TrainingConferencesAbout UsContact UsAdvertiseSQE.comRSS Feed

StickyMinds.com: brain food for building better software

Log In
 Clarify Your Search Criteria

Tips on Using Our Search Feature(s)
 
StickyMinds.com Home
ResourcesTopicsCommunityPowerPassBlogs
Home  >  Topics  >  Test & Evaluation  >  Detail: Three Pounds of Manure in a Two-Pound Sack



Three Pounds of Manure in a Two-Pound Sack

By Payson Hall

Send This Content to a FriendGet a Short Link to This ContentPrint This Content

Summary: Multitasking is not a magical cure for getting too much work done by too few resources. Listen in this month as Payson Hall eavesdrops on a coaching session between two managers about how to assign and prioritize work effectively.

HP
Steve was the newest and youngest manager working for Tazeem. He was having a problem with one of his team members and was meeting with Tazeem to discuss the issue and get coaching. "Mark seems frustrated when I give him a new task while he is in the middle of completing another assignment. Any ideas how I can better help him understand that multi-tasking is part of his job?" Steve asked.

"We can all become frustrated by untimely interruptions, Steve," Tazeem said. "What makes you think multi-tasking is an issue for Mark?"

"He's not getting his assignments done. Last week, I asked him to review some test results to see if the errors had a pattern. Monday, I asked him to work with the architecture team to help figure out a performance problem. In the Wednesday status meeting, when I asked about his progress reviewing the test results, he said that he hadn't finished his analysis because he was still looking into the performance problem," Steve said, sounding a bit irritated.

"What were your priorities for the tasks you assigned?" Tazeem asked. "Did you want him to analyze the test data before investigating the performance problem, or was the performance issue the higher priority?"

"Both are important," Steve said. "Mark is a professional. Can't I expect him to manage his workload? In this economy, aren't we expected to 'do more with less'?"

Tazeem was being gentle, remembering that Steve was a new manager who was having a problem that new managers often encounter when they rely too much on management clichés. "These certainly are trying times," she said. "Challenging for both the professionals and the people who manage them. While we ask a lot of the people we supervise, we must be sure that assignments are clear and reasonable."


"I'm not sure I understand," Steve said. "I thought I was pretty clear that I wanted both tasks completed." He seemed confused.

"Did you make clear how much time you expected Mark to invest in the test analysis and performance analysis tasks?" Tazeem asked patiently.

"I wasn't sure how much analysis would be required. I expected Mark would know."

"When you are assigning a task, it's good practice to ask someone how much time he thinks it will take." Tazeem explained. "It helps you help him manage his work load. Sometimes the person taking the assignment will have a good idea of what is required and can give you an estimate; sometimes he will need you to clarify how thoroughly you want the task completed; and sometimes he will need to do some work on the task to get an idea of what it will entail or to identify questions."

"But tasks take as long as they take. Even if I get good estimates, what good is the information?" Steve asked.

"Imagine that the test analysis task requires forty hours of Mark's time and that the performance analysis task needs forty, as well. Asking Mark to do both in the space of a week is unrealistic. If you ask for something unrealistic, the person you are tasking may not know what to do. Without additional guidance, team members often assume the most recent task they are given is the highest priority and stop work on prior tasks," Tazeem said.

"Isn't that a multi-tasking problem?" Steve asked. "Shouldn't Mark have been able to do both?"

"Multi-tasking doesn't solve the problem of how to do eighty hours of work in a week, Steve—in most cases, multi-tasking is less efficient than working on one task from start to finish. Sometimes it is necessary to divide someone's time among multiple tasks, but there is overhead in switching attention from one to the other, and neither task is made more efficient by the stop-and-start nature of doing the work. That's why it's so important to be clear about priorities. I would encourage you to talk with Mark and ask him how much time he thinks the tasks will take. Check to see what other tasks he believes are on his plate and how long he thinks they will take; then work with Mark to prioritize them."

"Do I need to prioritize every task I give him?" Steve asked.

"If you ask Mark to take on another task before he has completed everything he is working on, make sure you discuss how the new task affects the priority of his existing work. You are both less likely to be frustrated and more likely to get the right things done. You also might encourage Mark to ask you about the priority of any new tasks you assign in case you forget to mention it. That will help empower him to manage his own workload," Tazeem said. She noticed that Steve was making a note to himself. If he followed through and adopted the behavior consistently, she thought, it would be a positive step in his growth as a leader. {end}

What positive or challenging experiences have you had related to assigning work to others, monitoring their work, or dealing with the challenges of multitasking?


