Discussion Locked

Delayed project deadline

0 helpful
20 replies
Closed to new repliesThis discussion is closed. It stays readable, but it takes no new replies.

I am a project manager in the public sector with a software developer background.
We have a very high budget project in progress, and we facing the following problem:

Our supplier delivers project products just in time. Unfortunately, they work is in very poor quality, so we need to reject the acceptance and then we make a lot of replies and hints to the supplier. After they build our reflections into the stuff, they send it back, but the result is still not reach our expectations. So we reply again. And again.
Finally, the project just flush over it's planned finish line, almost two times now. My staff has other responsibilites meanwhile, taking part in other projects, and I also have three other projects to coordinate at the moment. As a result, my staff is getting unfocused, reviewing of the delivered products is getting late, and the moral is low.

I realised that I need to react, so I formally declared that the supplier must provide quality ensuranced products in a week, and they must not fail again. They replied that we are the folks who are incompetent, making too much assumptions, hints, constraints etc. and not even estimate the products in time.

At this level the conflict is seems unresolvable to me, so I am going to make a petition to our principal officer to make the rule. It will a painful step to all of us, but I can not see another way to regulate this hardly controlable and incooperative supplier. They want to invest the minimum efforts, getting the money as soon as possible, and they does not care about the project goals and the quality.

Could you provide any advice, please?

20 replies

by jamesrl

If the contract is vague and fuzzy, and a matter of interpretation, you need some help in resolving this through a mediator. Is there another public service customer that uses the same service provider? They would be a natural.

Failing that, someone in your project management office and/or purchasing office.

Unfortunately, when working with outside suppliers, you have to set longer timelines, to take into account the heightened need to review and the difficulties and delays of communicating with an outside organization.

You may have to reset the project timelines and expectations for the suppliers deliverables at the same time, painful as that may sound. First, get to a common agreement, then publish the new dates and reset it. Make sure you plan for more to go wrong....

james

by nobilis74

I can see now that I have a lot to learn! Here at my office that is a nonsense to let a project sink, reseting it and confessing that the PM has failed his job. Well, I suppose we should do this in some cases, but how could we? Our position is based on continous success, the unsuccessful manager has no more than two chances before he will be sent back to be PM administrator.
This is because out principals had to report to the local government. If he had to report failures, then his position is getting unstable, so our's are too.

Considering your advices, I am playing with the idea of reseting our expectations, and defining deliverables in a more exact way. After getting a more detailed description of our expectations, the supplier probably may not say that he could not imagine our desires. After that, when they provide with low quality deliverables, we shall be able to prove that they are incompetent or insuitable for the job. With such a proof I can stop the project without loosing or weakening my position.

Let me take a final note. Can you imagine that you managing a project without the authority to specify it's length and deadline? This is the case of me, so I can not change the dates on my own, especially since other important (and expensive) projects are depending on this project.

Thank you for your comment, James!

by tony hopkinson

and that'a a poinyter to a very big problem with big projects.
I've been asked to put forward and cost a proposal to meet a requirement. Think about double it add ten. Get told to take out my leeway because someone else is cheaper. Get the project, short 20% development time below my second number, because us devloper typea always over-estimate and besides we need the business. First casualty quality.
Exactly the same problem when you're told as a PM, we need this by then and we can pay this much. First casualty Quality.

by wayne m.

This sounds like a communications problem. First, take a quick inventory of what has taken place so far.

Carefully review the specifications that were given to the contractor and compare these to what was found during acceptance testing. If the faults were not listed in the specification, one must do a better job in preparing the specifications. If the faults are in clear violation of the specification, one should strongly consider replacing the contractor; someone knowledgeable in contract law would need to be involved. If the faults were covered, but the wording was ambiguous, again one needs to do a better job in preparing the specifications.

One needs to be very precise in identifying the acceptance criteria to a contractor. Nothing can be assumed nor unclear. The responsibility for clear communication relies on the purchaser and the responsibility for meeting the specification falls on the contractor. Analyze the problem and then take the appropriate action.

by dr dij

like iRise

or sophea's prophesy

they both define the required end product (software) in visual terms and actual screens similar to rad prototyping but with actual reqt's underneath, all visually, so real easy to get correct reqts from users 1st time and for developers to match the required functionality

by nobilis74

Hmm... You may be right, althrough all of these choices means to me that I am failed my job.

