Hacker News Reader

Comment by SatvikBeri | Hacker News Reader

Well, I can't know for sure. I went from charging $X/week and getting lots of resistance to charging $2X/week with the guarantee and getting almost no resistance, and several clients specifically mentioned the guarantee.

I didn't ask for anything up front then, but I probably would now that I have an established career and more options (although I'm not consulting anymore.)

Edit: also, $2X was honestly well above my market rate at that point, I charged that much initially because I was already booked. But that's more evidence that the guarantee tactic worked.

View on HN

You should consider charging fixed costs instead of hourly as well. It's easier for clients to make a business decision of your cost when they've got a guaranteed outcome.

Speaking as a consultant, that introduces a new host of headaches and problems...

I'm a consultant too and hourly has many problems as well.

The core issue with hourly is it means you earn less if you work faster and smarter. This disincentive is very bad for you and the client.

Also, the client is going to hold you to your estimate anyway and once you start getting close to it you're both going to get stressed and not do your best work. The client will start securitising your time sheets, you'll start eating hours to avoid angering the client, and you'll both waste hours every week on timesheets and billing. That's all time you're not putting into the project.

I'm not saying fixed price is always the best option but I prefer them. The only time I've been burned by them was when I wasn't the one deciding the fixed price. If you're suffering from scope creep and bad estimates, there's ways to fix that.

I actually tried that! Then a project took 5x as long as I expected, mostly due to problems outside my control. Never again.

Why did it take 5x longer and why was it outside of your control? Could you have changed the terms about which part was fixed price to fix this?

Note that I charged by week, not by hour, which eliminates a lot of these issues.

The problem with the project was essentially that the data was incredibly messy, and by the time I found this out the project was already well underway. So I could have abandoned the project, lost the fee, and screwed over my client...or put in a bunch of extra work to get the project over the finish line. I chose to do the latter.

I'm not confident enough in my ability iron out all such uncertainties in future projects – software is already hard to predict at a job, and even harder with new clients. Charging by week seems like the best compromise overall.

> Note that I charged by week, not by hour, which eliminates a lot of these issues.

A fixed number of hours per week? How did you keep the client happy when they wanted to know how many weeks?

One issue is if you work smarter or faster is you're not going to earn more and there's only a fixed number of hours in the week. Your client is likely to be unhappy in the end when the costs get out of control too.

> The problem with the project was essentially that the data was incredibly messy, and by the time I found this out the project was already well underway. So I could have abandoned the project, lost the fee, and screwed over my client...or put in a bunch of extra work to get the project over the finish line. I chose to do the latter.

- Could you have asked to see the data first to avoid this?

- Could you have offered an initial paid discovery phase that would have given you time to go over things in more detail first and even prototype it?

- Could you have broken the project into chunks where some parts were fixed price and the really unsure parts were hourly?

Not a fixed number of hours per week – more like "Here's the project proposal, I estimate it'll take 5 weeks, here's roughly what I expect to get done each week." Some weeks would be easier than I expected and I'd get ahead, some weeks would be harder and I'd end up working more. If a major event happened that would make the project take a lot longer, I'd renegotiate.

> - Could you have asked to see the data first to avoid this?

> - Could you have offered an initial paid discovery phase that would have given you time to go over things in more detail first and even prototype it?

> - Could you have broken the project into chunks where some parts were fixed price and the really unsure parts were hourly?

I could, but (1) most of my projects are quite different from each other, so the risks change, (2) the benefit seems pretty small. My clients weren't clamoring for more predictable costs – mostly they wanted transparency if a project was going to take longer than expected, as well as the ability to wrap up or cut scope if that was the case.

Reply on news.ycombinator.com