Personal Documents on the iPad using the SAP Mobile BI App

I was chatting with a customer and old friend from the Business Objects User Group , Simon Trill, and he mentioned in his review of BI4 and specifically using WebIntelligence reports on the iPad he was able to do simple security of mobile content using a personal category.

So if you want users to be able to display a WebIntelligence that they hold in their “Personal Documents” or what’s now known as “My Favourites” on an iPad through the SAP BI app then simply create a personal category called “Mobile” as you would do in “Corporate Documents” “Public Folders”, and assign the report to that category.  Simple as that !

I created a simple report called “Sales by State” and added it to a newly created “Mobile” personal category.  As you can see in the screenshot below the report from the personal category is there but also the content in the corporate “Mobile” category is also still visible.

I haven’t done any extensive testing but this approach may be of interest to someone to build upon further.

Question to the SAP Usergroup UK – Data Visualisation SIG … Does the academic view of a dashboard really deliver actionable insight?

At the recent SAP User group UK – Data Visualisation SIG I ran a hands on session based directly on the Stephen Few Dashboard Design Competition of late 2012  which was set to  “showcase for the current state of expert dashboard design”  (Full guidance, rules and specifications  of this competition can be found at http://www.perceptualedge.com/blog/?p=1308).

My objective of this hands on session was to get the SIG attendees discussing a real life scenario in small groups and come up with an initial dashboard design in under one hour.  I’m sure anyone who made an entry into the real competition (including myself) spent many more hours on their submission but I felt confident the attendees could encapsulate their thoughts in sketches, some being more legible than others, in a very short amount of time.

I was asked by Lars Schubert @graphomate via twitter to share the results, so here goes …

How do the SIG’s sketches compare to the Competition Winners and Losers ?

My official entry

The winner

The Runner Up

Stephen Few’s own solution

Observations from the designs

• Hopefully, you can see there is a lot of similarity in the designs,
  • Data grids
    • Common theme not easily readable in the photographs were Top 5 / Bottom 5 style visualisations to help drive the teacher in the scenario to the pupils that need attention that particular day/lesson
• Various types of  graphs from Bar Charts, Sparklines and even a 3D Radar Chart
• Minimal text or commentary
• Distinct regions to  group comparable components and data areas
• In the majority of cases interactivity was definitely part of the design
  • “When you click here ….. ”

Group discussion

• Each sketch was explained to the whole group by the author and in the majority of cases it was said that the brief was not sufficient in the “real world” and they would have liked user engagement to gain greater understanding of the requirements.
• Honestly there was audible gasps when the competition winning design and Stephen Few’s own design were shown to the group.  Both were so very far away from anything the attendees had designed themselves.  Not one person in the room had remotely considered a layout that principally had one row per class member in a sorted list.
• There was another audiable reaction when the competitions “Runner Up” was shown, an attendee said out loud, “That’s Better”.   Certainly a very emotional reaction, but the bigger question is WHY did was the “Runner Up” considered better then the Winner ?   A topic for the next SIG meeting I feel!
• It was generally felt that both the competition winner and Stephen Few’s design had a place in a dashboard solution but they would be 2 to 3 levels down a drill path. It was widely agreed they both were missing a summary to clearly give the teacher in the scenario instruction not only of who they needed to focus on that day, but through predictive business rules indicate who is also at risk of  falling behind in the focus areas

And one more thing …

Unbeknownst to me one of the Data Visualisation SIG attendees had spent a day attending a Stephen Few course on Table and Graph design in London a few days prior to this SIG.  He had mentioned to Stephen that we were going to be using his competition as a hands on session and even offered for him to attend, Stephen declined explaining he was leaving the country before the day.  PHEWWWWW

However, in their discussion my name was mentioned and over their lunch break Stephen dug out my entry and gave an impromptu critique of my design. The overriding theme I believe was that my design was “cluttered” and needed the teacher in the scenario to look around the dashboard to get a complete picture of a student.  I have blogged about the thinking behind my design previously so I won’t explain my thinking again, if you’re interested take a look here http://wp.me/p248fQ-2w

It turns out in the discussion between Stephen and the SIG attendee it was mentioned that the competition scenario was based on a real use case and is was suggested that the teacher was a friend of Stephens and had asked for a one line per child design.  If this is true then it would have been great if this was in outlined in the scenario briefing document, Ho Hummm

SAP TechEd Madrid 2012 – A few personal take-aways

Having spent the last week in Madrid with colleagues from itelligence UK at SAP TechEd 2012 I wanted to reflect on a few things that caught my eye.

