Oz Analytics

Governance: Influence versus Action?

Information and Data Governance Forum

Sydney 14 - 15 December 2010

At short notice, I was invited to come along to this forum and was able to attend most sessions on the first day. I made space in my day because I was curious about the Information Governance community in Sydney.

My day-to-day role is in the broader field of Information Management. A subtle distinction to an outsider, but in practice I find there is a world of difference. Sceptical? I understand, but let me try to convince you. I'm sure that I will also offend people, but what the hell, life's too short to be agreeable all the time ...

Influence versus Action

The forum presentations and discussions highlighted the IG (Information Governance) emphasis on the big picture of corporate policies and the need to sell executive management on the idea of IG. My experience of IM (Information Management) is that there is a greater focus on directly delivering value to business users of information.

In summary, IG = influence while IM = action.

I can hear the howls of protest from my IG colleagues already. I stand by these words because my actions match them. To me, IM is only worth investing in when there is a direct pay-off for the business. Yes, I'm talking about demonstrable return on investment (ROI).

My IM responsibilities primarily manifest themselves as projects. Projects with a budget and timescale to deliver an agreed set of business benefits. These benefits are measured and used to determine whether-or-not the project is successful.

IG is also something I own and while it is vital to achieving greater IM value (ROI), it is an enabler. It drives indirectly a great deal of the value that a good IM project delivers.

I believe that this is important to remember when considering investments of time and money in IG. Here are a couple of measuring sticks that I find are useful when considering IG investments:

• Will the IG capabilities be used in existing or in-stream IM business activities? If they are not, then don't do it. Keep them in your back pocket for a time when the organisation is a more mature practitioner of IM.
• Do the new IM benefits that the IG capabilities make possible, exceed their cost? Or in other words, is the ROI sufficient to justify making this investment instead of another?

In my experience IG rarely meets these criteria, and so, if I'm really pushed to give someone practical advice, then it is that you can also simply consider the 'right' level of IG as being those IG activities that you can afford to do within your existing resources (people and technologies) while you are busy delivering on your IM initiatives.

Don't get me wrong. I value IG strategically, but I think that the connections you can make between IM and value are more persuasive to the wider business community. And that often means that I can fund IM but not IG - but I can use IM to implement improved IG. Governance by stealth, if you like.

The Information Governance Community

But what about today's Information and Data Governance Forum? Well, it was a surprise on a couple of fronts. Firstly, it was a very small group and I was expecting to see more people. Perhaps it is too late in the year and being in the middle of the festive lunch season put many people off. But at least those that turned up all seemed to be 'hard core' IG managers. So there was some real experiences to be learned from.

There was also a predominance of government organisations represented, followed by a number of large industrial companies. On reflection it was refreshingly free of financial and consulting people. I feel safe to say that, having now moved 'out' of finance and into media. 

You can fool some of the people all the time, and those are the ones you want to concentrate on.  George W. Bush

So a number of the presentations were made from a fresh perspective. There was a lot of talk about the legal implications of poor/missing IG, records management and retention of information. 75, 80 and 'unlimited' years of records being required in some cases. It makes my 5 - 10 year retention policies look easy.

It was also interesting to learn that metadata is discoverable under Australian law. Something I had not given thought to before.

Finally, there was some talk about classification of information. This is something I have marked down as being too difficult in my own organisation. For semi-structured and unstructured data I am pursuing a strategy of employing ever smarter search technology rather than trying to classify a small fraction of the vast amounts of information that a media conglomerate produces each day.

I guess you can say that I am still betting on technology delivering a result faster and cheaper than cultural change. Maybe I'm still a man with a hammer that sees everything as a nail.

It's what keeps me up at night.

I blame the users. If I could just replace them with a bit of software then all my problems would be solved. Now where did I leave that hammer?

Actionable Information Management Principles: The Reality

A couple of months ago I wrote about IM principles relating to people and process. I had intended also writing about principles relating to technology and content but then I hit a wall. This wall was that the principles were too detailed for the people I needed to communicate with: namely the business community. If I continued with the principles in their current form I would likely fail in my effort to have people understand why I was taking the organisation in the planned direction. Not a good look and certainly bound to fail at some point down the road.

