It's better to estimate the hours required, quote for those hours and agree a completion date in advance.
With time tracking you're inviting unnecessary scrutiny and failing to establish work boundaries that can leave the client with no sense of beginning and end. It also leaves you constantly multitasking and bookkeeping.
If a client is insisting on time tracking it's probably because they're paranoid about getting ripped off or needing the insight into what they're paying for.
Upfront estimation with deadline gives them the peace of mind they need, establishes the boundaries and frees you to get on with the job.
If you underestimate do better next time, if you overestimate then great have a margarita.
That's a bit extreme. Lots of people track time and hourly billing is common in many industries. If it was self sabotage, the practice would have died out naturally.
> It's better to estimate the hours required, quote for those hours and agree a completion date in advance.
I think project-based pricing can work well if you have clearly-defined scope, if you've solved this kind of problem before and there aren't many unknown unknowns, and you invest in a lot of upfront planning and estimation.
Most of my work hasn't been like this. I'm usually filling a role, not completing a single, well-defined project. And I'm often solving problems that I haven't solved before, where a client and I are uncovering complexity as we go.
> With time tracking you're inviting unnecessary scrutiny and failing to establish work boundaries that can leave the client with no sense of beginning and end.
I could see this being the case with micromanagers and toxic clients. But I think time tracking and sharing a certain level of detail at regular intervals helps clients, and potentially other team members, understand the effort needed for various activities. That helps everyone build their sense of what it takes to complete the work.
> Upfront estimation with deadline gives them the peace of mind they need, establishes the boundaries and frees you to get on with the job.
Peace of mind comes down to trust. And trust can come from a SOW/contract, as well as from a relationship that's built through clear, regular communication and delivering incremental value over time. The contract is important, but that's the low bar of trust.
> If you underestimate do better next time, if you overestimate then great have a margarita.
In my experience, software project estimation is very difficult and the risks grow with the size of the project. At a certain size, if your estimates are wrong for a flat-fee projects, there may not be a "next time."
It's currently fashionable to hate on hourly billing, but I personally don't know how I'd make it as a consultant if I switched to project-based pricing. And yes, I'm familiar with Jonathan Stark's writing on the topic. More power to people that can make that work!
