Project Management Book

...chapter 10 continued

This project management book must not be copied in whole or in part to any other website. This project management book must not be reproduced in whole or in part in any form without express written permission of the copyright holder M Harding Roberts.

Team leader's tolerance

At the start of a stage your team leader says to you: "boss, I've heard about this thing called tolerance which I understand to mean that I can change my plan in some ways without getting your prior approval - but I'm not sure what I can and can't change..." How would you as project manager respond, what would you let your team leader change without your approval? Maybe you're thinking: 'nothing at all'. But if he's told you the stage will cost 10,000 hours and wants to change his plan so that it shows 9,999 hours you would doubtless be happy for him to do that. And a change to 10,001? Perhaps you'd set something like this as his tolerance limits:

Stage milestones and end date:	0% tolerance
Stage cost:	5% tolerance
Hiring and firing:	0% tolerance - he can't hire and fire team members
Change of project scope:	0% tolerance - he can't unilaterally reduce scope in order to meet dates etc.
Quality:	0% tolerance - he can't sacrifice quality to meet the date (now there's a novelty!)
Dependencies:	0% tolerance - he can't change his plan such that it would screw up other parties

How much freedom has the team leader now got to change the plan without your approval? You may be thinking 'not a lot', but the answer actually is a huge amount.

The team leader can switch tasks between team members, resequence tasks, make some tasks bigger than they were others smaller. He can make a lot of change to the plan without getting anywhere near any of those tolerance limits. There are two points here: first you are explaining to your team leader what he is allowed to do but much more important you have just explained why you pay his salary: he must move things around if that would improve the outcome. He can't change key dates, unilaterally descope, hire and fire, screw up dependencies, sacrifice quality or go more than 5% over his budget but within those limits he must manage to best effect.

If you do not have this five minute discussion with novice team leaders they are sometimes frightened to change anything in their plan in case they breach a limit they didn't even know was there. You are empowering them - indeed obliging them - to change their plan if that would be beneficial.

(Quiz. Six team leaders each have 5% cost tolerance on their bit of the plan. They all use their 5%. By how much has the overall budget increased? 5%.)

Project manager

What should the project manager do to keep a grip?

Keep his finger on the pulse - wander about, talk to people informally.
Weekly meeting with the team leader(s). See below.
Manage issues and changes - this is covered in the next chapter.
Monitor stage agreement commitments - as we mentioned in the stage agreement chapter, don't let managers forget what they have committed to you.
Consolidate sub project reports. If there are just 2 team leaders this is not too difficult. If there are 4 sub projects each with 6 team leaders the project manager will need a project office to add it all up. This does not mean the project office reports status to you (or to anyone else), it means they add up some numbers for you. Better still, in a project with, say, 8 team leaders, designate one of them as the senior team leader, or perhaps project leader, and have him consolidate the plan and status for you. Keep ownership of the plan and status reporting strictly in the chain of command, don't hive it off to an admin support office. If you do that you may find the team leaders start to disown and even disdain the plan and status data.
Revise the plan. If one team can't make their dates you may need to move resources between teams or direct another team to move their dates to reflect the first team's slippage.
Control contingency. As we discussed in the planning chapter, the last 5 days of your stage plan may initially be labelled contingency. When there is no alternative, you give some of that contingency elapsed time to a team leader, they reset their plan and their team's targets and let's say you've then got 3 days contingency remaining. The trick is to sell to the project team the tightest possible task by task plan you can get them to sign up to at the beginning of the stage and to hold in your back pocket things like contingency. Lash them regularly like galley slaves into meeting those aggressive targets, and only when you conclude the targets are genuinely impossible do you give them some of your contingency and reset their targets. Moving dates in this way is not slippage or failure, it's how you avoid Parkinson's Law. If you and the team meet the date you committed to the sponsor you have succeeded. If the team think they can spend the contingency whenever they want to they'll spend it for you ten times over in the first fortnight!
Report monthly to the sponsor. See below.

These days some sophisticated clients of large software development projects demand that their suppliers include contingency in the project plan and budget and show them (the client) where it is in the schedule. Then every month report how much of the contingency has been used and for what, and therefore how much contingency is still left. This level of maturity is rare among sponsors in process-based companies - we all have an informal education job to do to explain that contingency is there to help them make their project successful.

Project office

The project office has no part to play in controlling the project.

All they do is administer the processes, add up numbers and maybe physically produce the report you'll show the sponsor. This is not to say their role is unimportant. On the contrary it's vital. On a large project if change requests aren't administered properly there could be chaos - but the management of change, i.e. making decisions about change requests, is the project manager's job. Similarly, the team leaders produce, revise and own the plans for their teams - their plan is a tool they use to help them manage the work of their team - it is not a bit of administration to be shuffled off onto the project office.

The project office no more runs the project than the Managing Director's secretary runs the company.

Project sponsor

How does the sponsor know what's going on and keep overall control of the project?

Monthly status meeting with the project manager (or more frequently if he so chooses)
Health check reports
Stage go no go authorisation with agreed tolerance e.g.
- plus or minus 1 week on the end date
- plus or minus 10% on stage costs
Authorising revised budgets and dates

Let's say you as project manager have committed a stage cost of 10,000 hours to the sponsor and you have 10% tolerance - you must come in between 9,000 and 11,000 at the end of the stage.

But you realise at the end of week 1 that it won't be 10,000 it will cost 12,000. You've only spent tuppence so far, but you must inform the sponsor now of this projected overrun.

The sponsor reluctantly agrees to approve a revised budget of 12,000 hours with new limits of 11,000 and 13,000. But at the end of week 2 guess what? Sorry, sponsor, I think it's going to cost more like 15,000. The sponsor will conclude you don't know what you're doing.

Going back once within a stage with your begging bowl happens all the time. Going back twice as above will be a bit painful for the project manager. But the pain of going back twice is nothing compared with the pain of coming in after the end and telling the sponsor you've overspent the upper tolerance limit. If you do that you get hung, drawn and quartered (and then fired). This is an encouragement to come forward during the stage as soon as it becomes clear you'll need more budget. The sponsor must be able to control what is spent in his name before it is spent.

Bear in mind the budget, 10,000 in our example, includes the change and contingency budgets both of which are estimates of the amounts of money you think you are actually going to spend, in exactly the same way that the travel or testing budgets are estimates of amounts of money you think you are going to spend.

If the project will under run its 9,000 lower tolerance limit, why should the project manager have to come forward and tell the sponsor? So he can spend the money elsewhere. (Please email us if you ever see this happen.)

...next

Project Management Book
Copyright M Harding Roberts 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025
This book must not be copied either as is or in modified form to any website or other medium
www.hraconsulting-ltd.co.uk

Home Sitemap Contact Us Project Management Articles Project Management Book