Controversial may his decisions be, I still think Bernanke sets a great example of leadership. Having probably the toughest job in the world, in the face of great uncertainty and unfavorable public opinion, he led a series of bold actions, without an eye blink.
Part 1:
Part 2:
Collective problem
solving sessions, ranging from formal workshops to brainstorming meetings and
online forum discussions, are often not effective. There are a few common
mistakes that I saw happening again and again. Some of them are due to bad
problem solving method and some others are due to lack of a systemic approach.
This post will serve as a "what not to do" list with tips to make
such sessions more effective applying systems thinking in problem solving.
Complex problems often require a bigger group with different views to
participate into brainstorming, discussion and formulating solutions. However,
many sessions go off the main topic, lots of discussions but no clear result,
or worse, create/deepen conflicts and force compromise. These often lead to
pessimism about collective problem solving and lead to using less effective
mechanisms such as upper management intervention or hard sell an one-sided
plan.
From my observation and
experience leading sessions with widely diverse groups, here are components of
a formula for unsuccessful collective problem solving:
1. No clear process of
problem solving
To bring in different views effectively in a short time requires a careful
design of the inquiry and problem solving process. This process is often
reflected in an agenda. Having no agenda or spending not enough time to explain
and reiterate where the group is in the process is prescription for random
conversations while important inputs are not appropriately voiced.
Tips: Scaffolding the working session.
Careful develop a detailed agenda that takes into account not only the process
but also group dynamics and contribution that each group of participants can
make. Make sure you have the right mix of participants to understand, explore,
validate and decide. Whenever possible, synchronize the agenda in advance and
make sure participants come prepared with information and thoughts. This step
is as important as scaffolding before starting to build.
During the session, spend
ample time to explain to participants each step and intended purpose of each
step in reaching the final goals. The facilitator(s) needs to keep the flow but
allow flexibility to accommodate useful conversations and changes without major
deviation from the original process. Summarize results so far and clarify the
following steps once in a while to make sure that participants can follow the
process.
2. No upfront common
understanding of the objectives, value drivers and constraints
Assuming that every
participant has the same understanding about the background, especially
objectives and constraints, often lead to unproductive arguments and
conversations.
Tips: Define purpose and system
boundary. Start off the session by asking the sponsor to set the ground and
talk about the value drivers and objectives of the problem solving exercise.
Explain and agree on the constraints, but at the same time make sure everyone
understands why the constraints exist, just in case some of them need to be
revisited later (see point 5 below).
3. Jump to solution
discussion before the problem is properly understood
Sounds very simple, this is the most common mistake I observed. Under the
action imperative, especially when there are many experts in the room, groups
tend to start immediately discussing about solutions. In addition, to be
objective and avoid conflict, participants are often request to present data
related to their argument. However, participants with different mental models
interpret data differently, resulting in grid-lock arguments and conflicts.
Tips: Bring everyone to a helicopter
view of the system. Build a systemic problem model visually before solution
discussion. Data presented should be positioned against the appropriate part of
the model. Once presented against the model, data will help not only to give a
detail picture, but also to develop and verify the model of the problem. This
is like taking people out from their local views to a helicopter view of the
whole problem. Once the system is properly understood (and optimally documented
visually), the solutions normally emerge naturally.
4. No clear view of the
outcomes
Discussions often are divergent and take long to go to conclusion, sometimes
still not good enough to be used in the next steps when participants have no
clear view of the desired outcomes. Words are normally NOT enough.
Tips: Visualize and build the
outcomes as the discussion goes. Visualize and discuss early in the session
how the outcomes should look like. Design the process so that each step will
help to move the group closer to the outcome. Result-based conversation are
more productive, and visualization of each step toward the outcome will help
participants appreciate the progress much more.
5. Try to solve problem
at the wrong level/area and avoid to solve problem at a different level/area
Many issues that people try to solve are only symptoms of problems somewhere
else. For example, process exceptions in many cases are created by breakdown in
related processes or wrong organization structure. Trying to accommodate the
process exceptions will not make the problem go away, even make it more
difficult later to address the root cause.
Sometimes, recognizing that
the problem needs to be addressed elsewhere, narrowly-defined boundaries, rigid
deadline and lack of upper-management support, the team still create a
quick-fix and avoid the root cause.
Tips: Solve the problem of different
systems at the same time. While examining the problem in the current
context, it is important to put the system of inquiry in a bigger perspective.
Bring people with broader view and responsibilities to the process. Diagnose
the other related or incorporating (i.e. mother) systems, take not only
functional but also structural view of the problem. Once problems elsewhere are
identified, come up with careful analysis of dependencies and plan to initiate
problem solving at different levels and locations in parallel. Contingency
planning is important to make this happen without compromising, breaking the
deadline or exploding the scope of the project.
After all, collective
problem solving involving different views and functions requires a lot of
preparation, thoughts and skills that most people think. Inadequate attention
to these often result in frustrations, conflicts, compromise, and
create/reinforce boundaries and cylos. The tips presented here are a starting
point to think about a more effective process.
Uploaded from Heidelberg
It has been a while since I blogged and many people asked me about what I have been doing recently.
To answer, here are some of the things I have been doing since end of last year in no particular order:
Transition is a fun process though I have been through many, and again, I will strive to live up to the challenges to prove who I am.
Uploaded in Heidelberg, Germany.
Being a fan, and to a certain extend, a practitioner of visual problem solving and visual thinking, I always am intrigued by different tactics of expressing problem in visually-comprehensible ways. Hans Rosling's presentation at TED were remarkable presentation of global data in such a way. Here are two videos to mark my come back to blogging after a long period. Enjoy!