So I had a rethink and decided to simplify the principles into a more compact form that I could communicate more easily.

My rethink centred around the information assets I am responsible for. An information asset is any information, person, process or technology whose primary purpose is to support management decision making or regulation. They include data items, measures, reports, dashboards, alerts, data marts, warehouses, business rules and processes and any person involved in creating, maintaining or using these assets.

With this in mind, my principles have become:

1. Every information asset has a single owner and definition but has one or more stakeholders.
2. Manage Data across the whole lifecycle, retaining and disposing of detailed records and data while preventing deterioration and loss of data and metadata such as definitions and data models.
3. Data quality is dealt with as early as possible and is measured in absolute terms.
4. Decision making is fact-based, established through an analytics culture that hires numerate people and supports all staff with appropriate training and knowledge sharing.
5. Business processes are tightly bound to the underlying information assets and data is quality managed to ensure that it is fit for purpose.
6. The IM function works through an information budget & funded projects and can enforce its objectives related to corporate data. All IM initiatives deliver towards a defined and agreed target information architecture.

These principles underpin and explain all of my IM strategic actions. The benefit of doing things this way is that the people I communicate our strategy to can relate complex, difficult and challenging strategic actions back to a fairly simple and straightforward set of principles. Principles that are easy for them to decide whether-or-not they support. If they support the principles, then that informs their understanding of the actions. Ultimately, they have helped me get the decisions to fund my IM initiatives. 

For me as a change agent, it's so far so good. I have initial funding to back transformative IM initiatives. These initiative are not yet completed but at least now I'm in the game with a chance of winning. Stay tuned as I hope to report back on successes and failures along the way ...

How Many Analysts Does It Take To Screw In A Lightbulb?

An old joke but a question I face each time I am asked gather a team to create a sustainable analytic capability in an organisation. Well I've finally collected together my thoughts and I believe I have a model that can be used to plan the resource requirements for a range of analytic objectives.

A couple of things to be aware of if you want to use this:

• It is based on my work with large commercial organisations - those with 1,000 employees or more (sometimes 10's of thousands more).
• I almost always use specialist BI toolsets such as Business Objects, Cognos, SAS, Informatica, Infosphere, etc.
• My experience has been that the model works equally well for business as usual (BAU), project activities, or a combination of both.

It may be that this model is less applicable outside of the 'large commercial organisation' domain - but I don't yet know why it wouldn't be. I have used this model across a range of industries and have talked with many people working in many types of non-commercial environments. Nothing they have said has indicated that they have fundamentally different analytic needs from my own. I am interested if anyone reading this has a different opinion.

Anyway, enough prevaricating, here is the resource model:

 

 (Click to open a larger version in a new window)

 So the minimum team required for an organisation wanting to have a modest analytic capability is 6 people: A business sponsor and a team of 5:

• Business Subject Matter Expert who understands what the business requires analytics for - what will work, and what wont. This person is also an analytics evangelist (but usually not a 'tools' or techniques expert when they start in the role) who will work with the business as a trusted advisor. This is the person that understands the real business requirement for analytics.
• Data Steward who is responsible for data conversion and ongoing quality. This is the person that understands how to deliver business data that is 'fit for purpose'.
• Data Business Analyst who is able to understand business requirements and create new and original (innovative) solutions to the analytic requirements of the business. This is the person that understands the 'nature of data' and can apply this knowledge using diverse tools and techniques like UML, BPR, SQL to give structure and process to the solution.
• Analytics Manager and Technology Solution Architect who manages the analytics team and is the main designer of the technology solution that will be used to deliver analytic products to the business. This is the person that understands how to deliver analytic services.
• Report Writer, help desk and trainer who gets business people to use the analytic solution and then helps support them and further improve the 'analytic products'.

If you are really, really lucky, then either the Data Business Analyst or the Report Writer also have the skills to use structured analytic techniques. But 9 times out of 10, this isn't the case. Small team such as the one I describe, are effective at data integration and starting the business in using reports and dashboards for insight. Structured analytic techniques will be important a few years down the track as the organisation's capabilities grow.