1. Lowering my expectations is unethical, I think. I have a strong opinion of what the supplier could do. I was partly created the software development rules and corporate standards of IT department. Because I am the one who is looking after the project from selecting the best supplier to prepare the sw architecture plans and implement the software, you are probably right when you suppose that my expectations are high. How can I ensure the quality any other way? Should I really lower the expectations just because the supplier is neglectic? I am getting confused now...

2. Finding a new supplier is almost unthinkable after a public tender has finished. This supplier was selected by it's application. (Honestly, there only was two applicants.) Since the funds are originated from public taxes, we can not refund another tender in the same scope. I think I need to find a resolution within this project, since I think that I made some vast mistakes when I let the supplier to make us play his game of giving low quality pieces of products time over time.

3. Bringing the process in-house is just fine. In that case we have full control, excellent communication, clear responsibilities. I would love it. Unfortunately, we have not enough human resources, since we are only managers and experts, managing outer suppliers.

Thank you for your thoughts, they bought me aspects that I did not count on before.

by rojackson

The problems that you are facing are classic. We have deadlines for a reason. If the vendor cannot deliver the expected quality on time and they have failed to do this repeatedly it is time to fire them and move on. You will not get out of this trap, with the selected vendor.

I suggest Tony Collins' book, Crash: Learning from the World's Worst Computer Mistakes (Simon and Schuster). You'll see variations of your project repeated over and over.

A good PM knows when to tell the Sponsor that it is time to quit wasting money! That's part of your job.

by tony hopkinson

Things went nipples up a long time ago. Whether you haven't been clear in scope and requirements or they've gone off half cocked without appreciating them, I have no way of saying. Real world experience suggests a bit of both. Prioritise, get something workable out of them and close it off.

To be quite honest this sounds like you've gone down the big bang approach , something I've seen fail so many times it's untrue. Phased, iterative approaches reduce the risk factor on all sides and while it in theory takes longer and costs more than a totally successful big bang, one deliverable at the end approach, I've never seen a succesful one. At best it works, does exactly what you said, pity it took so long it's not what you need anymore.
You've got to come to some sort of agreement though, unless you awarded the job to a complete set of eejits, for every fault you can point out in their effort, they'll be able to point to a change in scope or requirement, that caused them a problem. In that case both of you have failed to keep control. The other regular scenario is where the price/timescale is set based on what is required, not what can be achieved. Been on the recieving end of that one too, in your position and theirs.

by sridharpandu

Suggest the following to bring the project on track.

1. Do a bit of micromanagement untill the project is on track by reviewing the daily status. A kind of JAD (Joint application Development). Be active in all Project management meetings.

2. State your quality parameters. (Remember that aesthetics and product quality are seperate things). Quality is measured by number of defects per function point. So state the number of defects you are willing to accept. Usually 1 major (not critical or showstoppers) defect per function point is acceptable quality. To measure the number of function points visit http://www.ifpug.org.

3. Revisit the estimates of Size (Function Pints) and the effort (Person Months). If you feel the FP count is accurate determine the effort
effort = FP * productivity
Productivity = number of delivered FP / Person month

A productivity factor of 0.6-0.8 for technologies without RAD feature like Java, C, C++ is an acaceptable industry standard. Developr 6i, PowerBuilder, VB etc 1 FP/person month is acceptable. Dont tinker with this industry standard as there is only a certain amount a human can achieve and humans dont always work to the maximum capacity like machines nor is their efficiency at its peak through out the day.

4. Redraw the schedule. Ideally a 100 pm project shoudld be completed in 10 pm with 10 people on board but humans are not that efficient and industry averages show that we work only at 70% efficiency. So the schedule should be accomplished as a thumb rule in 1.3 times CUBE ROOT of (effort) in this case a schedule should be drawn for a minimum of (1.3 * 4.64) = 6 months and not lesser than that. So you would need atleast 16 people (100 / 6) on the project. Now draw the schedule using MS Project with these number resources. Yo'll get a practical and achievable schedule. While drawing the schedule ensure that no task is scheduled for more than 8 hours. This ensures that achievements can be tracked daily rather than at the end of the week. And any delay is apparent at the end of the day if any of the task is not marked complete. Dont mark tasks as partly complete in the WBS either they are 0% or 100% complete, nothing in between. This will provide an objective asessment of the state of completeion rather than a subjective assessment.

