I recently read the following article in Harvard Business Review (HBR):

THE OVERCOMMITTED ORGANIZATION: WHY IT’S HARD TO SHARE PEOPLE ACROSS MULTIPLE TEAMS — AND WHAT TO DO ABOUT IT
BY MARK MORTENSEN AND HEIDI K. GARDNER  Sept. - Oct. 2017

The main points raised:

1. - Multi-teaming is the practice of sharing resources across multiple projects
2. - Developed in an effort to optimize human capital utilization in an effort to increase efficiency (reduction in idle time for employees) and reduce cost
3. - Benefit of knowledge dissemination between teams

Issues raised:

1. - Unrelated projects are not delivered on time because shared resources are putting out fires on other projects 
2. - Projects become interdependent 
3. - Loss of team cohesion 
4. - Stress, fatigue, and burnout 

Background data:

1. - HBR shows that 81% of teams are multi-teaming 
2. - 95% in knowledge intense industries 
3. - Sr. Managers can be working on 25 projects in a given week

I certainly agree with the issues they raised; HOWEVER, by approaching the issue only with a  human capital lens they focus on the wrong solutions to the  root cause, and therefore miss a key operational insight (and solution).

Setting up an Example 

Companies are trying to reduce cost.  HBR talks about a Senior Manager having 25 projects that they have to deal with per week.  Let’s look at a simpler example where a person has 5 projects that they have to work on each week.

1. - We will call the projects Monday, Tuesday, Wednesday, Thursday, and Friday for simplicity.  
2. - Each project team has 5 full-time team members, plus the multi-team member.
3. - Each team member has a weekly salary of $2000, and the multi-team member has a weekly salary of $3000.

First Scenario (Baseline)

Let's look at a baseline case where we have an expert serving each of the five teams.  The weekly project cost will be to $13,000 per team which implies  $65,000 per week in project cost.  The expert is only utilized 20% of the time so she has 80% un-utilized time.  We will assume the projects are due in 7 weeks.

Baseline Project Timelines

If we look at it from a cost accounting standpoint and total the employee expense we have the following graph:

It will get more exciting.  I promise 😉

Let's look at the Multi-Teaming example.  Looking at this from a project cost standpoint each team has a weekly cost of $10,000 for the team + $600 for the multi-team member.  Adding up the 5 project teams we have $53,000 per week in project cost.

Based on cost accounting logic and the idea of utilization maximization we are saving $12,000 per week and all people are 100% utilized.  Awesome!

Sample results for the same scenario above.

We are delayed 6 days on every project if we use the same logic as the baseline project schedule.  Lets look at the costs:

Looks like we saved $36,000 in the process.

There is one small issue with this analysis.  There are multiple studies that show multitasking causes a switching penalty (see references below).  It has bee shown that people switching between tasks can require 40% of their available time to get up to speed.

If we use the same scenario as above let's see what happens with the multitasking switching cost.

We are delivering almost two and a half weeks late across our projects.  Digging into the project cost:

With multitasking and multi-teaming we are now close to $25,000 above where we were with the expert on every project scenario.

One other small detail.

We are doing these projects for a reason.  They will generate revenue or cost savings when they are done.  For this example each of these projects will produce $10,000 per week when they are complete.

Here is a summary of the total cost of doing each project if you include the lost revenue from the project delays:

Now I have just invented this scenario, and there are many ways that you could manipulate the outcome.  However, I have been very conservative here.  If many companies have one person working on 25 projects throughout the week this scenario only multiplies.  Also, I don't know many projects that are completed in just 7 weeks.  The longer the project the more opportunity for compounding delays when multi-teaming and therefore multitasking.

Solve for (X) Insights

1. - Try and have complete teams.  This will eliminate all of the issues raised by HBR, and will likely have the lowest total cost in the end.
2. - Run the numbers.   See what the impact will be for your set of projects if you have to multi-team/multitask
3. - In any non-production environment (e.g. you are not making Fords) no person or resource should be utilized more than 80%
   1. - The more unknowns and/or creative the projects the lower the utilization expectation should be
4. - If you only have one person that can do a job or task, you must give them enough excess capacity to handle unforeseen challenges and multitasking switching time.  50% utilization might be a good starting point.
   1. - To do this you must prioritize and select only those projects with the highest priority or impact
   2. - Again, run the numbers to see what the scenarios and risks are for your organization

I hope this was helpful and gave you a different perspective when looking at any type of operational scheduling.  We are always happy to help if you find yourself in a position where you need to make tough decisions.

Please signup below for the free video series  that covers:

1. - The 8 critical elements that control your operational success
2. - How common business metrics can kill your operational performance
3. - How to design, or redesign, your operations for success now and into the future

All the best,

Chad

1. American Psychological Association: http://www.apa.org/research/action/multitask.aspx Sanbonmatsu DM, Strayer DL, Medeiros-Ward N, Watson JM (2013) Who Multi-Tasks and Why? Multi-Tasking Ability, Perceived Multi-Tasking Ability, Impulsivity, and Sensation Seeking. PLoS ONE 8(1): e54402

2. Proceedings of the National Academy of Sciences Sep 2009, 106 (37) 15583-15587; DOI: 10.1073/pnas.0903620106. Cognitive control in media multitaskers: Proceedings of the National Academy of Sciences  (http://www.pnas.org/content/106/37/15583) Eyal Ophir, Clifford Nass, Anthony D. Wagner