6 years, 11 months ago

Functional Organization at Microsoft

@iamjaygreene and @jimkerstetter of @CNETNews are not surprised by the departure of unpopular Windows boss Steven Sinofsky from Microsoft.

Some pundits (e.g. ZDnet’s Larry Dignan) had predicted that Sinofsfy would survive if Windows 8 was a
commercial success. By letting him go immediately after Windows 8 went live rather than waiting,
Ballmer has clearly signalled that it is not about Windows 8 success but
about something else.

In pieces written in the weeks before Sinofsky’s departure, Greene and Kerstetter mention the following issues.

  • Sinofsky successfully battled with Ray Ozzie for control of Windows Live Mesh. Ray Ozzie left Microsoft immediately after Ballmer folded Windows Live Mesh into Sinofsky’s organization.
  • According to unnamed critics within Microsoft, Sinofsky created a rigid product development process that puts more control in
    his hands and diminishes Microsoft’s ability to innovate.
  • In a similar fashion to Scott Forstall at Apple (who also lost his job recently), Sinofsky zealously promoted his group’s work at the expense of the rest of the company.
  • Manu Cornet’s cartoon of Microsoft’s organization chart is thought to be a reference to Sinofsky.

The comic is a set of 6 organizational charts, edges with arrows show who reports to whom. Amazon's is very traditional, each manager has exactly 2 people below her. Google's is colorful (nodes are colored red, green, yellow, blue) and is extremely messy. Edges are overlapping all over the place, it's unclear who reports to whom. Facebook looks like a social network with bidirectional arrows and a distributed structure. Microsoft's is divided in three sub-structures that are pointing guns at each other. Apple's is a circle with a large red dot in the center, and everyone around it reports to that red dot -- the arrow heads are particularly large and even the people two levels away from the center red dot also have arrows point at them coming directly from the red dot. Oracle's is divided into two sections, the first section is labelled 'Legal' and is huge, the second section is labelled 'Engineering' and is tiny.
Original cartoon by Manu Cornet

But this story isn’t just about personality clashes and organizational politics. Sinofsky has championed an approach to organization structure, which he calls Functional Organization, and this is described in a book called “One Strategy: Organization, Planning, and Decision Making,” (2009) co-written with Harvard Business School professor Marco Iansiti.

The Functional Organization builds management reporting lines around job functions — such as
product management, development, software testing. This may be contrasted with a Product Organization where multi-disciplinary teams work on specific
feature sets together.

Sinofsky and Iansiti argue that functional
organizations create clearer road maps for workers to march toward a
final goal. However, critics within Microsoft disagree. Apparently referring to Sinofsky’s Functional Organization, Charlie Kindel, another ex-Microsoft executive is quoted as saying that “it represents a siloed perspective, it represents an us versus them perspective”.  Another former senior executive (unnamed) has referred to the approach as “Soviet central-planning”, where tight control from the top squeezes out innovative thinking from below.

Announcing Sinofsky’s departure, and the appointment of Julie Larson-Green as his successor, Steve Ballmer wrote “The products and services we have
delivered to the market in
the past few months mark the launch of a new era at Microsoft. To
continue this success it is imperative that we continue
to drive alignment across all Microsoft teams, and have more integrated
and rapid development cycles for our offerings. …  Her unique product and innovation perspective and proven ability to
effectively collaborate and drive a cross company agenda will serve us
well as she takes on this new leadership role”.

(BBC News 13 November 2012)

So is this the end of the Functional Organization in Microsoft? Martin Fowler talks about the oscillation between FunctionalStaffOrganization and
TechnicalStaffOrganization, essentially the same dynamics (he reckons) as drive the
boom-bust cycle of EnterpriseArchitecture. (PreferFunctionalStaffOrganization). So perhaps now the cross-company silo-busting agenda will have the ascendency for a little while.

Read more »

6 years, 11 months ago

Is Organizational Integration a Good Thing?

Some members of the UK Government are keen on integrating health and social services. In his first speech as Minister of State for Care Services, @NormanLamb said

“The consensus behind integrated care is pretty universal. In government,
in think tanks, in patient groups everyone sees it as A Good Thing.” (Transcript of speech at @TheKingsFund, 11 September 2012)

And Junior Health Minister Dan Poulter is just as passionate. Integration of NHS and social care “is like the holy grail”, he told the Guardian recently (30 Oct 2012).

But not everyone agrees. Jane Young is a disability consultant and campaigner. She asks Would the integration of health and social care promote independent living?
(Guardian 8 Nov 2012), and argues that it would not.