1.  Predictive Analysis

I was very impressed to see the integrated version of Visual Intelligence and Predictive Analysis with a simple workflow around Acquire  Predict and Share.  The combined app looks great and by sharing a codebase with SAP Visual Intelligence this should allow not only the development team at SAP to deliver even more even quicker but certainly helps simplify the proliferation of end user tools.  In this case the persona of a Data Artisan can widen out towards a Data Scientist in the one tool, and then share the resulting content using the growing functionality in the Visual Intelligence wrapper.  Practically this could be achieved as simply as the end user of Visual Intelligence clicking a button in the product to expand into a predicative persona and by entering a license key the product will morph into the fully featured Predictive Analysis tool.  But still with access to the growing functionality of Visual Intelligence.  Let’s hope I’m close when Predictive Analysis goes GA soon.

2. Design Studio

As many of you will know SAP Design Studio (Code name ZEN) went generally available during SAP Sapphire 2012 and was a focus in many of the Analytic presentations. The key things I took away from about are very succinctly outlined in the video below by Timo Elliott with Ian Mayor. http://youtu.be/9mzihrcpFik

Simply, if you want to deliver pre authored Dashboards and SAP BW is the ONLY data source then SAP Design Studio should be your first choice.  But contrary to the video my personal opinion is that for pre authored Dashboards over multiple sources including BW and non SAP data sources don’t leap straight to the product called SAP Dashboards.  There are many options including, Web Intelligence and Exploration Views, start with the Use case and Persona and work back from there to select the right tool.

Also, there was a great customer presentation by Mark Cooper from British American Tobaccos (BAT) showing how “Analysis OLAP saved my life”.  When technical issues in a large scale BI4 over SAP BW project kept coming, Analysis OLAP just kept running seamlessly.  At the end of the presentation there was a clear statement that Design Studio is core to their strategy and it is the planned reporting service for “Formatted Reports”, “Analysis” and “Dashboards” from 2013/2014 onwards. A very bold and public statement of direction.

3.  SP05 and the new Mobile BI app due very shortly

SP05 for BI4 was also released days before SAP Sapphire 2012 and there is a great summary blog by Jonathan Haun, so no need for me to repeat the content when Jonathan explained it so well.  http://bobj.sapbiblog.com/2012/11/15/whats-new-in-sap-businessobjects-4-0-sp5

The next version of the Mobile BI App is due very shortly and will house not only Webi and Crystal reports as it does today but also Dashboards (Xcelsius) and Design Studio content.

Something to be aware of is that there are a few steps to take when deploying Dashboards to the Mobile app:

• SP05 for the BI4 Platform and the Dashboard client are prerequisites
• The dashboard needs to be migrated to using the HTML5 versions of it’s components.  There will be a helpful wizard but be aware not all components are available in HTML5 so there may be some rework or redesign necessary.

4. Web Intelligence delivering Mobile Interactive Dashboards

I was sitting in on a hands on session creating analytic content for mobile use (Webi reports on the iPad) when I had one of my infrequent moments of inspiration.  I got carried away and tweeted something like “Web Intelligence on the iPad is another nail in the coffin of Xcelsius”.  I will save the full details for another blog but I spent half an hour between sessions with a cup of tea mapping out a few of my thoughts below:

5. Visual Intelligence

Well, Visual Intelligence just keeps getting better.  During SAP TechEd version 1.0.7 of the product was released.  To me there a few key advancements …

1. UNLIMITED DATA

Removal of 15m cell limitation at acquisition time for offline sources
2. 32 Bit support

Not my personal favourite but I’m assured by a number of people I have spoken to that this is very important going forward.
3. Edit an acquired data source

Sounds simple but you can now add or remove columns from an acquired data source by selecting Data > Edit source.

4. New data type conversion option for Manipulation Tools    Available for number and text data.

So much easier than writing custom formulas for the end user.

6.  Meeting up with old friends and making new

To be honest the most fun I had this year was catching up with old friends from when I worked at BusinessObjects as it was then and especially meeting face to face so many people I engage with on Twitter.

Thanks to all the following for taking the time to chat.

