Project Management

Overcommitted? Face the music and talk to your client

When you realize that you've overcommitted on a project, Chip Camden says you need to come clean with your client as soon as possible. Find out what other tips he recommends for handling this dreaded situation.

You use your most diligent combination of research, rock(et) science, and haruspicy to come up with your best estimate for the project. You pad the estimate a bit, and then give it to your client, who happily signs off on it. You start right to work and everything goes according to plan for almost an entire day, and then...

  • You discover one very large task that you either forgot or never knew about, without which the project cannot succeed.
  • You realize that one of the project's activities is going to take a lot longer to implement than you thought.
  • Your prototype, on which you based your estimate, proves to be inadequate -- to the point where that entire approach will have to be scrapped and re-engineered.
  • Your estimation was based on a thorough knowledge of 95% of the problem, but the remaining 5% proved to be entirely within the wrong end of the Pareto principle.

Your palms start to sweat, your head swims, and you start to feel a lot like when you're in over your head. But this time it's not that you don't know what you're doing, it's just that your estimating was epically off. There's no way you can finish by the date you quoted the client.

This is bad news for both you and the client. If you're billing hourly, then it's really bad news for your client. If you're doing this on a fixed-price contract, it's still disappointing for your client, but it's horrible for you. Your first impulse is, "Well, I'd better get cracking!" You put your head down and start working as hard as you can. This seems reasonable, but it often leads to these three undesirable actions:

  • Not communicating with your client. You don't want to waste precious time with status updates. Besides, you'd like to cover this whole mess up until you can make up the extra time.
  • Procrastination. When you feel overwhelmed by a project, it's easy to let yourself get distracted from it. You end up working on anything except that train wreck in progress.
  • Burnout. You begin to hate the thought of starting work every day. Living under a bridge is starting to sound more attractive. You look for ways to escape, and if you're not careful, you might find them.

All of the above will only slow down the project even further. Try some of these instead:

  • Tell your client the hard truth. Maybe they'll be able to work something out, like allowing you to revise your estimate. Or maybe they can add someone else to the project to handle part of the load. Or perhaps some other piece of the project can be postponed. Or just maybe, they'll think of something else you overlooked that makes the whole problem go away. And even if it turns out to be just as bad as you think (or worse), they're better off knowing now rather than later -- and at least they should appreciate your honesty.
  • Keep your expectations realistic. "I'll just figure out a way to squeeze it in" never works, and you know that. When you say that to yourself, your trust in your own schedule evaporates. Be honest with yourself about how long each task will take.
  • Start getting things done -- but not everything at once. Don't panic, take care of yourself, and complete one thing at a time.
  • Learn a lesson or two. Think about how you could have avoided this mistake. What did you need to know that you overlooked? Did you fail to factor in a high enough Uncertainty value? And if you took a fixed-price contract for this job, do I really need to spell out the lesson you should take away from that? (Hint: dee, oh, enn, apostrophe, tee, exclamation point).

Hey, bad estimating happens to everyone sooner or later. It happens to me so frequently that I'm starting to doubt that sheep entrails have anything to do with predicting the future. So you want to buffer yourself against the ill effects of a really bad estimate, here are four tips to keep in mind in the future:

  • Bill by the hour, or some other unit that means "I work longer, I get paid more."
  • Don't give them an estimate unless they ask for one.
  • Set the right expectation with your client. "This is an estimate, and only an estimate." Tell them about all the uncertainties that you know and mention that you might not know about all of them.
  • Always estimate higher than your "real" estimate.

Do you have any "oops, I overcommitted" war stories? If so, share them with the community in the discussion.

Get weekly consulting tips in your inbox TechRepublic's IT Consultant newsletter, delivered each Monday, offers tips on how to attract customers, build your business, and increase your technical skills in order to get the job done. Automatically sign up today!

About

Chip Camden has been programming since 1978, and he's still not done. An independent consultant since 1991, Chip specializes in software development tools, languages, and migration to new technology. Besides writing for TechRepublic's IT Consultant b...

