Helping self-employed tradesmen and construction companies connect via a digital marketplace
Project 5 at General Assembly (client project)
Team: Myself (lead), Anna, Elvis, and Mike
Client: Scout4 (Paloma)
Duration: 2 weeks
Methods: Competitor analysis, user survey, interviews, task analysis, personas, design studio, paper and digital prototyping, usability testing.
Tools: Pen and paper, Google Forms, Sketch (Bohemian Coding), Marvel, Trello, Keynote
This is how the project started.
Launching in summer 2016, Scout 4 aim to become the UK’s first ever Labour marketplace for construction companies and tradesmen. The construction industry remains in an analogue world: the companies call around to find labour, and the tradesmen call around the find jobs. Their aim is to make the construction industry a digital marketplace; enabling the companies and the tradesmen to connect via a mobile app.
Our job was to focus on the tradesmen journey through the app by specifically looking at the on-boarding process and Tinder concept of the job search.
Proposed problem: At this point we understood from the brief that tradesmen, who are often self employed, continuously move from project to project and they struggle to avoid unpaid downtime.
I was lucky to lead a team of enthusiastic and creative people for this project. We adopted the Double Diamond process which consisted of 4 phases over the 2 week sprint. I implemented and led a Kanban board to organise the workflow and delegate tasks.
Sprint 1 — Research Phase
This phase is about understanding the brief and the potential user base, and refining the researched information into a design hypothesis.
Initial target users
We began our research phase with the brief’s target users in mind; we focused our search to tradesmen who were 30–40 years old and either British or eastern European.
We began with a smart-phone friendly survey so that we could get quantitative data and to recruit participants for testing. We distributed the survey using Twitter, Facebook groups, Reddit and our contacts, but unfortunately only got 12 responses from tradesmen. Already this spoke volumes about the receptiveness of the proposed target group to the smartphone.
Interviews and findings
We compensated by conducting a lot of interviews, and in total managed to meet with 30 tradesmen and 5 employers. We arranged visits to two of the biggest constructions sites in London and also approached tradesmen on the streets while on their breaks.
From an early stage in our research phase we found that the tradesmen who were Scout4’s proposed target group “rarely have downtime” and “have never been out of work”.
They said that they have “so many calls from construction companies” and even when we were talking to them got this text about a new job opportunity. We wanted to understand why this was the case so we kept exploring the reasons why some tradesmen found it hard to get jobs.
Further research showed that having a large network of trusted colleagues was essential in finding work quickly. Across all of our interviews it became apparent that word-of-mouth was the number one way you find a job; “this game works on contacts”, if you have a large network it’s easier to find a job, if you don’t it’s harder.
This particularly seemed to be the case with highly skilled tradesmen. One interview with an employer of a self-employed work force said that “good people are always busy” which showed us that the quality and level of skill of the workman determined how easily he found work.
At this stage we felt the proposed target users did not need help in finding jobs with an app because:
- they use word-of-mouth to find jobs because they have a large and reliable network of friends,
- and were experienced and highly skilled so sought after by employers.
Sure, they may benefit from a digitalised experience but there was no real need.
A change of direction…
We thought it would be interesting to interview a different demographic, so We headed to Lambeth College to interview students who were looking for their first apprenticeship in the trade.
These young tradesmen were so keen to work for “safe and respectable companies”. They relied most heavily on job sites like Monster and Indeed. They explained how time consuming it was to keep signing in to different websites, filling out different application forms and writing cover letters when the jobs are most likely to be out of date. Since everything in their lives is available in a few clicks from the palm of their hands, they were excited by the prospect of this all happening with just one click: “It would be easier if you don’t have to keep writing a cover letter. It would be better if you could just click a button and that’s it.”
It became clear that there is more of a need to bring some ease to these young job seekers lives. We believe they could be the early adopters of the app.
We amalgamated our research into 3 personas, who we believed will be the potential user-base for Scout4:
- Andrius is our old hand. He’s been in the industry for over 2 decades. He’s highly skilled and the younger guys on site look up to him. He continuously looking to get the best paid job at any time so support his family.
- Danny is our entry level professional. He’s got a few years of experience under his belt but has to rely on a small, sometime unreliable network of friends. He’s pretty aspirational so is always wants to improve his skills.
- And finally Ben, our school leaver who desperately wants to test his new skills and work for a respectable employer, but has no network to rely on.
We chose to focus on Danny because he represents the biggest group of potential users of the Scout4 app. He is young and tech savvy but also holds a number of years experience to be useful to employers.
We wanted to re-position the target users for the app because we believe that the younger generation will be early adopters of the app. Therefore, we focused the rest of the design phase on helping Danny, because by helping his pain points we are likely to help the needs to Andrius and Ben too.
Bringing all our research together, we repositioned the original problem statement from “tradesmen struggle to avoid unpaid downtime” (we found this was not the case), to:
Younger tradesmen struggle to find work quickly in the analogue world of construction
We were also able to propose a design hypothesis: by designing an app that is built on quality, efficiency and trust we will help Danny find jobs. The Scout4 app needs to:
- have quality job results so he doesn’t waste his time look at irrelevant jobs,
- be efficient so the time between finishing one job and starting another is as quick as possible,
- and be built on trust in order to simulate the benefits of word of mouth.
Sprint 2— Design Phase
This is about taking the hypothesis and brainstorming potential solutions, and providing a final proposal through user-tested validation.
