Before
- Lots of detail about Upwork in general, with no mention of offerings specific to Enterprise
- Relied on Enterprise customers identifying with brands
- Worked heavily to sell the product before a sales conversation has happened
- "Get a Demo" CTA could feel like a big commitment
After
- Clearly grounds Enterprise customers so they know they're in the right place
- Concise and to the point, focused on a few main ideas
- Leads with value to clients rather than features
- "Schedule a Call" CTA is flexible, and shows awareness of client needs
Other improvements
- Redesigned the following confirmation screen for qualified leads to more clearly set expectations.
- Designed screens for unqualified leads to give them context to why they didn't qualify for the enterprise offering and provided other options.
- Added a more prominent link for freelancers who accidentally ended up on the page.
Before
- Generic "Get Started" CTA didn't speak to the headline or the overall focus of the page
- CTA redirected to a 2-step sign up process where users select what account they want (client or freelancer)
- Sign-up messaging did not reflect where someone is coming from or confirm that they're heading in the direction they expected
After
- CTA matches the action in the headline
- CTA pops up an on-page modal for a 1-step sign up for a client account
- Sign-up messaging acknowledges the content of the previous screen and the action someone takes to get there, and sets expectations for what is coming next
Reflections
It was easier to get a read on the impact of an isolated change like a new CTA, but harder to separate two significant changes to the next step.
And though most of the traffic to this page is prospective clients, there were some prospective freelancers as well. With the new sign up defaulting to a client account, we saw an increase in freelancers signing up for the wrong type of account.
Problems
Top-rated freelancers tended to get more invites than they could ever hope to accept, while freelancers who were newer to the platform had a difficult time finding enough work.
Hypothesis
Allowing freelancers to refer other freelancers will:
- redistribute invites to those who need them the most
- help clients fill jobs even if their first choice isn't available
- create a foundation for building freelancer networks
Research
We first presented the idea to freelancers as something based around helping fellow freelancers find work and helping clients fill jobs. There was a lot of support for the idea even without a monetary incentive.
Solution
Throughout the process, we spoke to the value behind helping people find work or hire. We also provided guidance for how to include useful information for both parties about why you were referring someone for a job.
Results
Although the initial MVP did not see a large amount of users choosing to refer a freelancer, there was a significant impact on fill rates for jobs where an invited freelancer referred another freelancer.
A majority of the messages included in referrals were thorough and thoughtful, often times apologizing for not being to take the work, providing an explanation about why they weren't taking this, and detailing why the referred freelancer was great for the job.