Believe me the joy is in bringing the project back on rails not taking the easy way out by closing it or fixing the blame. Believe me the sense of accomplishment is much more.

Will continue monitoring this discussion so keep posting your progress.

ALL THE BEST.

by hozcanhan

Nobilis , I celebrate your patience if its the case as you explained. Trully profesional suppliers get it right from the first or worst case the second time. You will not be solve it without a clash . The question is what kind of clash ? A constructive clash that at the end your company loses you or the supplier loses your company . You must take written notes down , you must not leave anything in words or just telephone talk . You must have papers with checklists and a signiture on them . Those papers must be completed to the full and signed every delivery and every return . Use a digital camera and include those pics supporting your views with the reports. It will disolve all blames against you . Make a file and collect all papers in them . When the big meeting day comes , take out a paper after every sentence you finish !

by dc guy

I worked for a large municipal government for most of my life. It is frustrating. Your problem is outsourcing. There are a lot of really good people out there in the private sectore, but THEY are not the ones whom their managers assign to government contracts. You get the beginners, the slackers, the incompetents, the whiners, because the outsource firm knows that you don't have an effective method for complaining. Your processes are too slow and complex.

Public sector agencies are slowly catching on to the fact that they simply have to do their own IT in house. As difficult as it is, it's easier, faster, and cheaper than outsourcing, because public sector agencies are structurally incapable of getting good work done by outsource firms.

Eventually a really large, politically sensitive, budget-crippling project will fail irreparably due to incompetent outsourcing. Someone very high in the organization will lose their job over it, perhaps even an elected official. Then you will be given the resources to do it yourself.

by wayne m.

Having worked as a beginner, slacker, impcompetent contractor to the U.S. Federal Government, I think I can assert that part of the responsibility falls on the government side.

A lot of contractor employees are average (this is not an insult, but should be a self-evident fact) in their area of expertise. For short-term projects or as start-up for long-term projects, contracting makes sense, but getting to DC Guy's comment, I do not understand the complete outsourcing of IT. I do not understand the trade-off of losing direct hiring and firing ability while gaining the ability to pay even higher salaries. This may seem like I am cutting my own outsourcing throat, but I am really making a pitch for more inhouse positions.

The main point is that by outsourcing all IT work, the government has lost the knowledge and ability to manage the work. It is no wonder that so many of these projects veer off course.

by nobilis74

Since my first post more of you have suggested that the problem can be originated in two PM faults:
1. inappropriate communication
2. unclear expectations

So I firstly reviewed all related documents (contract, upper lever requirements, project status reports) ? not for the first time now. I did this to find ambiguous terms, expectations, common communication problems, misunderstandings, bad techniques etc. And I have found some ? not so many.

I also had to realize ? thanks to SridharPandu ? that I did not managed the project by a methodical manner. I did not used man productivity indicators, function points and factors. My approach was pretty unscientific. However, I think that this kind of micromanagement is only appropriate when you work within a team, day by day, and have detailed inputs about the real work hours per person; the effectiveness of time planning and software architecture planning; the arised and solved problems on a daily basis; the other tasks carried out by each one member etc. All these information can be collected, structured and organized to constantly make smooth corrections, and then populate your decisions to the whole project organization, while you are still managing PM documentation. It could be quite difficult.

Well, I think all above stands for in-house projects. In my case, it is an outsourced one. I can not see how many people are working actually on the project in fact, can not measure they effectiveness as a team or as individuals, etc. On the other hand, I can have they products at every milestone, and I can state that the products are all right or they need to be reworked. This is what I can do, nothing more.

So, after this long introducing, let me say what I finally did.

I forced the supplier to prepare the products to quality measurement process. They ordered to finish the documents and polish every line of the documents, otherwise the will pay a penalty, according to the contract. (This is not a real option, because ? as far as I know ? no one was punished for low quality work in the public sector. Quality is a subjective think in most cases.)

