SessId010

Critical Issues in Managing Very Large Projects (Like ERP) Marie Benesh, Eileen Strider

SEE SESSION NOTES AT THE BOTTOM OF THIS SECTION

Maybe you have a favorite recipe for chocolate chip cookies that makes 3 dozen cookies -- enough for a party of 20 people. The recipe calls for an 18-ounce bag of chocolate chips, one and a half cups of flour, and so on. Now suppose you had to make chocolate chip cookies for all the kids in the state high school wrestling tournament -- 6,000 teen-age boys. We can be sure of one thing: multiplying the recipe by 300 and preparing it in your own kitchen with the help of 299 other people just isn't going to work.

Anyone who has ever worked on a very large project knows that small projects don't scale up easily. In this session, we'll share large project experiences to extract learnings that can amplify our effectiveness on the next large project we undertake.

Applications: Handling problems of scale; Making large projects manageable; Improving software development; Scaling small projects up to large projects; Solving special problems of large projects.

Do you have experience on a large project? We'd like to hear what your thoughts are on how large projects are different than those with teams of 5-10. And, of course, we are always interested in new ways of solving those problems and mitigating the special risk associated with large project endeavors. MarieBenesh

How large is large? That's an interesting question to cover in this. I'm on a large project, but not an ERP/IS style project. We're a mixed hardware, software, networking product architecture, with a number of suites of products in the architecture, actively employing 300-500 R&D folks depending on how you count. Many subprojects pretend they are small projects and ignore the architecture and product direction. Some parts have scaled up - and then scaled back. Lots of gold to mine in this one. - JimJarrett

Good question. We seem to know, both intuitively and from past track records that "small" is easier to successfully complete -- so, I guess it's natural for people to think small and try to ignore all those complexities that come with "large". And in some areas of the project, this is probably exactly the right thing to do -- but not with areas like architecture and product direction! Does the scaling up and then scaling back work effectively in your opinion? -- Eileen Strider

************************************************

Critical Issues in Large Projects-Session Notes

Eileen Strider (Strider & Cline) Marie Benesh (Omega Point Consulting)

Decision-Making

Issues:

o Survivor � decisions made via ad hoc alliances

o Decisions not made

o Decisions made by default

o Not enough information

o Speed of decisions

o Informal Power Vs. Formal Power o History o Relationships

o Passive/Aggressive

o Lack of governance

o Related to managing change

o Follower mentality � no leader

o Have a way for input and requests to get into the system

o Execs (4) [hour meetings, 8-10 min. per package, not problems solvers],

o [�change council�, decision-makers, commitment] � VP�s (16) � Agents (4-8) [1 chairman, get trained] Ideas: o Governance structure- SEPG

o Use the �Change Model�

o Agent who builds relationships

o Gets new info to feed to farmers

o Tests locally

o Draft 1 agent per 25 engineers

o Determine scope

o Bring change to central group o Assess, refine for rollout, package it

o Send package to agents

o Get email feedback at SEPG

o Contact requestor to solve issue

o Send out resolution to everyone

o Send to resolution to Change Council

o After Change Council approves, it goes out to 1500 agents

o Posted on Central web site

o Roll-out

Other Ideas: o Very specific decisions can be made at a small level

o Use a decision tool (Marie�s form)

o Lack of decision weakens the product

Design & Development

Issue: Requirements incomplete

Not full understanding of how data or system is used Not enough business involvement

Idea: Get an understanding of how the system will be used before design

Issue: Under-estimating cost complexity

Knuckling under on cost without changing scope

Idea: Use interactive approach

Storyboarding- prototyping

Issue: Overly aggressive timeframe

Issue: Poor Coordination

Idea: Break down barriers up front

Try not to make assumptions Use decision-making process

Issue: The word is �estimate�

Idea: Allow contingency

Allow time for training Getting ready for the project

Multiple Parties

Issues:

o Differing motivations, agendas, and goals (unspoken)

o Who�s in charge o When this changes, then what??

o Honesty, Integrity, Vendor disclosure

Ideas:

o Good leadership o Removing fear o Shared vision

o Clearly states expectations of the team

o Win/Loss structure- shared

o Remove barriers by class

o References - Check Them!!!

o Checkpoints throughout project

o Agree for basis for renegotiation upfront

o Roles and responsibilities

o Communicate rationale behind goals

Integration Interlock:

o Schedules Vs. Dependencies o Set up staging area o Does your stuff work with my stuff?

o Project Survival Guide (Steve McConnell) o Hourly builds o Continuous builds o Schwab o Triangle: Speed � Cost � Quality

o Misunderstanding the technologies and their role o Pay attention to this o Eliminate the technology

o Expectations of the customer

Project Management & Reporting

o Clearly defined goal

o Who has the authority to manage to that goal

o Roles and respect � right people in the right roles

o Failure to �chunk� properly sub- projects

o Have a PMO to coordinate �chunks�

o Validate completion � measurement

o Project Administrator needs time o Collect data o Commit the resources

o Take care of people � Plan it! o Burnout o Respect o Safety o Conflict

o Use a PMO

o �Dimensions of Authority� o Formal- bestowed on you o Competence o Respect (character)

o Peer reviews

o Accountability (held to it)

o Use tools for reporting and tracking resources

o Clear roles and responsibilities

Communication

� Requires a safe feedback mechanism � safe culture

� Clearly communicated roles and responsibilities (temperament and skill)

� Info dumping as a substitute for real communication

o Ie. 27 3-inch binders for a requirements doc.

� What communications media to use? o Written? o Behavior? o Speech?

� Who is our audience?

� What needs to be written down? o Clarity o Get to the point

� How much emotional content can the media contain or express?

� Can we draw a picture of emotion?

Ideas: Communication

o Communications training o (MBTI, clarity, media, form, timing, audience participation, learning modalities)

o Feedback into the system o A safe environment

o Check for understanding

o Be aware of the timeframe that info will be important � determines the media o Web page � post-it note

o Assign resources (people) to do communication

o Check-ins

o Contact

o Create awareness that it�s �ok� to experience ups and downs

o Retire for a couple of days

o Check-in with yourself o Where am I?

o Reward yourself (massage, tennis, etc.)

o Contact with someone outside the change o To ventilate to o To be listened to

o Mentor

o Persuasion and speech criticism class

o NLP

o Remind yourself and people of what�s not changing

o Have fun everyday

o Forgive myself in advance for not being perfect

Updated: Tuesday, January 9, 2001