• Mani Srinivasan  @mani_srini
• Lars Schubert  @graphomate
• Roger Mathias  @rogermathis
• Mark Cooper @McBobJ
• DJ Adams  @qmacro
• Tim Guest  @TimTheGuest
• Nick Wall  @nickwallSAP
• Varik Torsteinsen @varikt
• Martin Walker
• Paul Barker

Stephen Few 2012 Perceptual Edge Dashboard Design Competition

Back in August this year Stephen Few opened up a competition to Design a dashboard using a defined data set for several purposes:

Quote”

• A showcase for the current state of expert dashboard design.
• An opportunity for me to use the submissions to teach best practices by critiquing several of them on this website and in the second edition of the book Information Dashboard Design, which I am currently writing.
• An opportunity to provide sample dashboard designs that could actually be used to improve the quality of education in schools, for this competition involves the design of a dashboard that could be used by teachers to monitor the performance of their students.”

Full guidance, rules and specifiactions  of this competition can be found at http://www.perceptualedge.com/blog/?p=1308

I wanted to share my thoughts and design process for discussion .

Design Process

Thoughts

Design Ideas

My Competition Submission

I believe the definition of the competition as defined in the Excel spreadsheet can be summarized as:

Design a dashboard that would be used by a high school teacher to monitor student performance. The particular person who will use this dashboard will ordinarily look at the dashboard prior to each class session to update awareness regarding the students in that particular class in an effort to prevent or resolve problems and to help each student improve as much as possible.

Information has been provided for a single Algebra 1 class, along with some summary information regarding the entire school and the school district, which can serve as context for evaluating this class’ performance. Concern yourself with the design of the initial dashboard that this teacher could use to rapidly monitor what’s going on with her students in preparation for the day’s class.

After reviewing this brief I approached the challenge from the perspective that as the user will be engaging with the information daily, and therefore I want them to be alerted to “Which Students need focusing on today”.

I am fortunate enough to have attended Stephens’s workshops in London a number of years ago and he critiqued one of my early dashboard designs in front of the class.  I do subscribe to the majority of his visualization techniques and have used this insight to build out a set of key themes which should drive an Action.

I have  laid out the dashboard in a number of quadrants:

1.1     Quadrant One

• Students to focus on TODAY
  • Attendance
  • Behavior
  • Achievement

The primary focus of this quadrant is to clearly identify the students that the end user needs to focus on that day.  This quadrant is divided into 3 areas, Attendance, Behavior & Achievement and laid out in a way to allow to allow easy identification of the students that need focus (Red traffic Light) and identify commonality of students over multiple areas

1.2     Quadrant Two

• Attendance

5 Worst Performers This Term

• List of the Bottom

5 students in the areas of Days Absent & Days Tardy with a Class Average number to allow comparison

• Behavior

5 Worst Performers This Term & Last Term

• List of the Bottom 5 students in the areas of Discipline Referrals & Detentions with a Class Average number to allow comparison.

1.3     Quadrant Three

The primary focus of this quadrant is to clearly show students who are currently below the class average in the particular areas with a trend to show if the current score is linked to a declining trend and thus be a cause for concern.

The subtle highlighting behind the Sparkline starts at the data point of the Average score up to a maximum of 100%. This technique allows the end user to quickly assess that even if the student is listed as below average at one point in the time series displayed is there a cause for concern.

• Achievement – Below Average Students

Students listed all have a below average current math’s assessment score

• Standardized Math Assessment Score

Trend of math’s assessment scores from the 6th Grade to 9th Grade and the Latest score.  All displayed in %

• Assignment Scores

Trend of math’s assignment scores from assignment 1-5.  All displayed in %

• Achievement – Declining Students

Standardized Math Assessment Score

Trend of math’s assessment scores from the 6th Grade to 9th Grade and the Latest score.  All displayed in %

• Assignment Scores

Trend of math’s assignment scores from assignment 1-5.  All displayed in %

1.4     Quadrant Four

The primary focus of this quadrant is to clearly show students who are below their personally set Course Grade Goal.

The use of a bullet chart and data table allows the user to easily identify students currently running behind on their course grade goal and their previous performance.  All displayed in %

• Achievement – Students with Course Grade Off Goal
  • Current Score, Goal and Previous score

This button will navigate the user to a different interface offered by SAP BusinessObjects to allow them to free format explore the data set via a web browser

This button will open a Help screen offering a glossary of terms, contact points and other pertinent information

1.5     Technology Used

SAP BusinessObjects Dashboards, formally known as Xcelsius

And the winner is …..

The winner of this contest can be seen with a full explanation at   http://www.perceptualedge.com/blog/?p=1374#comment-296665

Great design but my only comments are as follows:

1.  Can you really gain actionable insight in 5 seconds of reading ?

2.  Can this design become a reality outside of photoshop ?  The winner replied to my question about technology as follows:

Quote ”  @Andrew: as I mentioned, I am really lucky that I work with some super talented programmers who do the real heavy lifting in taking my designs and making them work. This can be in a variety of technologies, from .net to apex to sharepoint to straight html5/jquery. I’m just happy I don’t have to do it, and am amazed that they can!”