The table below gives you an idea of how these roles become more specialised as the team grows:

 

 (Click to open a larger version in a new window)

So these are the roles and skills that I believe are required if you are to be successful. They do reflect my personal biases for:

• A strong business focus and led approach where IT's main role is providing the infrastructure platform
• Delivering a self-service analytic capability into the business wherever possible (as opposed to building a large team that provides analytics to the business)

Like the BICC advocated by the Gartners and Forresters of this world, I'm sure that there are a number of models that work. The above is one that has stood the test of time for me.

BI Personas

Boris Evelson of Forrester has recently published a matrix of 'BI Personas' on his blog. As he says, this is his first attempt - but the concept struck an immediate chord with me. The matrix is trying to present on a single page a view of the many different dimensions of BI creation and use that exist today.

His blog contains the original matrix of 5 types of personas across 7 dimensions. Here is my attempt at producing a version 2.0:

(Click on image to see a larger version of this summary view) 

This matrix has added 2 new roles (Data Entry and splitting Developers into 2 types) and a new dimension (Reliance on Business SME). I've also changed some of the details in the matrix so I'm not sure what Boris will think of all this. From my perspective, the changes better reflect practices that I have seen in large corporations around the world:

• I am a strong advocate of making business subject matter expertise a critical dimension in any BI capability. It delivers a huge analytic punch that contributes directly to the amount of insight (and thus value) a business gains.
• BI and Data Integration developers rarely span both skill sets although if money is tight I will go with hiring a data integrator everytime. In this case, the BI Application Developer role is often wrapped into the Power Report Writer role as they are the ones who best understand the very specific requirements of vendor-specific constructs such as Universes and Frameworks (see persona definitions below).

Medium sized businesses will undoubtedly see a collapse in the number of roles (rows) they have. Differing levels of IM maturity will have a similar effect on the number of dimensions (columns).

Here is a more detailed matrix that includes persona descriptions and other explanations (and attributions):

(Click on image to see a larger version of this detailed view)

I have also tried to define each of the personas:

• Executives - Business people who consume highly structured reports or dashboards
• Analytic Experts - Business experts in structured analytic techniques and related toolsets
• Power Report Users - Business people who consume reports or dashboards frequently and who are highly skilled in the specific BI tool used
• Average, Regular Users - Business people who consume reports or dashboards frequently (typically each week)
• Casual Users - Business people who consume reports or dashboards infrequently (typically less than weekly)
• Data Entry People - Business people working in Call Centres, etc. that use transactional applications to create and modify data
• BI Application Developers - Experts in Presentation Layer toolsets that support data access and report writing. Examples include Business Object Universes and Frameworks in Cognos
• Data Integration Developers - Experts in ETL and data quality toolsets and methods. Examples include DataStage, Informatica, SAS and SQL

I hope this helps people communicate the complexities of modern BI activities. If you have comments (good and bad) and ideas to improve the matrix then please contact me.

Please note that Boris is the originator of the model so he gets the credit. Any errors in the above I proudly own alone.

Analytic Auteurs

In my last blog I was looking for a label that I could use to describe what I do analytically. I don't have an answer yet but I've found a new candidate: Auteur. Sounds pretentious enough to me to be interesting. Bare with me as I think that there might be something worth knowing.

In the January 29 edition of the New York Times, I read an article by Steve Lohr called 'Steve Jobs and the Economics of Elitism'. In it, John Kao talks about the 'auteur model' of product innovation. The basic idea is 'innovation driven by a single key figure'. Steve Lohr goes on to talk about Steve Job's description of himself as a Team Leader:

"In choosing key members of his team, he looks for the multiplier factor of excellence. Truly outstanding designers, engineers and managers, he says, are not just 10 percent, 20 percent or 30 percent better than merely very good ones, but 10 times better. Their contributions, he adds, are the raw material of “aha” products, which make users rethink their notions of, say, a music player or cellphone."