25 comments
info
info

we found that ClickandCall's (www.2Callus.com) helped us comumnicate instantly with clients and improve sales and customer service. This VOIP feature enables clients to connect over their phone or PC instantly to our business. Saves us on long distance costs, too. Its the best web communications tool out there. They even take messages for any "missed" calls!

agusacil
agusacil

I often feel guilty when I made a wrong estimate,sometimes it reduces my self confidence for the next estimate. After reading this article I see that this situation occurs very often to other people too so it is a relieve :-). However it doesn't work for me not to give estimate. My customer always ask for estimate before a task is approved. Because of the underestimate experience, we tend to offer customer for a time and material basis when possible. Otherwise we will put a big buffer to cope with unexpected. Again,thanks for this article.

reisen55
reisen55

One rule I have is "never" become involved in something I know little or nothing about, particularly if I do not have a non-production environment to test. Learning and client time is always a bad thing and it is usually these situations that result in OHMYGAWD situations. The concept of coming clean is good but BEST BEFORE everything goes to pieces. Warning the client in advance that there are unknowns involved, and that the project plan (I now subscribe to that - write them up and submit to client) can be either solid as a rock or built of vaporous air. Critical systems require extra care. My associate became involved with VoIP phones over the summer for a new client - a lawyer - and blew that up because he had no prior experience there and thought he could learn it on the fly. Nope. I do not do phones and would have found a qualified consultant on that one. Come clean, tell the truth and outline in terms the client can understand (no acronyms or tech talk) plus a bit of goodwill (bring food) and perhaps a discount if you can afford it. Do not over-charge.

shane.freman
shane.freman

I've heard this a number of times in the last couple weeks, "Don't give your customer an estimate." I'd like to know who your customers are and what I can do to find some of them. Typically in the SMB space, customers aren't going to give their developers a blank check and in most occassions you're up against other developers in an RFP situation. I would love to hear some feedback about how you are selling your services without giving an estimate on either price or time. I would also love any info on providing estimates rather than fix pricing anything. It could be entirely possible that I'm looking at this all wrong but it just seems like our customer base doesn't fit this model very well.

ssharkins
ssharkins

rock(et) science -- I love it! ;) Because I work hourly, I don't run into this problem too much. Another thing that keeps me out of hot water is the type of projects I take on -- they're all small -- a few weeks work of work rather than months or years! That makes estimating a lot easier, so I just don't run into this too much. In truth, it's one of the reasons I stick with smaller projects -- trying to estimate a huge project truly is rock(et) science and I flunked that course! ;)

Sterling chip Camden
Sterling chip Camden

Have you ever quote a fixed price, and then made a bundle because you were done much sooner than you thought? I can only remember one such project (of any size) in my career. I felt a little guilty for taking all that money. Very little.

Sterling chip Camden
Sterling chip Camden

There is really no need to feel guilty or inadequate -- the key is to learn from past mistakes and improve. And it doesn't hurt to build in a buffer to protect yourself. The only downside to overestimating is that someone else who is more competitive may get the job instead. But maybe they bid it too close and they'll be regretting it later.

Sterling chip Camden
Sterling chip Camden

That seems to be the hardest thing to get right -- knowing how much you don't know. I think it's OK to take jobs where you're learning, as long as you make that clear up front and you allow enough slack for it in your estimate (if you provide one at all). As long as everyone understands what's going on, nobody should get in a fuss over it. But often, consultants will know that they need to learn some piece of the work, and they'll make a guess on how long that will take based on extrapolating from what they do know -- those are almost always underestimated.

ssharkins
ssharkins

A small project is easy to handle this way. The specs are short and most of the code already exists in some library. It might take less than a week to develop, test, and debug, but I tell them it will take about 5 to 6 weeks. They're happy when it comes in early on on time, and that gives me the flexibility to work the project into my existing schedule. I don't recommend this method -- it's just what seems to work for me. There's no model for it, it's based solely on my experiences. Where are these clients? They are harder to find than they once were, that is definitely true. It's not my estimation method that's the problem, I think this stuff is staying in-house now.

Sterling chip Camden
Sterling chip Camden

Mine are software developers, so (a) they understand the problems of estimating somewhat, and (b) most gigs are ongoing. Since I bill hourly, the estimates I do provide are usually crystal-ball-budgeting- oriented, rather than do or die. SMBs that aren't software companies will have quite different budgeting needs. One of my clients sells to government agencies -- there's no way they can avoid fixed-price.

Sterling chip Camden
Sterling chip Camden

Definitely good advice. But even on hourly projects, a bad estimate can throw off bigger project plans and sour your relationship with your client. There's still the temptation to cover over the error and try to make it up on your own, which for me at least gets rather discouraging. I find it works out better for everybody just to be open with your client and let them help you help them if they can.