SAP Data Geek Challenge 2012

Nick Smith @Nicfish set out a Ultimate Data Geek Challenge at the beginning of September to “Let the rest of the world know your data skills”.  http://scn.sap.com/docs/DOC-31008

Well, my data skills aren’t the greatest mathematically so I wanted to use this challenge to push my understanding of SAP Visual Intelligence a few stages on…..  It sure did that.

I decided to use the data set of USDA Nutrient Data , which after a bit of googling is the  United States Department of Agriculture  Nutrient database that holds the nutritional information of numerous food and beverage products.

Step 1 – Where do I start ?

Well with Visual Intelligence (VISI)  it’s extremely simple to acquire data for analysis and chop out columns  you are not interested in.

Another great feature is the inbuilt enrichment of data, in this data set it is only used to create “Measures” but it is very useful for both Geographical and Date enrichment. It will in fact build out a Time and Geography hierarchy from a City or Date dimension field.

Step 2 – Have a look around the dataset

In looking around the data set it was easy to notice that a lot of information in terms of a product Hierarchy is held in the Product Description field but separate by the ‘Comma’.  Traditionally this would be a nightmare for a Web Intelligence report developer to work with and would often need a service request on the DBA to split out the field into multiple fields.

In Visual Intelligence this is an easy user task and enables them to really take a ghold of their data, Split column by <Comma>.

With a few moments of renaming  the Description can be split out into multiple columns to aid analysis.

Step 3 -  Visualising to aid analysis

There is only so much analysis you can do by eyeballing a many thousand row spreadsheet so this is where visual analysis really aids understanding. If this Data Geek Challenge experience is anything to go by it really showed me that questions lead from question and take you places in the data you never thought you’d end up.

Question 1 - What food group is “Bad for me”?

Question 2 - Just How Bad ?

Spotting the outliers in a bubble chart really help understand to exceptions

But spin the axes around an a different picture forms

Question 3 – What should I really not eat?

We all know Sugar is not great for you but it seems on first look to be intrinsic to what I consider to be “Bad for me”  but where should I try and lower my intake in my regular diet ?

I’m not a big candy (sweets) eater but as any parent there are loads of different boxes of breakfast cereal in the house.

Should I be worried my 4 year old son has Cheerios every day for breakfast ?  It looks like yes

Question 4 – What is the worst thing to eat for breakfast?

Well, how about Cheerios with instant tea instead of milk

Question 5 – What should I eat for breakfast ?

What’s high in calories but low in sugar and I should really consider eating for breakfast?

Wheat with corn beverage ?

  

Appetising ? ….  Maybe not

Summary

Well did I think I’d end up banning my youngest child from eating Cheerios as a result of this challenge?  Absolutely not.  But this challenge has certainly enabled me to get close to SAP Visual Intelligence and appreciate more that getting closer to data and putting the analysis in the hands of dare I say Data Scientists, well analysts can only be a good thing.

 

Part 2 – The Reprise

I owe a lot to John Appleby @applebyj and @Ayooshha at Bluefin Solutions for getting me started in blogging there encouragement and dare I say persistence changed my attitude to social media.  I have mentioned before in my blog that Johns 10 tips to getting started in blogging did inspire me.

This week I had to get to grips with number 11 quoted by @BoobBoo

11. Do not be afraid to be wrong, people will challenge you but if you have passion, good grace and knowledge the conversation will most likely be rewarding and informative.

Yet again, great advice as the conclusion in my blog is fundamentally wrong and it was kindly pointed out to me by Ethan Jewett @esjewett

Do you realize what you did in this exercise? You added up sugar for several different types of cheerios.  Regular cheerios only has about 4g of sugar. You added regular, banana nut, yogurt burst, and chocolate varieties together. Did the same for several other types of food as well, and for the food categories in the bubble charts….

And yep, Ethan was bang on right.

I made at least three fundamental mistakes

1. I assumed Cherrios was a product not a Brand.   In my house my youngest son eats Cheerios, I had no idea that various products were made including Banana Nut, Chocolate and Yoghurt.
2. I didn’t drill down to the lowest level of granularity, if I had I would have seen the individual products individual data values and not the summated amount at Brand level (diagram below)
3. I didn’t validate my conclusion.  In haste I didn’t stop, think and validate.  A good lesson learnt.
4. Get the source data right.  In breaking out the label field the way I did there isn’t a consistent hierarchy, Level 3 for one product maybe in level 4 or 2 for another.

