Gigable Freelancer Profile
Enhancing the freelancer profile for informed hirer decisions
SUMMARY
GIGABLE is an online service platform for the freelance gig economy, connecting businesses with networks of freelancers such as food delivery drivers, waiting staff, security guards, warehouse operatives, kitchen porters and event stewards.
The platform allows merchants, or hirers to post jobs and to select, manage and pay freelancers. It also allows freelancers to accept jobs, send invoices, communicate and receive payments.
Gigable’s current freelancer profile needs more information to give a hirer confidence when hiring. Current features are limited to:
Profile pic
Average rating summary (non-interactive)
Performance statistics
Licences held
Vehicles owned
Bio
Favourite
Report no show
Messaging
THE PROBLEM
There’s not enough information in the freelancer’s profile for a hirer to make a confident decision when hiring. This was established by the company through feedback from the merchants, and from internal research. The key question to address on behalf of the merchant is:
“How confident are you when hiring a freelancer?”
I was tasked with enhancing the freelancer profile so that the hirer can make more informed decisions. With the goals and objectives of the organisation already established, the project is more stakeholder-centric than user-centric design as it involves aligning the design with the stakeholders’ desired outcomes and ensuring that their needs and concerns are addressed throughout the design process.
As a follow-on and related project, we aim to improve the onboarding process for the freelancer to make it easier to use.
THE BRIEF
Create wireframes to redesign the freelancer profile for:
Informed merchant decisions, fostering better collaborations.
Simplifying the profile onboarding process for easier sign-up.
THE PROCESS
Competitive Analysis
I conducted a competitive analysis with 11 companies of relatively similar scope to identify profile features shared, not shared, and missed. They included Fiverr, Airbnb, Uber, Airtasker and Upwork and I compared their profile features with Gigable. Over 50 features were identified across all platforms. Some features not used by Gigable appeared on several sites, and would be obvious candidates for exploration. Features which were used in a high percentage of cases would be explored further, and initially it could be assumed that what worked well for them could also be deserved of our attention.
While certain features would not be relevant to us as the competitors spanned different industries, it was a fair assumption that features which appeared on a high number of sites and not on ours should be taken seriously. These included nationality, customer reviews, ID verified, platform level, compliment badges, achievements within the company, cost/rate, languages spoken, skills, portfolio/gallery, and satisfaction rating. Others, like gig category, gig title, response time were most likely less relevant to our model.
One particular feature piqued our interest, a ‘Fun Fact’, used by Uber as an icebreaker to potentially spark conversation between a passenger and driver. Although only used by a single platform, we thought this feature could portray the human side of the freelancer, and it was decided to rule it in for now and test it.
However, one feature was omitted by all except one platform, yet is deemed important (according to UX research findings by Baymard Institute) - only Fiverr featured a ratings distribution summary.
Above are the profiles from Gigable and Fiverr, comparing their features. For the full deck click the link below.
Link to Figma competitive analysis report
As-is and Possibilities
Focusing initially on the mobile platform, I compiled a list of all features considered relevant at this stage which could potentially be used in the profile page. I prepared wireframe mobile screens in Figma, with an as-is and possibilities comparison. In order to compare and contrast features on a like-for-like basis the as-is screen shows the features currently in use, yet bears no resemblance to the current visual design of the application.
This was an exercise to test with the team in Gigable to identify which features should initially be ruled in or out, and to explore structure and hierarchy while considering the information architecture.
As-is, and possible features
Survey
We compiled a survey, as it would be cheap and quick, and easy to get feedback from a large customer base. Or so we thought…
Collaborating with the sales & marketing and design teams, our survey consisted mostly of closed questions rated 1 to 5 on a Likert scale, with a couple of open questions to gain some qualitative data.
The survey was distributed by email to almost 500 merchant clients. The response rate was below expectations by some distance, with less than 20 responses, well below the number needed to gain truly informed insights. However, the product manager informed me that they already had insights into many of the survey issues from previous feedback and research. I was not made aware of the research methods, nonetheless, I was assured the survey, along with the existing research would provide good data to make informed design decisions.
We asked “How IMPORTANT are these aspects of a freelancer's profile to YOU while selecting freelancers? - star rating, performance (no shows etc), profile picture, bio, time on platform, vehicles owned and licences or certificates. The options given were Not Very, Somewhat and Extremely.
Performance (no shows etc) was extremely important to almost 90% of respondents. Furthermore, when asked to rate “A score that shows a freelancers reliability would give me confidence when selecting them” 88% agreed or strongly agreed.
Gigable’s platform currently shows completed, cancelled & no show stats as numerical values. While redesigning the UI could enhance this feature, a Reliability Score, which would be a percentage number based on these statistics, could be introduced to reduce the cognitive load when trying to gauge reliability and performance.
Star Rating was extremely important to 78% of respondents. Furthermore, when asked if they would find a Ratings Distribution for a freelancer useful when choosing a freelancer, 67% selected 4 and 5 on the Likert scale. Likewise, when asked if they would like to see written Reviews from other businesses about the freelancer, all respondents selected 3, 4 or 5.