Sterling chip Camden
Sterling chip Camden

this is a test of line wrap in techrepublic when using the Chromium browser, otherwise known as Google Chrome.

reisen55
reisen55

Here is a standing issue. One of my best clients is an optical house I have supported for over 10 years, and right now they have Dell Poweredge 1800 server that I upgraded, and really two networks, 10 computers that they bought without my imput about 3 years ago. Dell Dimensions. Why? They were cheaper than Optiplex. OK, I lived with this and someday they will upgrade. Dell just ran a promotion for a Vostro 200 with a 19" monitor for $329 (good price) and a $99 downgrade to XP Professional. We can Ebay most of the existing hardware to mitigate the cost. But - is there a DRIVING BUSINESS NEED to do this? Nope. Everything is really working well and I have no headaches. I can increase to gigabit speed with a small switch, and the 17" to 19" is a nice thing, but that is about it. There is nothing drastic failing there because I do my job and keep this place humming. Every so often they ask me and I honestly have to admit there is no NEED to do it and there will ALWAYS be a bargain price on another day. Could I argue and make a ton of money? Yes, but I am not going to do that. EVER. I saw that happen at MicroAge of Mahwah NJ which is why I got into independent work.

a2makarov
a2makarov

Overcommitted - this is something that happened to me permanently. I know one my bad sides (!) - is being too optimistic. That's why working as an in-house consultant is better, because you don't feel that pain described above that close. :) What I really like in your articles is the message - be honest with you and your client, this is the only way to be a "solid" person and professional.

road-dog
road-dog

I don't get into these uncomfortable situations. If someone in the customer organization comes to me with additional requirements, I refer them to my POC for their concerns. Then I speak with the POC to identify this new requirement as gotta have, or full of crap. Gotta have means I update the estimate as best I can. Don't feel guilty for getting a bundle, after all part of your service was the ability to do the job. They paid for your expertise. If you have shortcuts that get it done faster, then that delta between you and the competitor translates to dollars.

reisen55
reisen55

And you may recall my other posts on various subjects regarding my talented, enormously gifted and totally eratic partner who got into a VoIP phone system for a new client, a lawyer, and detonated the whole system for 2 weeks, wrecked the relationship, and spent hours doing free work to avoid a lawsuit. Beautiful thing. I would never have gone there, EVER as I know nothing about VoIP and that is a CRITICAL FUNCTION of an office. You do not mess with these areas of business lightly. I regard phone technicians with total AWE. Their closets of cables make our little RJ45 hell-holes look easy no matter how many Cat5e cables are coming in.

ssharkins
ssharkins

You know that as soon as you posted "Pay No Attention..." we all read this! :)

Sterling chip Camden
Sterling chip Camden

In brief, "how much do you recommend that the client allow you to do?" Your approach is ethical, trying to save the customer money rather than making more for yourself. I think that will more than pay for itself long-term in business built from trust.

pinoysgotit
pinoysgotit

have had experienced working on projects were we all felt the project (the requirements) were not at all feasible but was still pushed through. as a lead i took part in making estimates, but knowing the team capacity (technically and man-power) i knew it would be impossible yet we were forced by management to still proceed. what ended up was we rendered sooooo many overtimes (even overnights!) for a month just to get the project done. oh yeah most of the time we make the deadline - but the quality was unfortunately soooo poor. this stuff happens and normal developers like myself cant really do anything about overcommitted/overconfident estimates (management) :( http://itpinoy.blogspot.com

Sterling chip Camden
Sterling chip Camden

It is truly amazing how much energy you can save if you don't fabricate any part of your story.

Sterling chip Camden
Sterling chip Camden

So the real lesson here is not to paint yourself into a corner with a commitment to a bad estimate.

reisen55
reisen55

That was a valid invoice and the sales rep confirmed Mt. Sinai would buy it. They did until they phoned Belkin and that account left MicroAge. There were other issues there such as in 1993 or so when a woman came in with an already old IBM Classic PC that needed a hard drive. We had an old old ST-251 on the shelf and I was prepared to give it to her for $20. Nope, management said $200. She bought it and I gave her as much shareware as I could to make up for the lapse. That is why I got into independent work, to give back to customers the outright theft and deceit I saw at MicroAge. That the chain collapsed and is now an IT service firm is no surprise to me.

Sterling chip Camden
Sterling chip Camden

... and I wouldnt' call that a "lapse", unless it was just a data entry error on invoicing omitting the decimal point.

reisen55
reisen55

The cloient looks for guidance and here is a problem too. I do not have firm rules on this one. They will upgrade sometime, but I do not have a business REASON to do so beyond a general dislike about the Dell Dimension systems as a business choice. I have experienced nothing but hell on occasion from these low cost boxes. But again, not enough to justify a total dump. Clients need guidance and as their corporate IT department (which is what I am), I would nominally argue more emphatically so that the client feels assured that guidance is being offered. Orders from the High Command can make the troops feel secure. That sort of thing. At MicroAge of Mahwah, I saw horrible ethical lapses. How is this: sales rep who sold Mt. Sinai Hospital a $13.95 Belkin cable for $1,300.00. No kidding.

Editor's Picks