We organised and I led a Design Studio to collectively generate (with my team and our client) as many design ideas and sketches as possible for a given screen. From some simple usability testing on existing apps, we found that a Do-Get interaction model was best: users filled in a form to be able to see jobs. We chose to design the Tradesman’s Profile screen, for three reasons:
- It is the screen that companies see first when they review candidate applications. Therefore, it would help us to prioritise information such CSCS card, year-of-experience, or availability of the tradesmen.
- It is also the end-result of the tradesmen on-boarding process. Therefore, it would help us driving the concept and design of the on-boarding process.
- Finally, its layout and design would be similar to Company’s Job screen that you had to design for our prototype. Therefore, we could reuse our ideas.
Through an iterative process of designing and testing, we went through three stages:
- Rapid paper prototype — Used to quickly validate a design idea in a cheap, low cost and focused way.
- Low fidelity wireframes — Brought the paper prototype to the digital space with straighter lines but still not involving colours so as not to distract testers from core functionality.
- High fidelity mockup — Built to represent a potential final look for the app by making use of colours along with higher resolution images and icons. This is the basis of the clickable prototype.
Through design, testing and a couple of iterations, we managed to keep the onboarding process simple, effective and progressive with 4 quick steps:
- Job Requirements. Only mandatory step to show the tradesmen relevant and personal job results.
- Application Requirements. Optional but necessary to apply for jobs
- Recent Projects. Optional but increase tradesmen’s chance to get a job
- Build Your Network. Optional but allow tradesmen to recommend and get recommended
Testing for the concept of the 4-step onboarding was mostly positive; users understood each stage and with one user saying it “feels like short form”. Despite this, during the last stage of testing in the digital prototype one tradesmen noted that the “Signing-up process is too long — it’s more for the employers with the office jobs as they have more time to search”. Therefore in the next sprint it would be worth spending time testing ways to speed up the process, perhaps by making the skip button more obvious.
Designing for efficiency
Part of the brief was to address the Tinder model applied to tradesmen and employers. The tradesmen and employers would express interest, leading to a ‘match’ (see diagram below which illustrates this concept).
Results of testing showed that when users clicked on ‘interested’ that their “details have been sent to the employer”.
We therefore changed ‘interested’ to ‘apply’ which meant that we could take a step out of the whole job application process. The diagram below shows how the 4 step process has now changed to 3.
Feedback from testing was extremely positive, with tradesmen saying how “it was so easy” and “there’s less toing and froing”.
Testing the Tinder swiping model
Our client came to us with the idea that each job would be a card and, similar to Tinder, you would swipe left or right to discard or show interest in a job. As mentioned before, the layout and content was driven from the results of the Design Studio; we saw what was important for the employers to see in a tradesmen.
The content here was understood and we felt from testing that nothing was missing, but the biggest change was the gesture interaction. Users automatically swiped, but when asked what they thought would happen they thought it was to browse between jobs. Therefore we intended the final iteration to be that swipe is to browse rather than making a decision on a job.
Next steps would explore more UI patterns to encourage users to swipe to browse in order to eliminate confusion of the interaction.
Designing for trust
As we found word of mouth the most popular way to find a job, we considered it to be the main competitor for the app. Therefore we wanted to simulate the benefits of word of mouth into the app as much as possible in order to incentivise people to use it.
We did this by:
- Add people you know during the on-boarding process. Connections are imported from your phones contact book.
- Recommend a friend on the job card. This is essentially what happens in word of mouth, but by bringing it in the app it can be done quicker and more efficiently.
- ‘1 contact in your network have worked for this company’ on the job card. Encouraging users to talk to each other.
This idea was well received during testing. One user said “References and network are very important because they are proof of your work”.
Finally, as Scout4 is yet to be launched, we had the opportunity to think about the visual design of the app. We were given the Scout 4 logo and colour palette and from that we choose our app colours. We made blue the dominant colour because it symbolises trust, which we believe is one of the key element of our app.
Finally, I had fun adding some personality to the step by step process by adding a graphic of crane, which stacked bricks as you progressed through the steps.
The culmination of all previous iterations lead to a final digital clickable prototype made in Marvel:
Summary and Next Steps
At the end of the 2-week sprint, we performed a 20 minute presentation for the client and their feedback was very positive. We were pleased to have provided strategic recommendations to them.
In summary, here are the main take aways from the 2 week research and design sprint:
- Re-defined target group to a younger audience because we believe that this tech-savvy generation are likely to be the early adopters of the app and are in need of building their network.
- Incorporated word of mouth because this is the existing most used and reliable way to find jobs and therefore the biggest competitor for the Scout 4 app.
- Created & validated a simple 4-step onboarding process that’s simple & easy to use.
- Disproved Tinder model; swipe is to browse. Contrary to the brief, users wanted to browse all jobs before making a decision.
- Simplified the process from ‘interested’ to hired. This enabled the whole job application process to be reduced by 1 step.
On a next sprint, we recommended to do more research on the younger generation. Some interesting findings came out of the 5 interviews we had, but this came quite late in our research phase. Closer testing could have been done on how they used job sites in order to find what they like and don’t like about them.
Also our final round of testing led to some more next steps for the next sprint:
- Design the Profile page
- Language selection for app
- Import contacts from social media
- Explore UI patterns on job card browsing
Thanks for reading!