The supplier reacted fastly. They blamed me (us) for the problems, and managed a private talk with my senior officer. I do not know what they sad then, but I am not even interested.

In the next round, the supplier created a new set of products for the quality review. Well, they were unable to correctly refresh at least the content table of the documents?

This is where we are now. I plan to finish somehow this project, and correct the products using our inner resources. I was hoping that this supplier will never get a reference of us, but now I know that they already won another, even bigger public tender of us. This is because we can not make such a tender description that is able to force our expectations while it is remain legal.

Thanks again for your posts; they will be very useful in my everyday PM processes. If there is any interest, I will continue posting news of this unlucky public project.

by tony hopkinson

cheap, and now you know why.
A simple time booking system will help.
You don't have to get anal and collect info about toilet breaks. But break the project into phases, agree dealines and then ask for a report of where you are at useful intervals. Regular minuted review meetings are also a must in my opinion. You don't have to use system X or Y or the process management product of the century, but you do need to know as early as possible if it's all gone nipples up. Deliverables and dates they are to be delivered are your start, it's common sense after that.
Rule one of being a PM, never trust anyone to be as professional as you are, until they start proving that they are, then watch them closely.

by rojackson

A strong contract with a very complete, objectively measurable Statement of Work and would have helped to avoid this, along with the willingness to say NO (see my earlier post).

As to letting these guys back in, part of PMI and best PM/IT practices in general is the use of some sort of vendor maturity rating system. If you don't have one in place, get one and then make the score for previous projects apply in your analysis for new work. You can also use a two tier screening process that keeps the lowest bid from taking effect until other criteria have been met to assure quality. Or you could just insist on CMMi Level 3 to bid.

Lots of ways to skin a cat....just make sure it doesn't skin you.

by nobilis74

Yes, this is the point. Currently I am working on a common Project Statement, a more precise role definition system, and on a declared scenrio that describes what is going to happen when the supplier does not play by the rules. One of the mistakes I made is that I did not defined kickbacks: I supposed good faith.

Based on your great idea, I am going to start rating the vendors. I will even declare to them that they will be weighted at the end of the job. This may can inspire them a bit.

What a pity that I can not use legally this rating table the next time when we spread a tender.

by jamesrl

We often have this issue where someone sees a problem and as IT folks we tend towards a tool as the solution.

But a wise person once taught me that first, you fix the process, then you look for ways to automate it.

The problem seesm fundamental - lack of communication. If the contractor found the requirements vague, then they should communicate that. If they were uncertain what was required, they should ask.

Solutions such as you are proposing may help, but at the core of the problem is something more fundamental - solve it first.
James

by dilipj

Hi
This is a real challenging stage in goofed development project. Please note that s single remark from client PM can put all development efforts of service provider to soil. Following can always bring the the deliverables on track.

1. Most important is there must be a willingness to set the things right from both side. Make the environment assertive.

2. This all happening is due to poor communication. Probably developer is not understanding properly the requirements. There must be a process to verify and approve the scope.

3. List out the problems - In Crystal Clear words.
4. Break up deliverables in small groups.
5. Have distinct milestones in each group.
6. get closely associated with development/rectification process.
7. have acceptance process for each milestone.
8. Don't jump on all deliverables at a time. Set up the priorities for deliverables.

Sorting out issues one by one, systematically will definitely bring the project on track. Please note that extra efforts with responsive management is required from both side with patience.

I have brought couple of, completely stranded, multi million, corporate level developement project on track with this approach (which always feels genuine)to happy ending with completion certificate.

by nobilis74

Hi!

I believe you. I can see that you know what you say.

I thought before the same. Then I must realize that my PM partner on the supplier's side thinks differently. He has other motivations. He does not closely concerned on getting the project on track. He wants to declare the least persons to the project, he wants to finish the project as soon as possible, because he has other responsibilities too. I suppose that the harder the client's PM, the better job will be done by the supplier. And when the client has a patient approach, saying all of us is professional, why should we declare every little terms, well the project of that client will fail.

I note all of your remarks, because all of you are right. Managing a complex project is a very hard task. We need support some time. I wish to thank you this support!

This discussion is closed to new replies.