So hopefully I have used the SAP Data Geek Challenge not only to deepen my understanding of SAP Visual Intelligence, but also see a new side to the value of blogging, open conversation and the benefits of peer review.

And one more thing, have a play with the data set yourself and let me know what I shouldn’t be eating for breakfast !

 

Download the data set   Usda_Nutrient_Data

Download SAP Visual Intelligence

 

 

What is a Dashboard really ?

I recently presented at the UK&I SAP User group Data Visualisation SIG (sapusers.org) on the topic of “What is a dashboard really, and how to select the right SAP tool for the job” and I wanted to share the content to a wider audience.

What is a dashboard Really ??

I believe that when individuals use the term “dashboard” they can very often have different ideas and interpretations as to what they mean.   In my experience these opinions can be as wide apart as a one pager with a few Key Performance indicators (KPI’s) to what you could call a fully featured Enterprise BI Application with Drill, Slice & Dice and Printing

What’s your definition of what a Dashboard Really Is ?

Wikipedia:

• “An easy to read, often single page, real-time user interface, showing a graphical presentation of the current status (snapshot) and historical trends of an organization’s Key Performance Indicators (KPIs) to enable instantaneous and informed decisions to be made at a glance.“

Dictonary.com

•  document presenting the most significant information about a subject on a single page

SAP

I recently saw this interpretation at an event I attended …   A DASHBOARD is…

• A visual display of…
  • Information for strategic decisions down to operational for daily decisions
  • Multiple dashboards can be linked based role to provide a storyline
• Created for the purpose of…
  • Alignment of organization to objectives
  • Monitoring of KPIs
  • Decision making and actions to improve business results

What definition does your company agree on ?

Why not try the twitter approach and define a dashboard in less then 140 Characters.  This can be harder than you think but the academic view from the author Stephen Few (perceptualedge.com) is only 34 words but 163 characters

A visual display of the most important information needed to achieve one or more objectives that have been consolidated on a single computer screen so it can be monitored and understood at a glance.

For those of you that know me personally will have heard me refer to myself as a “ Bear of Very Little Brain, and long words bother me.”  (A. A. Milne, Winnie-the-Pooh ) so I have tried to simplify this further

A display of “Grids and Graphs” that have been attractively displayed so they can be read and action identified in seconds.

By way of explanation I mean the following when I say “Grids and Graphs”

What is Right tool for “Grids and Graphs”

In the PPT presentation at the end of this blog I have screenshot lot’s of different dashboard style visualisations that were created and presented using SAP Business Analytic tevhnologies including Web intelligence, Explorer, Exploration Views, Design Studio (Zen) and Mobile BI.

You could say each of the visualisation examples meet my simplistic dashboard definition.  So how many tools does SAP BusinessObjects offer that can deliver a “Dashboard”  I counted at least 5, but if we include the SAP BW technologies we can increase that number even more

What to do when you are asked for a dashboard

In a previous blog I shared my thoughts on Kicking off a Dashboard Project and I have summarised a few of them here

• Always Ask WHY a dashboard is needed
  • Keep asking Why, Why, Why, Why, Why until you drill down to the real requirement
  • The key point for me is always ask what ACTION are they going to take as a result of using the dashboard
• Always Ask for “Mission Statement”
  • A dashboard to XXXXX
  • Then start asking around the following areas, as it will drive out the most appropriate technology to deliver the defined dashboard.

So my take away thought is ..

When thinking about Dashboards don’t limit yourself to the SAP product called Dashboards

What Is A Dashboard Really : Link to presentation

Are you interested in attending a gathering of individuals interested in Data Visualisation and Dashboards?

I have been pondering for a number of weeks in seeing if there is any value in getting together many of those who engage in debate, thought leadership and knowledge sharing through social media  in the areas of Data Visualisation and Dashboarding in a room together.

I have been greatly encouraged by interest by both potential attendees and corporately.  However, as you’d expect the first questions asked are along the lines of “What is the value to the attendee”, “Why would someone travel from far afield to Europe for such an event”, “What is the value to the hosting organisation”

To help me with building out my idea I would appreciate it if you could fill in a short 10 question survey I have put together.

http://www.surveymonkey.com/s/NKLT7WV

If you want to add any more comments for wider debate that the survey misses please feel free to comment via this blog site.