries; iWay software, Pervasive software, Talend
Niche Players; Syncsort
Challengers; MS

f. Data Quality Tools
Leaders; DataFlux, Informatica, IBM, SAP, Trillium Software
Visionaries; Talend, Attacarna
Challengers; Pitney Bowes Business Insight
Oracle
Niche Players; DataLever, UniServ, Datamantors, datactics, Innovative Systems, Human Interface

g. Marketscope (no Quadrant as this is too new) for BI and Information Management Services in Western Europe (too many to list)

Leave a Reply

Your email address will not be published. Required fields are marked *