Experiment: The A-matcher
This exercise is part of an interview with a team of one about “a mobile app that enables its users to recommend people nearby.” There are no more details than that sentence. 💪
Context —
Many websites and apps we use daily use recommendation systems that empower the user experience and drive specific actions like purchases, follow, and Recommendations.
Human or personal recommendations happen around us, in our work, our significant other, friends, followers, etc. In the professional talent field, there are three perspectives: who is looking for a job, who is looking for an employee (company), and people in between.
For this experiment, we will focus on a person in between who is willing to recommend his contacts to a company, value for the company, value for the middleman, and value to friends.
“Interslice.” They need to find an employee as soon as possible that fills the “Community Manager” vacancy and is asking their employees and “friends” for recommendations.
Company requirements: Proven Experience (+3 years), Excellent communication and interpersonal skills, Proactive personality, ability to get things done, and Fluency in English.
Finding talent is difficult for any company, especially small/medium companies; they usually ask their employees, friends, or contacts if they know someone to suggest; the idea behind all this is:
“If a friend/employee/believer refers someone; is a good suggestion, right? They know the company, the need, the culture, no one better.”
But how can we be sure that the suggestions match the company's needs, culture, or requirements and are a “good suggestion?” Finding people who want a job is always easy, but finding and hiring (A)players is never easy.
What does a friend/employee/believer need to suggest only (A)players?
Hypothesis —
The fundamental idea behind recommendation systems is a collaboration between different sources, tech + user(s), the user(s) + user(s), or tech + tech, working together and sharing information to match common tastes, likes, requirements, etc. The more information between sources, the more it sprawls towards infinity.
The solution background is to match key company requirements with user profiles to help users find A-players between friends.
Companies like Interslice will start their headhunter process through their already hired employees, especially their A-players (people with high results and a lot of trust), and there is no way to ensure results.
Companies need an app that helps any company’s believers to find A-players in their network and help companies hire the best talent in a matter of hours, “Tinder” is simple.
Performance indicators: (YC growth)
- Hiring rates from companies will increase by 5–7% weekly.
- Connections/Followers will increase by 5–7% weekly.
- Suggestions between users and companies will increase by 5–7% weekly.
1. — Do potential users/companies want to use this product? Yes, if there is any technology that will help companies reduce timeframes, efforts, and intermediaries when hiring to find A-players and people who add value, they want to use it and pay for it.
2. — Do potential users need to use this product? Yes, after seeing the results of a simple process, both “friends” and companies wanted to use it. Companies will use it to find the best talent near them and “believers” to increase their net value and trust.
3. — How often do users envision themselves using this product? Small/medium companies depend on job availability, so in the best-case scenario, they should be monthly active users. " Friends” building their net value should be weekly active users as they want to grow their value.
Alternatives Competitors —
To find good patterns when sharing people’s suggestions about people, we will use apps from different markets but do the same thing: suggest a person connect for a relationship, interests, etc. The competitors are Tinder, LinkedIn, and Facebook.
Tinder
The most straightforward app to connect people is suggested by software based on the information details of the profiles (interests, age, location, etc.). — Tinder connects me with my next significant other.
PROS (based on the Hypothesis):
- Foolproof UX allows you to say yes, no, or super yes with one finger, intuitive and clear behavior for anyone.
- The first Aha moment is when you can see a giant image of your next favorite person with the same interests.
- With a touch on the screen, you can access more images, personal details, music tastes, and information for the promised first chat.
CONS:
- Too focused on how the person looks, what about “ugly” people or people that do not follow the “perfect look”? It is not inclusive and doesn’t help.
- As a fast dating app, the information shared is not too deep if I want to dig deeper, but it makes sense as the final goal is to get a date to know better.
It is the most popular social network designed for professionals to connect and share their curriculum vitae. — LinkedIn connects me with my colleagues and professional partners
PROS (based on the Hypothesis):
- All my friends, contacts, and family are there, so it is easier to connect to someone and build my network.
- It helps me connect with people with 2 or 3 up-levels of friendship that are not on my regular radar.
CONS:
- It's too crowded, with too many things to do on the same screen; I’m unsure if all those features add value to me or my network building.
- The suggestions are out of context. Why do I want to follow a black leader or an LA resident? What’s the value for me and my network in that?
- The actions to manage, add, and remove a person from my network to grow my net value are the less critical actions; it seems what matters is added followers just because.
Instagram is an entirely visual social media platform that primarily enables users to share images or videos with their audience. — Instagram connects me with my colleagues, professional partners, friends, and everyone.
Design & Prototype —
The application could have many flows depending on the user, the vacancy, etc. For this project, we will follow the simplest method of sending a recommendation to a company.
Fast wireframes
Rapid ideation to figure out what from the competitors makes more sense to the type of users, The “Tinder” way looks good but too light, the “LinkedIn” version is too crowded, and the “Facebook” version makes a bit more sense but still too crowded.
“High” fidelity prototype (1st Iteration)
The first version of the prototype on users’ hands, with fundamental interactions, was well received; all the users could use the application, and even those who didn’t understand the language could do some necessary processes.
The flow steps are:
- Go to the list of vacancies
- Select one of the options in the list (the first one)
- Review the necessary information on the vacancy and scroll down to find the perfect match.
Feedback (1st Iteration)
In general, the feedback from 3 different users was focused on the following:
- Pills look like buttons; why can’t you click on them? Why are they there? The idea is to visually match company requirements with what the profile offers, what does match, and what adds more value without being part of the requirements (it needs to be clear).
- Why are no images? Should it have avatars or a simple category image? When people talk and think about other people, we should remove any possible option of discrimination; with avatars, the noise increases thinking about people's appearance, ethnicity, skin color, sex, etc. Without that noise, the user will focus on what's reading, the abilities, and capabilities.
- It feels simple, but the shades of blue feel incomplete, like a draft. I totally agree; the Cabana Design system is in the middle of wireframes and high-definition prototypes, so the feeling could be due to that. Next time, something less “draft/wireframes” could probably help with aesthetics.
“High” fidelity prototype (2nd Iteration)
Analyze —
After learning a bit more about recommendations and the math behind them, I am completely sure that there is a lot of space to build valuable recommendations using data from companies and possible employees. Between those, the gap between finding an employee and finding an A-player can be shrunk using technology and data.
After seeing all the reactions from C-level employees and more teams, I can see the pain of finding good people to build great teams, which is extremely important for companies and space to build a good startup.
