8

I've landed my first contract (hooray, self employment!) and the company is asking for time estimates.

Programmers are notoriously bad at time estimates, and I know I've been laughably wrong before. It's fixed bid, so I'm not concerned about charging. I'm just concerned about expectation management.

So far I've itemized the work I need to do, estimated the time it would take, then significantly padded that time. I'm still nervous about it, though. Is it acceptable to write in time for "unexpected delays"? I've estimated 4 weeks, and I'd like to add a 5th week for problems I haven't thought of. Is that something people do? Would you balk if someone gave you an estimate with that in it?

  • 2
    This is absolutely normal. "Unexpected delays" can even be caused by bad estimates - they don't require problems that you didn't think of to occur. So it's actually better to expect them ;) – back2dos Aug 18 '13 at 17:37
24

First, congratulations on your contract! Ok, enough celebrating, let's get down to business. ;) I've been a consultant for over 15 years -- here's my advice.

In project management, what you are talking about it "contingency" planning -- and you absolutely should do it, else you are likely to disappoint your client (and make yourself unhappy throughout the project). However, you should NOT specifically put it into the plan as it's own line item -- as then when you need it (and you most likely will), it only makes you look like a bad planner, and by definition you will be behind schedule.

There is a motto you live by: "Under-promise and Over-Deliver". Set expectations (in this case delivery time) low, but not so low that the client would be put off, and then beat the timeline (and demonstrate that you are ahead of schedule).

Instead, of one contingency block at the end of the project, you should distribute contingency planning throughout the project. Assuming you haven't already committed to delivering in your estimated 4 weeks, suggest and plan for 6, with your planned 4 weeks of effort spread out evenly over all 6 weeks. This will make you, and your client much happier, as you should generally be slightly "ahead of schedule" throughout. :)

Important: You should plan progress updates / partial demos on a frequency that: 1) mitigates the risk of building what they asked for but not what they want 2) builds customer confidence while not being overly burdensome on you. Be SURE to plan for this time working with the customer, giving demos, tweaking things, etc.

In a project of that length, that is most likely every three days or so.

Finally, when you are planning the work, front-load the most "risky" or "unknown" items first and plan the most contingency for them. Risk takes many forms -- and it's most often not the technical stuff. Generally, the biggest risk is that the customer doesn't TRULY know exactly what they want. You want to get stuff in front of them early and often to ensure you are in alignment. This means prototypes, mockups and such. If there is misalignment or misunderstanding, you want to find it as early as possible! Generally, if you find this out very early (before significant work has been done), you can renegotiate the contract to work for both parties. The biggest mistake freelancers make is delivering what the customer asked for, but not what they want. You need to understand that you are responsible for ensuring that both parties are the same page.

On riskier technical items, do enough of a proof-of-concept early so that you know you won't crash into a roadblock late in the game. Fight the tendency most people have to focus on the stuff they already know to "build momentum".

Have fun with it and I hope this helps. If you would, let us know how it went after you complete the project. Good luck!

  • Thanks, I'll remove the padding week and add it in elsewhere. I'll also let you know how it goes. It's a really cool project (involving lasers!) and I'd love to brag to someone about it. – canisrufus Aug 19 '13 at 0:49
  • I really look forward to hearing about it. :) – RobM Aug 19 '13 at 1:44
  • Absolutely. I wouldn't book "Oh I didn't see this coming time". I'd just extend my estimates to absorb it. If you didn't provide enough buffer be sure to let the customer know early though! Exactly why you front load... – Rig Aug 19 '13 at 1:50
1

First, distinguish between elapsed time and billable time. You will not manage to get a full 8 hours billable every day for weeks on end, because things will happen. You'll get sick or your kids will, or you'll need to go to the bank or take your car in. You need to eat and drink and go to the bathroom, check your email and take phone calls from the client. One way to deal with this is to plan for 6 hours a day billable. If you do that, then unexpected things shouldn't make you miss your deadline, assuming the total effort that you estimated was correct.

Then there's the matter of extra effort. We call the money we don't get to bill because we estimated badly "tuition for estimating school." It hurts. You should definitely add contingency into any fixed bid. It might be 10% or even 20% if parts of the project are really vague but you're being asked for a fixed bid anyway. And be prepared to dig in your heels if the unexpected delays are things like "here's the format for the file you need - or should I say formats hee hee hee" then you need to raise a change request and arrange for more time and more money to handle it.

But sometimes the extra effort is your own fault - you did it wrong and now you have to do it over, you thought it would be simple but it wasn't - and you'll have to eat it. That's when you'll be grateful for the 6 hours a day pace - you may still have to give them an hour or two free every day making up for your mis-estimate, but you'll probably meet your deadline and end up with a happy client.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.