Join the conversation below or start a new one in the Member Comments section.


About the Author
Payson Hall is a systems engineer and consulting project manager for Catalysis Group in Sacramento, California. He has consulted on a variety of public and private sector projects in both North America and Europe during his twenty-six-year career. He can be reached at (payson@catalysisgroup.com).

Back to Top
 
 

Member Comments
Add Your CommentExpand Comments
 
Comment:    
by Randy Tangco 4/7/2009

What is the most effective way to show that multi-tasking is bad if you have no access to the other work your resource is doing. Is there a quantification method to show that by spreading a resource thinly, the project is timeline is extended or quality suffers?

Author's Response:
4/7/2009    
Most of the hard data that multi-tasking is bad comes from cognitive psych - lots of experimental data that says interrupting someone who is engaged in a task requiring intense concentration results in about 15 minutes of ramp up time. Think of a time when you were down in the guts of code looking for a bug and the phone rang... two minute interrupt, then time to get your head into the problem again. When I was getting my comp sci degree, we talked about task switching overhead in the operating system... in the day task switching could account for as much as 50 percent of all CPU consumption (saving the state of the current process to handle an interrupt or bring in another process, managing I/O and memory).

Here's a fun experiment that you can try at home to prove the point to yourself: get two sheets of paper, a pen, and a stop watch. Draw a line down the middle of each sheet. We will put a list of letters on the left side and numbers on the right side.

On the first sheet:
1) Start the timer
2) Write the alphabet backwards skipping every other letter starting with "Z"
3) Write the first 13 multiples of 17
4) Stop the timer and note the time

Now the multi-tasking experiment. One the second sheet you are going to multi-task... write a letter according to new rule 2 and then a number according to new rule 3... no cheating... you must alternate
1) Start the timer
2) Write the alphabet backwards skipping every other letter starting with "Y"
3) Write the first 13 multiples of 19
4) Stop the timer and note the time

If you are like most people, the second task will not only take longer, it will have more errors (timeline extended AND quality suffers). Our brains were not designed to handle multiple abstract complex tasks at once. Changing gears has overhead. Note: Most software related activities are abstract complex tasks.

You ask "What is the most effective way to show that multi-tasking is bad if you have no access to the other work your resource is doing" - if by this you mean that you can't control the other work, you might check with whomever assigned the resource to you and ask, "What is a reasonable expectation about how much time I can expect from Sally each week?" There are a few things that might be helpful:
1) Ask Sally to track how much time she spends on your project (are you getting what you were told you could expect?)
2) See if you can arrange a time for Sally to work on your project uninterrupted. If you only get 20 percent of her time, try to arrange one day that she works on nothing but your project to minimize disruptions from her other work. Or set a time from 1-3 that she is yours.

Some pointy-haired bosses take some convincing... there is also an economic argument: Imagine Sally has 5 tasks to work on. Each task will take about 1 day. The work product from each task is worth $1K. Would you rather that Sally do one task each day so that you get your $1K each day, or have her work on all five concurrently? If she works concurrently, you delay getting the payoff at least 5 days, and if she gets sick or hit by a bus on day 4, it could be even worse.

 
Back to Top

May We Suggest...
Show All

Articles & Papers

Templates

Links

Books

Tools

Related Products
Testing Training Courses
Software Testing Certification, Systematic Software Testing, Test Management, Mastering Test Design, Just-in-time Testing

Software Engineering Training
Mastering the Requirements Process, Requirements Modeling, Introduction to the Capability Maturity Model Integration, Business-Driven Software Measurement

Agile Software Development Training
Scrum Master Implementation Workshop, User Stories and Estimation in Agile Development, Design Patterns Explained, Practical Test-Driven Development

 
Ads By Google
What's This?
 
 



Home   |   Resources   |   Topics   |   Community   |   PowerPass



© 2010 StickyMinds.com. All rights reserved.
StickyMinds.com is a division of Software Quality Engineering.
Privacy Policy    Terms & Conditions    Link to StickyMinds.com    Feedback


TechExcel, Inc.


STAREAST 2010 


Better Software Conference