Ron,
It is very often the proverbial finger in the wind. 
Depending on the size of the team and delivery required.  When I have smaller engagements there are a few rules of thumb I use.

1.  Don’t have the PM manage more than 3 engagments, preferably only two.
2.  Plan their utilization at 80% which means 32 total hours split 16/16 over two projects.
3.  Their percentage of time against the total project is 25%.
4.  Use an inverse bell curve of the PM hours high at the initiation(scope,sponsorship,risk mitigation) and high again at delivery( deliverables review, knowledge transfer, rollover) at the end and status meetings and other PM functions (status reporting, issues management, client comittment )at a minimal level in between.
5.  Full time PM required if more than 5 people or the project is fixed price and time or the client is at all risky.

Cameran

Tricky topic, as for me (custom software development) the line between project management and solution design / business analysis can
sometimes be a fine one.

I try to explicitly estimate the solution design and business analysis processes as tasks. So requirements gathering, functional spec
creation, meetings with the client over design issues, etc., become line items in an estimate.

For general project management, I count as

Answering miscellaneous questions for the team

Gathering status reporting on project progress

Communicating project progress to the client

Making adjustments to scope or resources as the project moves forward to respect whatever the most fixed element is (date or price), etc.

Invoice review and sanity check

Work product review

etc.

I tend to use a 15% multiplier on the total project hours. To be honest, this was just a swag by me of what it takes to keep a project
moving and I’d be interested in what other people think of that? I can’t imagine that it’s too high, but I could imagine being told that
I’m targeting something that is tragically low.

Regards,

Kevin C.

Hi,

We estimate our PM hours at 20% of total project effort (which I believe is about industry standard from what we’ve researched).  However, with high risk projects it tends to be closer to 25-30%. 

Hope that helps.

Kevin,

For Project Management, we use the following:

1) Small projects - less than three people, less than six months - 10% of
total hours
2) Standard projects, existing clients, modest deliverables - 15% - 20%
3) High risk projects, ERP deployments, aggressive schedules - 25% - 30%

Hope this helps.

Joe G.

Kevin,

Rather than PM hours as a % of project, for a project with at least 6 people on it, I tend to use a full-time PM.
If between 3-6 people, I use a PM for 50% of time.
If between 1-3, then, there is a Tech PM who spends up to a max of 30% of their time on PM related tasks.
For projects beyond 30 people, I would look for junior PMs as well.

By people, I also mean concurrent.

Hi Ron,
My project managers also do design, testing, etc.  I have them book their hours by milestone phase.  Design, Configuration, Testing, Migration, Live so I can see what phases they spend their time to look for opportunities for improvement.  We have a simple tool called Replicon for time capture.

My technical folks book time by module (invoice, payment, etc.).  This not only helps when looking for areas for improvement but helps with pricing future work.

Hope this helps. 

Scott S.

We typically factor in 20% for our project management time. The project mangaers are responsible for typical project management
activities which include project leadership; progress tracking; status reporting; etc.

Customers are at times resistant to having PM hours on small engagements where we sent a consultant onsite for just one week.

-Dave M.

The effort that a project manager will spend is very much dependent on:

--skill levels and experience of the team
--complexity of the product
--amount of hand holding that the client will need
--whether or not a client project manager is assigned (the client PM can arrange group and individual meetings, marshal the forces for user testing, etc.)
--whether you will use your delivery methodology or something imposed by the client.
--team dynamics
--team experience with the product
--lessons learned from other similar projects .
--etc.

The real answer is that there is no rule of thumb that will work for all organizations at all times.  The best way of estimating costs and what should be charged to the client is to keep records of type of project, original versus actual estimates of cost, amount of time spent by every person who touches the project, etc.  As you build this data, you will be able to make ball-park estimates that will keep you out of trouble during the initial sales and scoping sessions and will allow you to develop client-specific schedules of cost that avoid the client hot-spots ("I don’t pay for management—that’s your overhead!"). From this history will come a method by which you can scale effort and time to the current project.
Also from this record keeping will come the metrics that you may need when faced with the biggest fear of all—fixed price/fixed time projects for high maintenance clients.

Hope that this helps.

Regards,

Charles

We have come up with a very cost effective model for project management cost. We have several clerical tasks that we have assigned
to project coordination units, including client onboarding, project team mobilization, time reporting review, project plan update,
billing and exception handlings. In addition, we have taken the project coordination unit offshore that saved us money and expensive
time, turnaround time (12 months difference works to our advantage).

The Project Managers are relieved of clerical work and inaccuracies that may result because there is not anyone to double check the
integrity of information.

It took us 2 years to perfect this model, and each project site is updated meticulously and even clients have access to project site. I
would welcome any emails to personal email dkhaleeque at yahoo dot com .

Thanks,

Dilshad K.