That's when a light bulb went off in my head and I could relate this to my own experiences delivering analytic capabilities. How do they relate to Steve Jobs description? Well I always make a point of hiring trusted experts who I know are outstanding in their field of expertise.

 I know they are good because we have worked together before. This gives us a shared level of trust that is hard to beat.

They are the catalyst for exceptional delivery to the organisation. We think as a team and our combined skills are much greater than the sum of the individuals. What I also call the 'multiplier'.

By multiplier, I am referring to the team's collective ability to really change the way an organisation manages its information. Together we have critical mass. The Team is the catalyst that sparks creativity within the wider Analytic Community in the organisation. Like a chain reaction, the Analytic Community itself inspires their business divisions to develop ever greater insight and information management skills.

Now I never thought of myself as an auteur, but my favourite analogy for 'doing analytics' was making a movie. The analogy runs something like this:

As the director (think of me as a cross between Tim Burton and Spike Jonze) I assemble a team for 1 - 2 years and together we create the analytic solution the organisation needs. The solution is generally made up of two components. The first is a suite of information products (data marts, cubes, reports, dashboards, etc.). The second is the skills for the organisation to 'self serve' and create new information products without us.

We then we disperse to different places. Only to meet again when the next 'movie' opportunity arises.

My trusted experts are:

• Business Analyst - Script Writer
• Data Integrator - Special Effects
• Solution Designer - Cinematographer
• Information Designer - Set Designer/Head of Props
• Quality Control - Post-Production
• Change Agent - PR and Marketing

Why do the same people agree to work with me more than once? Well, you'll have to ask them to really know. But I can tell you what I tell them when I'm trying to get them them for one more gig. In no particular order I promise them:

• A real intellectual challenge
• A core team they can trust
• That they will learn from both the challenge and their fellow team members
• Very good remuneration - at least as good as they will get elsewhere
• Fun!

Don't get me wrong. The core team does get new blood each time and sometimes people just muscle their way into the team by being damn good at what they do. Never underestimate serendipity and the pleasure at meeting another person of whom you think 'wow - they're good!'.

Wow is the thing that keeps me going. If you can't be passionate, then the low points of any 2 year analytic journey are going to outweigh the high points. 

So that's the analogy that has worked best for me in the past. Writing this down makes me want to go out there and do it all again!

Footnote: To the cynical bunch of people that know me - I have not written this blog in order to stroke Steve Jobs ego so that he will send me a free iPad. But Steve, if you really, really want to ...

Social Networking and BI

I was sent a link to a presentation on social media and BI by US/Indian company erada. They specialize in developing solutions that work in a SAP environment.

In the presentation they quote Gartner and their recent (February 2009) survey of 1,400 CEO's who reported that they did not achieve BVA (Business Value Attainment) in their BI investments  What's really interesting is that less than 9 % reached their desired BVA when they went live. BVA is broadly a measure for stakeholders of whether-or-not a good investment has been made.

What is not clear from the Gartner analysis that I have seen is why you would expect to see instant return on your investment. If you buy a truck for your company you expect it to earn a solid return for you over the future years - not instantly the day you buy it. Isn't this the same for BI (and any other software) investment. Or am I missing something?

I suspect Gartner is trying to measure the perception that CEOs have of what they are getting at launch versus what they expected to get (based on promises) when the BI investment decision was made.

This is an important point and I myself go to great lengths not to over promise to stakeholders. It's also not necessary to do so as the benefits of the BI investments (at least where I have worked) are simply massive. The trick is to use language your stakeholders understand. Delivering value also helps ...

Here is a copy of the presentation. Persist with the way the slides are built as there is some good information to learn:

Social Network BI-Outside In_sept09

I like the slide pictured above as they make a good point that just churning out reports each day does not deliver insight.

I also like the sound of their SmartCube product that (according to their website blurb) "helps in redesigning the SAP InfoCube to reduce query time, load time and footprint on BIA." Anything that helps InfoCube work faster gets my vote!.

Need To Brainstorm?

