As a freelancer, how do you figure your compensation for complex projects?
Break the work into smaller milestones with clearer deliverables
that you can estimate. It ultimately comes down to “this will take me X
amount of time, which is Y dollars. Are the things in this milestone
worth that to you?”
将复杂的大项目拆分成多个可以评估的阶段。每个阶段可以交付的产品，你需要付出的时间，以及价值多少，并和甲方客户商讨。
Your clients need to say their budget out loud. They need to identify how valuable this problem actually is to them.
客户需要大声讲出他的预算，他们需要明确这个问题的实际价值。
I frequently have conversations where a client will say (for
example) “this is worth it to me if we can get it done in 8 or fewer
weeks (of budget), but not if it takes more than 10 weeks.”
在和客户的谈话中，我经常听到客户说：如果这个不超过8周完成就值得，超过10周就不做了。
If the problem has a bunch of unknowns or areas of risk
(feasibility, lack of design, really ambiguous scope, a really
wishy-washy stakeholder who’s going to change their mind a bunch of
times, etc.), then we try to address the biggest questions as fast as we
can. If I have to pull the plug （拔插头）on the project, I want to do it a week
or two in instead of a month or two in.
如果这个有特别的的未知风险（可行性、缺设计、范围模糊、老板还在不停的改主意），那么我们会试着尽快解决最大的问题。如果我们不得不停掉项目，那么最好一到两周就定下，而不是拖一到两个月。
Sometimes solving the problem is a lot more expensive than it’s worth. I’d rather find that out quickly.
有时候解决问题的成本比它的价值高，我更情愿的快速的发现这一点。
Ultimately it comes down to getting your client to name their budget
(and expected value). Then you gotta define the problem well enough that you can figure out if it’s even possible. Then set milestones and
deliverables at each phase (you deliver work and your client pays you!).
Each milestone has an expected timeline and cost, as well as a cost
cap.
最终，问题其实是，让你的客户说出他的预算。然后你需要很好得定义问题，看是否可行。然后设置每个阶段的里程碑和交付计划。你交付工作，你的客户付钱给你。每个里程碑有一个预期的时间线和花费和成本上限。
Sometimes you get into it and figure out that a given milestone is
way more complex than you thought. Communicate proactively.
有时候，你会陷入其中，发现某个里程碑比你之前想得复杂。主动沟通。
If you do it this way (in relatively small chunks, with frequent
user acceptance, proactive communication, and full view of the budget),
and then deliver reliably and regularly, clients tend to happily pay
invoices regardless of how you bill.
如果你愿意这样做（相对小的模块，频繁的用户验收，主动沟通，全面的预算）可靠定期的交付，无论你怎么计费，客户都会愿意支付的。
Clients just don’t want to pay a bunch of money for something that
might not even work, and usually aren’t savvy enough to know what to
look for here in terms of picking a good freelancer to do greenfield
work.
客户只是不愿意花一大笔钱买一些甚至可能不能用的东西， 通常他们也不够有足够的见识，不知道应该如何着手来找好的自由职业者做开荒的工作。
如果项目被缩减，你还能拿到原来的说的报酬吗？
Yes. I typically set milestones to be roughly every week or two and
invoice every two weeks. The client pays for work done thus far. If I
make an error in the work that's unambiguously my fault, I'll fix it for
free or at a reduced rate. Sometimes I'll fix things at a reduced rate
or comp some hours as a show of good faith.
是的，我倾向于设置里程碑，大约一到两周的客户为目前的工作付费。
如果是工作中明确了是我的错误，我会免费修复或者降低时薪。有时候我会降低时薪或者赠送一些小时，以示善意。
Most "errors" are actually "We said we wanted <this feature>
but actually now that we see it we want <that feature> instead."
大多数的“错误”，实际上是：我说我们想要“这个功能”，但是在我们看到它后，却想要“那个功能”来代替。
I get around that now by working really hard to understand the
problems and the various solution fitness criteria before I let clients
tell me about the solution they want... usually the solution they have
in mind is adjacent to but not quite the solution they actually need.
Ideally we craft the solution together.
在客户告诉我他们的想要的方案之前，我会努力去理解问题，和各种解决方案的适用性。通常他们想的解决方案和实际解决方案相仿，但是不完全同。理想情况下，我们一起制定解决方案。
If a client prefers to hand me mock-ups and specs and sequence
diagrams and things like that instead of doing that discovery work, I'm
going to build it exactly as specified... if they complain that
something doesn't work how they want, we'll go to the documentation and
I'll ask them to identify for me which requirement it doesn't meet (and
how). Most clients who want to work this way are not skilled enough at
their job to be able to work this way.
如果客户给我实物模型、规格和序列图之类的东西，而不是要去发现工作，那么我就会直接按照规定构建。如果他们抱怨一些东西不是他们想要的，我们会看文档，然后我会问他们，哪些需求没有被实现（如何实现）。大部分想这么干的客户并不具备这么干的能力。
The contract I used to use (I work for an employer now who handles
contracts) specifically stipulated that even if the project gets
canceled, I'm still owed the invoiced amounts.
我的合同明确规定，就算客户的项目被取消，我以及付出的劳动的钱还是要支付。
I also put a cap on how many hours per week of work I'll do (with an
option for written approval of going over the limit), which protects
both me and the client.
我设置了我每周工作时间上限（可书面申请超过上限），保护我也保护客户。
So far it's been fine. If I think a client is at risk of not paying,
I either won't work with them, or I'll work in much smaller units of
work, or we'll do something like "33% up front, 33% at the agreed-upon
halfway point, and 33% when the project is done"
到目前都还好，如果我认为一个客户有不付钱的风险，我会不和他合作，或者设置更小的工作时间单元，或者设置类似先付1/3，中间付1/3，项目完成1/3。