I have been made product leader for a big international customer, designing a physical product for them that will have to perform action A, B and C. But seeing as the needs of the customer will change and we wish to include other functionality the design needs to be flexible.

Now, my German colleagues have completed a design that does A, B and C and some additional features that were not asked for but that are really handy and nifty. They presented this to the client as a part of the joint-development-agreement. They claim that the design is complete and all that is needed is some finishing touches. The problem is that after studying their design I have found a number of weak design elements and decisions that basically will lock us in one design path that may not be optimal. The weak elements are easily corrected, the wrong design path is not.

I have had my fill of monumental Charlie Foxtrots when it comes to inheriting a design and I am not keen on doing it again. I told them that changes may be needed on the design but they become defensive whenever and regardless how I voice my concerns. I believe they have invested their pride in the project for better or worse.

I have discussed the matter with the team's superior and we are in agreement, but the team does not see it this way. I could pull rank on this issue, but for obvious reasons I don't want to spoil a good working relation, we will work together again in the future and it would be nice not to have an ongoing grudge.

How do I scrap the design without spoiling the relation?

Is it even possible?

In light of some of the comments that have been coming in I believe some clarifications are in order. The product will be a cast product, any errors introduced today will stick with the product for at least 5 to 10 years before any revisions can be made. That means that I will be held responsible for any and all errors in the cast that pop up during these 5-10 years without having a reasonable chance of correcting them.

It is ludicrously expensive to change cast molds.

  • Why do you want to change the design? The product fulfils the specifications, so why would you invest time(money) to redesign a valid solution? Did the colleagues know that they should later be able to also implement action D, E, F easily?
    – FooTheBar
    Commented Dec 5, 2016 at 14:47
  • If this product design is nearly complete and is accepted by the customer, what is it about the design path that prevents it from being viable? Why is simply fixing the issues not good enough?
    – user44108
    Commented Dec 5, 2016 at 14:48
  • The design will be difficult to produce, will introduce a number of possible errors and if we commit to it then we are locked to a design path that is less flexbile and could incur us serious costs. The design cycle from idea to prototype is long in my field, and it is difficult beforehand to determine what errors we might encounter. Hydraulics or fluid power as some might call it is finicky, the less unknown factors the better, Commented Dec 5, 2016 at 14:52
  • 1
    What relationship are you seeking to not spoil? They customer or the german team that created the flawed design? Is the design path it is on going to cause a certain failure in the future? Or is it just a path you would rather avoid? Have you done an FMEA? Commented Dec 5, 2016 at 15:36
  • 1
    There may be cultural issues here (I'm American, I've worked with Germans). Once a project milestone has been achieved (design approval by the customer), it's my experience that Germans are much more resistant to backtracking and change requests that would require re-approval than Americans are.
    – John Feltz
    Commented Dec 5, 2016 at 16:50

4 Answers 4


Leave the design as it is – unless you have a really compelling reason to change it now.

When the new requirements come in, and funding to match, re-design the system to cope with the new requirements that you know – rather than the ones you think you know, but which will always turn out to be wrong.

You always know more about the problem after you have solved it, and it is rare you'll solve it the 'best' way first time. I always want to rewrite a system after version 1 (I work mainly with software). But normally, it is best to defer this until version 2. Version 2 won't be perfect either – but then version 3 will come along. And each time, you'll know more about both the problem and the solution.

Of course, you might have a bullet-proof reason why a redesign now is absolutely essential. In that case, you'll present the argument to the team (and possibly the customer) in a way they can't argue with. If you save the customer from a disastrous rollout, then you won't have to worry about the relation.

From your clarifications, it seems you might have some specific examples of things you perceive as weaknesses with the current design. Ask the design team outright how their design will deal with those specific cases. If they acknowledge that it can't deal with them, then you can find out whether it is important to the customer that the design supports those cases.

  • 2
    I always want to rewrite a system / class when I revisit it one or two years after I wrote it. Commented Dec 5, 2016 at 17:02
  • 1
    This is a little too focused on systems where the cost of change is low. Failure to adequately design a correct solution for a Mars lander is much more costly than a similar failure for a web portal.
    – Eric
    Commented Dec 6, 2016 at 17:45
  • @Eric I agree, actually. But with a high cost of change and specific requirements that cannot be met with the current design, it sounds like there are clear, specific items that can be discussed and presented in a non-confrontational but clear way. My fourth and fifth paragraphs hopefully deal with that case. Commented Dec 7, 2016 at 10:55

Enforce a production cost analysis on the design so far.

Work out how much it's going to cost to tool the factory and to manufacture the product and how much the materials are going to cost. How much it's going to cost to package, transport, each of assembly, repairs, disposal, etc.

When you've broken down the total cost of the product lifecycle, you'll have a good idea of what's costing the most and can seek to cut back on the expensive aspects that aren't likely to reap profits.

  1. Do you have enough time and resources to get the redesign done? If you don't, then either you have to fall back on the current design or you'll have to ask for more time and resources.

  2. Almost anything is cheaper than having to retool after you ship a design to manufacturing, so the best time to have a redesign done is now.

  3. Consult the team leader and your management ASAP, give them the benefits of redesign vs. living with the current design and have them onboard with your preference for a redesign, with the understanding that you'll fall back to the current design if the redesign goes haywire.The only question in my mind is what to do with the client, who approved the current design. You'll have at some point to get the client to sign off on the redesigned product rather than stick with the current design. You may need support from your own management in approaching the client. The client is the most important player. If you, the team and your management agree and the client doesn't, then the client wins. The Golden Rule in this case is that he who has the gold makes the rules.

  4. Let your team know that you, the team leadership and the management have signed off on redesigning and ask the team to get going ASAP.

You and the team are in disagreement. If you pull rank and you get a successful redesign implemented, you should look good to the team and to the management. Nothing engenders confidence in you like success.

Once the redesign is done, it meets your criteria and it's successful, make sure to thank the team for their good work and to praise the team for being willing to stand up to you for what they believe in.


It is not just a design issue.

They presented this to the client as a part of the joint-development-agreement.

You will need to also convince the customer the design path is wrong and you cannot pull rank on them. You will need to admit to the customer a poor design was presented.

If you pull rank on your team they are not going to be very supportive in convincing the customer the design path is wrong.

Maybe start over is the correct long term design you just need to be aware of the social challenges involved.

You need to start calling it a limited (not wrong) design path.

  • There is nothing wrong with the current design, except that it lacks the built-in modularity to acommodate future enhancements that would add a couple of generations to the life the product line. The current design is not a design with a future but it's not necessarily a poor design. I'd argue with the client for the redesign on the basis of the benefit of enhanced product line longevity and I'd tell the client that it's a choice between good and better. Commented Dec 6, 2016 at 17:50

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .