Freelancer's Guide to Upwork
Freelancer's Guide to Upwork
It has taken some time to learn what how to get good projects through Upwork. I figured I'd write a guide in case anyone else finds my learnings helpful.
This is what has worked for me.
My Portfolio
You can check out my portfolio for context and inspiration: https://upwork.com/freelancers/brycechamberlain.
Project Portfolio
Clients are more likely to buy if they can confirm that you do good work. Before you start work, spend time building a portfolio to show what you can do.
Skip this if you already have a portfilio.
- Target 2-4 public projects you are proud of.
- Find a platform that will host your app for free or cheap: Huggingface, Kaggle, GitHub Pages, shinyapps.io, Google Cloud Run/Firebase, etc.
- You can use public data to build an app. I keep a list of public data sources at https://github.com/superchordate/data-viz-talk.
- Consider finding collaborators to work on a project so you can share the work of building it.
- Use AI to move quickly and to polish your work.
Building Your Upwork Profile
Budget a few hours to build a good profile.
- Add your portfolio projects as you create your profile.
- Upwork has a process where you can have a non-Upwork client provide feedback that'll show on the site. Consider finding a few people to do testimonials for you so you aren't starting from scratch.
- Try out the premium subscription to give yourself the best chance for success, at least for a month. I think it's only $20 and gives you "Connects" which let you bid on projects. I've heard you are less likely to get matches without premium, but I had premium from the start so I can't confirm this. I can say that I've always felt like premium was worth it. If you win a project, it can be very valuable, so it's worth it to give yourself the best chances.
Finding Work
Now that you have a profile, it's time to find a project!
You need to respond quickly. I think projects get snatched up quickly but take a few days to get to contracting and turn off the search result. So if a project is a few days old, the client is probably already in talks with the finalists.
- Every day, spend 30 mins searching projects and applying for any that look like a good fit.
- Only reply to projects posted within the last day or two.
Some tips on finding a matching project:
- Setting skills on your profile should result in you getting projects that roughly match your expertise. Try experimenting with your skills if you aren't seeing the right category of projects.
- There are a lot of bad projects on there. Low payments, rude posters, bad scoping. Don't be phased - just scroll past them! Eventually, you'll find some good ones.
- Don't bid on projects outside the U.S. - time zone stress is not worth it. However, don't use the "US Only" filter since this seems to drop a lot of US-based projects due to a bug somewhere. Just scroll past them.
- Use the thumbs up/down ratings on search results to rate projects that are good or bad fit, so that Upwork can start showing you better results. I find the algorithm gets pretty good after a bit if you feed it your preferences. In particular, use thumbs-down to rate projects that pay very low amounts or that have unreasonable requirements, so that you see less of these over time.
- Cast a wide net as far as what types of projects you look for. Some niche tech will get very few applicants and this is an opportunity if you are willing to learn new tech. I often tell clients I'm confident I can learn new tech and won't bill for time spent learning, and that's led to some good opportunities.
- When starting out, you really need feedback and badges. Consider accepting low rates initially to build your profile and gain experience as a freelancer. Similarly, if business is really slow, it's probably worth bidding a lower rate to keep busy and learning/growing with projects.
Submitting a Bid
Once you find a project that you are interested in:
- Boosting your proposal: If the boost is cheap, I'll usually do it to show that I do care. Often the boost is very expensive though. It's unlikely that you'll win any specific project, so I don't think it's worth it unless it's a project you really, really want.
- Create a ChatGPT/Claude project to start building out context for AI to write good cover letters and answers to questions. Write proposals yourself a few times and give these to the AI so it can replicate your personal style. Add new answers over time so it has more data to pull from and create better answers.
- I use Grammarly premium and feed everything through it to improve the language and make it more concise. Grammarly is better at this than AI, I find. It is critical that you come off as professional and intelligent as possible. At minimum, see if you can get a free trial to try it out.
Extra Tips
- Stick with it! Earlier this year I was like "I'm done with Upwork the rates are too low" and then I scored a sweet gig at a high rate. There is good stuff there, it can just take a while to surface.
- I find I get interviews on about 1 in 50 applications and that I win the proposal about 50% of the time if I'm able to get to a live interview. If you are seeing fewer wins, consider changing something: take an interview prep course, bring your prices down, change your cover letter strategy, apply to fewer projects - ones that have a very good fit, etc.