The following excerpt is from research carried out by Baymard Institute, and while it may be argued here that freelancers are not ‘products’ there is a valid correlation - there is a customer (merchant) trying to make an informed opinion about a number of products (freelancers).
“In our latest large-scale UX testing of product pages, it was clear how important product reviews are to users: 95% of users relied on reviews to learn more about products.
In particular, a product ratings distribution summary at the top of the reviews section allowed users during testing to get a feel for how the product has been rated overall. In fact, the ratings distribution UI was the most utilized feature of the reviews section, and was relied on by users even more than the actual review content.”
Baymard Institute
Ratings Design UX Research: 5 Requirements for the ‘Ratings Distribution Summary’ (65% of Sites Get it Wrong)
Our existing platform already has star ratings, but not reviews and ratings distribution. They would now need to be incorporated into the design, and tested. Typically the star rating is in a prominent position and is clickable to reveal these details.
Features which respondents mostly agreed or strongly agreed with as being important, included:
Confirmation the freelancer verified their identity
Compliments given to the freelancer by other businesses (e.g. that they were a "Good Communicator")
Time on the platform (when joined)
Last gig on the platform
Language skills
Primary occupation outside of Gigable
A list of skills had the majority weighted at the upper end.
A photo gallery showing examples of a freelancer's gigs, experience or skills was evenly split across the spectrum, proving difficult to gauge its importance.
On a personal note, I really liked the fun fact used by Uber (as did the stakeholders) as it gave an opportunity to inject a little warmth and humanity, but the survey revealed the majority strongly disagreed it would give them more confidence in hiring…surely a missed opportunity!
Ideate
Having given careful consideration to our findings, and collectively deciding what to include and what to omit, it was time to visualise our discoveries and I created a series of wireframes in Figma. Deciding it best to pick a single platform to keep things simple for now, we chose mobile. We also started a series of collaborative workshops on Miro to discuss and create further iterations which were conducted over a number of weeks.
The first draft was a features dump to include any and all features with potential - more emphasis on inclusion and less on layout.
Star ratings would be clickable, linking to employer ratings and reviews. Teams is a new feature we wanted to adopt, where a group of freelancers work for the same merchant. A freelancer’s achievements within the company, CV, presentation video, location & availability, plus website and social links are features we wanted to explore.
A reliability score was very important to us as we needed to convey a profile’s performance and reliability simply and effectively. Taking cues from the existing UI, I included a graphical illustration and breakdown of the ratings in a bar chart. The reliability score is a percentage number calculated from the performance stats.
Existing UI, above, now represented as a bar chart graphic
Miro workshop screenshots
The prototype went through multiple rounds of refinement, gathering feedback with continue testing on a number of iterations and discussed at workshop sessions.
As well as addressing individual features, UI design and information architecture was considered.
Some of the feedback from draft 1 included:
Move the profile picture left to create space for information on the right
Reliability score bar, taking a cue from Apple’s UI storage bar graphic we could incorporate the performance stats into this and delete the performance bar illustration.
Move teams, ID verified icon, social icons, education, occupation and last gig to the top
Ability to see ‘last’ applicant or ‘next’ applicant in navigation
In the 2nd draft, placing the profile pic to the left of the screen allows for more information in the top fold. The title has changed to show the number of applicants to the current gig. The performance graphic is gone without any loss of relevant information, the star rating graphic is simplified and the ID verified icon is in a more prominent place.
Another iteration followed, and feedback included:
Reliability score - it was questioned as to whether the bar graphic was in fact needed at all so it was decided to remove it and place the percentage number in the top section. The performance breakdown could be added to a drawer which could be expanded, or possibly presented in a different way.
The favourite icon to change from a heart to a star, for consistency.
Hierarchical structure, certain features to be moved up or down accordingly.
PROTOTYPE
Throughout the design thinking cyclical process I created a number of iterations which were tested, reviewed and reiterated. To view the full screens click the link below.
Next Steps
And this is where the project currently lies…
The organisation have shelved the project for the moment, as internal focus shifts priorities. I’m told it will be picked up again, as improving the product’s feature is seen as a priority business goal.
There were limitations from the outset:
The objective from the start was to improve the product, but there were no metrics available to show why it wasn’t working as well as it might. As a result, no success criteria could be established.
To define the problem clearly at the outset I would have liked to have conducted usability tests and inteviews with the product users, but time constraints prevailed.
The survey response rate was too low to be confident in its findings.
While each iteration of the wireframe was tested, it was with the same small stakeholder group, as opposed to actual customers.
Nonetheless, we conducted a large amount of research, analysed and took learnings from multiple platforms and improved upon every iteration. The feedback from the organisation has been very positive, and we feel we have added real value to the product.
The wireframe prototype is at a point where remaining interactions need to be completed, along with additional functionalities such as multiple profile applicants to the same job. Ideally, I would like at that point to conduct usability tests and interviews with a number of merchant clients to both validate the designs and gain further data to take the project forward.
Regarding the onboarding project to come, it’s clear there will be a lot more personal detail required from freelancers in the next update. For this to happen it will be necessary to improve the current clunky onboarding process. The system will have a far greater degree of success in obtaining these details if we present the user with a smooth, intuitive, enjoyable and frictionless process.
More to follow…