“Rather than medical treatment, disabled people need assistance to
perform such varied everyday tasks as driving, bathing, dressing,
typing, cooking, parenting activities etc. None of these functions
is normally carried out by medically trained professionals, so on this
basis it is illogical for the Department of Health to be wedded to the
integration of health and social care services.”

Meanwhile Jeremy Hunt, the Secretary of State for Health, sounds an ambivalent note.

“But structures are only a means to an end.  What really matters is better health and care outcomes.” (25 October 2012)

 What are the problems that integration might tackle. There are many symptoms of poorly joined-up services. Jonathon Tomlinson documents some from his practice as an East London GP.

  • Adverse social factors, such as poverty and social exclusion, have a critical impact on
    the efficiency and productivity of healthcare.
  • It is impossible to discuss diabetic control or smoking cessation with
    someone whose housing depends on her benefits which have just been cut.  
  • Patients cannot follow routine healthcare advice when their lives are disorganized as a result of financial stress, or when they cannot afford to pay for prescriptions.
  • Hospitals, clinics and surgeries are full of people who don’t know where else to go. Hospitals beds are blocked by patients who lack sufficient social
    support for them to be cared for elsewhere. 
    Hospital staff report readmitting the same patients week after week
    because they cannot cope at home.

Based on: A perfect storm: welfare meets healthcare (June 2012)
(slightly reworded)

I agree with Jeremy Hunt that outcomes matter more than structures. Obviously this covers the individual needs of patients and their carers, but also includes broader economic and social outcomes, such as higher quality and value-for-money, to be achieved through innovation and leadership.

Hunt describes integration in terms of “a culture of cooperation”, “meaningful contact” (e.g. between GPs, consultants, local authorities and social care providers) and “bringing people together”. But how are these things to be achieved? By better processes? By heroic leadership? Hunt merely appeals to new structural mechanisms – specifically the Health and Wellbeing Boards, and Healthwatch – which will somehow bring about a sufficient level of “meaningful contact”.

I presume that Jane Young has no objection to some level of “meaningful contact”. Her main objection to “integration” seems to be that she doesn’t want to see the Department of Health managing services that do not require medical training, thereby implying that organizational boundaries should be primarily aligned to skills rather than outcomes.

But it seems to me that “meaningful contact” alone cannot bridge the structural barriers to joined-up care. If patients are getting the wrong (expensive and inconvenient) care package because there isn’t funding for the right care package, this needs to be addressed during the budgeting and commissioning phase, not by better coordination in the delivery phase. Surely we need to start by understanding what overall capabilities and processes are required for effective management and delivery and governance of care, before we start allocating responsibility for these capabilities and processes to various agencies.


Let me just take a step back for a moment. Something
called “integration” is being put forward as a structural solution to
some set of problems. But there is a great deal of confusion about what
“integration” actually means, what this “integration” might achieve, and
whether there are any unpleasant side-effects. Some people may think that “integration” between A and B merely means establishing effective channels of communication between A and B, while others may think “integration” means shared planning and commissioning, integrated governance, or even full merger.

In my
opinion, structural solutions to complex problems is (or should be) the
job of the business architect, and I believe that business architecture
can play a vital role in clarifying the requirements for “integration”
and working out the practical details. So we need to apply some of
the business architecture viewpoints to thinking about the integration
of care.

We might imagine that the ultimate in integration would be to put all healthcare and social care into a single monolithic organization, but there needs to be some differentiation of structure even within an apparently monolithic organization, so that would merely reframe the problem rather than solving it. There is still a challenging architectural question – what structuring and organizational design principles to use for carving up responsibilities and negotiating exchanges between different units.

There is nothing logically wrong with the idea that
responsibility goes with expertise, as Jane Young favours, except for the fact that it doesn’t deal with the observed symptoms. Evidence-based healthcare is taken very seriously, and there would be strong objection to applying some quack nostrum without proper study, but the evidence base for organizational change in the NHS seems to be very much weaker.

See my earlier posts Resistance to Architecture and Illusion of Architecture.


(Update) Of course integration is not just a concern for the public sector. Compare the latest changes in leadership at Microsoft “aimed at
ensuring the firm continues to be a dominant player in the sector”. Microsoft CEO Steve Ballmer said “The products and services we have delivered to the market in
the past few months mark the launch of a new era at Microsoft. To continue this success it is imperative that we continue
to drive alignment across all Microsoft teams, and have more integrated
and rapid development cycles for our offerings.” (BBC News 13 November 2012) See my post Functional Organization at Microsoft (Nov 2012)


This is one of a series of posts on The Purpose of Business Architecture.

By the way, places are still available on my Business Architecture Workshops (January 29th-31st)