Do you want to run a brainstorming session with your analytic team? This could be hard going if you don't prepare for the session. Here are a couple of short videos found on youtube that show you what not to do:

 

and what you should do:

 

These two short videos hit their mark. They were created by students at the Stanford Design Institute and forget the excessive (for aussies) 'earnestness' of the US students because what they show in a few short minutes is great information.

How Not To Transform

I read with great interest a recent article on the departure of Telstra's Senior vice-president of transformation - Tom Lamming. This Accenture partner had been taken into the company by CEO Sol Trujillo as a part of the US Executive influx at the start of his time at Telstra. He has now resigned from Telstra and returning (I imagine) to Denver and Accenture.

Lamming started out at Telstra as a consultant where he was paid up to $15,000 a day. Presumably less after he joined Telstra as an employee. What do you get for that sort of money?

But first, what was he paid to do? According to reports, his stated objectives were: a $12 billion, five-year transformation program that sought to develop common business processes for Telstra's operating support, customer care and billing systems. The aim was to save the telco hundreds of millions of dollars as 1,500 legacy systems were consolidated down to 'between 250 and 300 by 2010'.

This was all part of Trujillo's business transformation of Telstra. He talked about the network transformation at Accenture's Global Convergence Forum 2007. It gives you an idea of the scale of transformation being attempted:

Accenture Global Convergence Forum 2007, 11-13 April, 2007 in Rome, Italy. 'Driving Transformation Through Innovation' by Sol Trujillo.

So perhaps this was a good investment for Telstra - it certainly should be if the transformation delivers.

What I am particularly interested in is the interim report card for the transformation project 4 years into the 5 year project. Here it is as far as I understand it:

On the plus side:

• Seven million Telstra private customers have migrated to the new platform (an application that crashes a lot is still an application)
• 400 of the 1,500 legacy applications have been retired (with one year to go).

And on the negative:

• Only 250,000 business customers had been moved across
• Reportedly, the budget is blown by more than $1 billion
• Of the 1500, by 2010 Telstra hopes to reduce that figure by 80 per cent. But with a year left, only 400 have been switched off.
• More than 3,300 change requests have been issued
• Related to the last point, the 'fixed price' contract with Accenture has reportedly blown out from $350 - $400 million to almost $600 million. Plus there must be a significant cost to keeping all of the remaining 1,100 legacy applications running
• Presumably no net savings have been made so far (given all of the figures reported).

On the face of it, this is a pretty poor result. Don't be fooled: it is a poor result. No wonder CEO's, CFO's and CIO's are so worried about starting major transformation projects like this. 

Telstra are not alone in achieving this level of poor performance, but it makes me weep when avoiding these problems is so easy.

Can anyone spell 'fiscal discipline' anymore?

So where was the effective governance in this program? Why was such a transformation project agreed to in the first place if governance wasn't the number one priority? Apparently performance tracking of the project was missing - or at least ineffective.

Transformation is always difficult. You succeed by breaking the problem down into small, measurable chunks that are delivered on an agreed schedule throughout the life of the project. Big bangs are bad. Period.

When you've done this you then put into place people with the backbone to govern the project and demand that the project deliver as promised - and if they don't then to understand why and judge whether-or-not investment continues.

Why is this so difficult?

This may be why I am a turn-around specialist rather than a permanent line manager. That and the fact that I spend a lot of time on governance and proving benefits as they are delivered. You'll have to take that as true for now - but please check with people I have worked with and for. I hope that they back me up on this.

The point about all of the above is that hopefully Telstra will learn something from this experience and bring the transformation project to a successful conclusion. I for one will cheer for a Project Manager that delivers this outcome. Even at a starting rate of $15,000 a day - that would still deliver bottom-line profits to Telstra.

I'll be brushing-up on my American accent for my next interview!

I'll leave you with Mr Lamming's own opinion as reported in the papers:

Quote:

"Telstra's transformation is the largest, most comprehensive and successful I have ever seen, and we have implemented it at speed," Mr Lamming said.

May he be as successful in a future engagement - but hopefully not in Australia!

Check back here in a year's time to see the end result of the 5 year program and the final report card.