For higher end market (north of $900 for 7 hours of work), 100% remote, no agencies/only direct, no fixed price projects, I apply the advice I outlined here:

https://www.wisecashhq.com/blog/how-to-have-clients-find-you....

EDIT: I've been freelancing since 2005, and 100% remote since 2010. Happy to answer questions!

I run a consulting/support business around Prometheus.

I've found that giving talks and weekly writing blog posts (https://www.robustperception.io/blog/) are great for acquiring customers. Beware though it's often at least 6-9 months before any of it directly bears fruit.

A great point - it's important to do that ahead of the moment when you'll ready need to have clients (and it takes time for sure).

For consultants working with a single client, the right moment to do that kind of marketing is right now, while you're busy with your gig.

Thanks for sharing.

My big question, which I'm still struggling with, is how much to specialize.

I'm one of those guys who wants to do everything. After a few years of Java development in financial companies, I now work on real-time web UIs. I'm now eyeing distributed, soft real-time systems with Elixir, and a bit of applied machine learning.

If you look at my blog, you'll see stuff on Clojure, React, and a few other things. So far, the React stuff has helped me land a client.

My question: Is this counter-productive? Should one specialize? How does one adapt when the specialization wanes and a re-branding is required?

One trick is to separate your skills from how you market some of them.

You can remain pretty much (internally) a generalist and a tinkerer (as you are, and I am, too! but make sure not to spread yourself too thin), yet make sure to market /some/ special skills appropriately (and in very specialized/niche fashion).

You'll decide which areas to market based on how you feel the market can generate recurring revenue.

For instance, I'm a tinkerer, yet I have a couple of well-defined niches :

- in the past, "Rails maintenance work" (easy to find, and useful while my second child was still a baby)

- Ruby ETL (http://www.kiba-etl.org/) (more specialized, but very visible in my little sphere)

- "SaaS bootstrapper with experience implementing products (and their billing)"

- anything FinTech related (as in WiseCash & other projects)

- scaling your data processing (no matter the technology, Elixir, Node, Ruby, ...)

The underlying skills are all very "aligned" around products & data, yet each small niche appeals to a subset of potential clients.

Hope this helps!

I do consulting for a big company and am a bit like you.

Jump between Java, .NET and C++, backend and frontend (both native and web) depending on the project.

Also since I majored in systems programming, I tend to dabble in other languages.

On my personal experience and those on my circle of friends, some specialization is required to keep being marketable.

Switching among technology stacks inside the company is relativity easy, as it is only a matter of being available when the right project is looking for new devs.

Outside on the job market, agencies and HR departments tend to focus on the last couple of projects, so if they aren't the key technologies that are looking for, it is very hard to convince them that we also have knowledge in the specific area.

In any case, just the size of standard libraries for modern languages, make it almost impossible to know everything.

Let alone the third party tooling and most relevant libraries outside the standard.

The content of that post is amazingly on point. Anyone looking to build a clientele should follow it.

Nice, thanks for the infos :)

I also liked the infos of Brennan Dunn, but these seem mostly related to keeping customers and appearing "professional".

I had a pretty negative experience with TopTal, where the first interview/screening thing involved me solving programming puzzles on some sort of code editor. After spending an hour on a puzzle whose solution turned out to pretty much be "return len(A) - A.count(X)", I gave up in frustration.

One of the site's founders emailed me later to tell me that he knew that those questions didn't actually really test programming skills, and that they were trying to improve, but the damage was already done.

I passed their tests, spent 50 unpaid hours on a coding assignment, for which I learned react native (not a requirement) and got passed over because their requirements were poorly worded and I didn't ask for clarification (it's a coding assignment, why shouldn't I assume they can at least spell out their own requirements well.)

They'll give me another shot at it, but honestly it's not worth the effort. I discovered from the engineer that the average asking rate for the other 70+ developers in Canada is $45/hour. I'm not going to work for that - as a "top 3%" developer it's so much easier to make $65+/hour full-time, remotely. Why would I settle for sporadic work and lots of risk for less money? Do yourself a favor and steer clear of those guys if you know what's good for you.

Jesus, 50 hours? I knew that they needed a project, but I thought it would be more in the 8-10 ballpark. And for $45/hr? The way they presented it to me was "you set your own rate and you get that, guaranteed", which I'm not sure how it works, since obviously some clients want to pay less.

Yeah, you can set your own rate, but they don't send you clients if you want more than $60 and hour, unless you've seriously proven yourself with them. Even so I think for reasonable rates of $120/hour you're shit out of luck. You'll be very hard pressed to break even with full employment + benefits. Give them a skip.

I had the same experience. It's very ironic - one of the reasons you want to be a freelancer my be to escape stupid interviews with those sort of "did you pay attention that one time in uni and remember enough of the algorithm to ace this?" puzzles. And here comes toptal asking me to do the same. When I had specified I was looking for Frontend work.

Exactly. Most of my value is in telling the client "you'll never needs this. Here's something that will do the job you need right now for a fraction of the cost, and won't prevent you from scaling later on", and not the fact that I can write optimized code to find the midpoint of an array so that the two halves contain an equal number of items divisible by 3 in half an hour.

Exactly the same reason why I'm not applying for TopTal. I started their tests and realised what it is about is the opposite of what I consider to be good programming and problem solving skills.

Remember doing these tests on codility. I failed. If anyone wants to try these tests, there are some samples: https://codility.com/programmers/lessons/1-iterations/

What went wrong, was the question impossible answer with given info?

It wasn't impossible to answer, it was just a puzzle. It required you to realize some specific property of the problem, or be stuck implementing a suboptimal solution in a fiddly editor with no debugging features.

It also had pretty much zero overlap with actual software development (e.g. it completely ignored actual business needs, which is a big part of what devs should care about).

They give you two or three tests from Codility. You can prepare for them, too: https://codility.com/programmers/

It's actually quite fun to do. I succeeded only one of the three within the time limit, and sent the results of the second one an hour or two later. I still got denied, but was told I could reapply in a couple of months.

The thing is, I'm earning a nice 70 euros per hour with local work. They don't advertise the rates, but I've heard and read it's 20-30% lower than that.

I think the problem is that it's very hard to figure out if the developer you are hiring is skilled or not ... And if you do not know whether you will get crap or hight quality, you don't want to pay the premium.