Debunking third-world myths with the best stats you've ever seenNew insights on poverty and life around the world
I got the link from a colleague for a talk in TED, which stands for Technology, Entertainment, Design, though the scope is getting wider every year. The conference aims at bringing people together to share their special ideas. This is the most impressive one I found, creating new creatures. A good one to stretch our imagination.
Web 2.0 Summit is often a very important event to discuss new developments and trends in the area. One of the most interesting sessions of the event this year is the presentation from Mary Meeker of Morgan Stanley. The presentation addresses technology trends in general and is very interesting, except that it cites quite a few Cisco's info (which is a clever company, but still biased, though).
The slides and video however was not included in the official site. I found them so included here.
Continue reading "Web 2.0 Summit 2007 - High Order Bit Presentation from Morgan Stanley" »
Most of people I talked to was very happy with the information they got from Google Search. Natural language searches on the other hand try to synthesize information and give a more definitive answer. I tried AskJeeve and found out that I most of the time, it gave me information similar in type with Google. A few exceptions are for simple definitive questions, e.g. "how is weather in Rome in December" comparison below.
Recently, Powerset and TrueKnowledge launched beta service with promise to give people more power to get the correct answer to what they want. Barney Pell, Powerset's CEO pointed out that to a keyword-based search engine, "book for children", "book by children", and "book about children" are all equivalent to "book children".
According to TechCrunch, "Powerset is both indexing the web and working to convert natural language queries into database-understandable queries. True Knowledge is only tackling half the problem - the conversion of queries. They are not indexing the web. Instead, True Knowledge is grabbing data from structured databases, like, for example, the CIA Factbook."
From descriptions, both services are quite interesting. I tried to register for both Powerset and TrueKnowledge, but haven't got any invitation, yet. It will be quite challenging, though, to change the short form of queries that people got used to. The new generation's preference, short attention due to information overload and the trend for mobile device searches will probably prefers the keyword search.
Below is a short video about features and architecture of TrueKnowledge. I particular love the fact that it gives the answer immediately and cites facts leading to the conclusion.
The slides are very informative and surprisingly comprehensive.
Fast-growing entrepreneurial companies often face a daunting task of maintaining autonomy while continuously improving the structure and controlling mechanism. As business environment in and out side company is normally not clearly defined, a rigid structure and control system will limit the future of company. The answer is a good balance between a flexible control system that emphasizes cohesion and deliberate plan to grow trust.
Nothing is like being a part of a fast-growing entrepreneurial company, exciting, ambitious and still chaotic. Growing out the first period, there is a strong sense of satisfaction with empowerment, without which, the company would not be this successful. Many people, especially management still remember the days when everyone knows everyone else, did everything and still worked as one.
However, the company has changed with many new generations of people joining, management simply don't have visibility into everything anymore. Management team is challenged by the enormous amount of works and has to rely on trust to give autonomy to units. Without autonomy, the company runs the risk of limiting its future, or even slowing down and doom. Trust is difficult when there are many new people, and when most of the management are working in new capacity than they had been doing before.
On the other hand, management starts to be under pressure from new stakeholders to reduce the risk of losing control of the organization, which they are fully responsible for. Growing the company through the initial days, management often has fear that employing too much control will destroy "the magic" and hamper the growth. In other words, management often finds it difficult to make a compromise between trust and control. In fact, management has no choice, though, but combine these two seemingly contrary measures of reducing risks while managing a bigger complexity.
This post presents a set of recommendations to strike this balance. The measures pointed out below are applicable to the whole company or its autonomous unit. It is still worthwhile to note that application should varies with situation of the company/unit. If most of the people identify themselves well with company, there should be less control necessary. If people have low commitment and there are high potential gains from deviant behavior, control is more important.
The measures proposed included a flexible and lean control system with deliberate trust growing plan. In addition, coordination, negotiation and monitoring are important to reduce the drawbacks of control and to avoid too much interference that harms trust building efforts.
Flexible Control System
Traditionally, companies immediately establish the formal structure with effort to define tasks, authorities and information flows. An issue with clearly-defined structure is silo thinking that reduces collaboration. A thorough exercise like this requires a lot of effort and a large part may not valid for long in a changing environment like in fast-growing entrepreneurial companies, anyway.
Managerial cybernetic sheds new light into the concept for structure,
that is a better fit for entrepreneurial company. It is stable relations that
allow people and other resources to operate
together as a totality define structure. The organisation's structure
is thus viewed as a network of stable 'real life' on-going
communications, or organisational processes, and not as static formal
reporting relations. Objects, such as communication devices or
information systems, thus constitute an organisation’s structure.
Fast-growing entrepreneurial companies should focus on understanding, establishing and modifying these stable communication relationships, which happens directly by conversations or indirectly through coordinate actions as an outcome of sharing a context or culture, among its units and roles.
In addition, these companies should spend time together with their units to define and agree on the coherence boundaries. These are guidelines to make sure identity of the company, company's operational direction and synergy with other units of the same company are taken into consideration in unit's decision and operations.
Fast-growing entrepreneurial companies also need to tune their incentive system. The system should encourage collaboration, tight compensation to level of support (e.g. service quality) to other parts of organization, or support achievement of whole organization.
Leadership development and training are effective measures, though less often thought of as mechanisms to ensure compliance and coherence. These help enhance management community, create sense of direction and involve key members from unit management to development of strategy and policies.
Finally, these companies should also facilitate sharing with new generations by reviewing resource allocation agreement more often, recognizing contribution and seniority but emphasizing growth of the company or unit as a whole.
Coordination, Negotiation and Monitoring
Continue reading "Balancing control and trust in fast-growing entrepreneurial companies" »
Recent Comments