LinkedIn - Connection Management Feature

GROUP PROJECT

LINKEDIN: CONNECTION MANAGEMENT FEATURE

 
 
How do we design a better platform experience by helping users find and remember their connections in intuitive ways?

55% of LinkedIn users have over 500 connections, and yet lack a built-in way of organizing connections or adding notes to better leverage those connections later on. As a user’s network grows, it can become especially difficult to find older connections. Additionally, users only spend an average of 17 minutes per month on the platform, a shockingly low figure compared to other social networks.

How do we create a feature that lets users organize and contextualize their connections to make them easier to find and use later on? How can this drive platform engagement?

 

 

PROJECT INFO

2 WEEKS (OCT - NOV 2019)

3-PERSON GROUP

ADDING FEATURE TO EXISTING APP

SKILLS

PROJECT PLANNING / DELEGATION

BUSINESS + USER RESEARCH

DIGITAL PROTOTYPING (HI-FI)

USABILITY TESTING

TOOLS

TRELLO

SLACK

GOOGLE DOCS / SUITE

OTTER.AI

WHIMSICAL

SKETCH / SKETCH CLOUD

INVISION


MEET THE TEAM

ME

 
Luca.jpg
 

USER INTERVIEWS + CONSENT FORM

SKETCH SYNTHESIS

APP SCREEN INVENTORY

HI-FI DIGITAL PROTOTYPE

USABILITY TESTS

 
 
Camila.jpeg
 

USER INTERVIEWS + SCREENER

PROTO-PERSONAE

STYLE GUIDE

HI-FI DIGITAL PROTOTYPE

PRESENTATION SLIDES

 
 
 

RESEARCH PROTOCOL + SYNTHESIS

BUSINESS + COMPETITIVE ANALYSIS

USER FLOWS + APP MAP

USABILITY TESTS

PRESENTATION SLIDES

 

We each wanted to draw on our respective strengths while being able to sample all parts of the process: Camila and I tackled the design process, while Morgan drew on her anthropology background to lead the research process. All three of us participated in the initial sketching, affinity mapping, project planning, and strategy decisions.

We also wanted to make sure that we were able to resolve differences of opinion. Most important was communication, so we checked in daily for at least 15 minutes, though we often spent considerably longer together when it came to hashing out design decisions.


RESEARCH / UNDERSTANDING THE COMPETITION

Many social platforms and contact management systems include organizational capabilities, so we conducted competitive analysis to determine the ways in which these were implemented and how we might incorporate them effectively. We found that flexible tagging schemes were widely-implemented features that would be familiar to users and well-suited to our challenge.


RESEARCH / UNDERSTANDING THE USER

USER INTERVIEWS

We conducted thorough research in order to understand LinkedIn users. We first created a screener to gain insight into potential interviewees’ platform familiarity and usage habits, then interviewed five users, all of whom:

  • Had over 500 connections (we wanted to know how they managed large networks)

  • Used LinkedIn at least several times per week (we wanted some degree of platform familiarity)

  • Added several connections in the past month (we wanted insights into how users navigated the connection process)

  • Had used the platform for at least several years (we wanted to know how people remembered older connections)

 

AFFINITY MAPPING + SYNTHESIS

We then created an affinity map, which brought to light a variety of user behaviors, desires, and pain points.

IMG_7743.JPG
 

USERS REMEMBERED CONNECTIONS IN MANY WAYS:

  • Name

  • Employment / education history

  • Date of connection

  • Mutual contacts

  • Closest association / where connection originated

  • Face / avatar

USERS WANTED ORGANIZATIONAL FLEXIBILITY:

  • Users wanted to organize by when and how they met contacts, as well as other memorable characteristics

  • Users wanted to add personal contextual notes to contacts (some wanted to be prompted at the time of connection)

  • Even those users who self-identified as “lazy” recognized the value of well-organized connections

USERS WANTED A MORE RELEVANT PLATFORM EXPERIENCE:

  • Users didn’t like “noise” from random posts and notifications, and only wanted updates about their close connections

  • Some users had become more selective about adding connections

  • Users felt the app and website were cluttered and unintuitive

 
 
If there was some kind of genuine way for the platform to rekindle relationships… because who cares if you’re connected to 1,100 people if they barely even remember who you are
— Interviewee
 
 
I definitely add a lot of random-ass people and sometimes I’ll just be scrolling through my feed and be like ‘where the hell did I meet this lady?’ like I don’t remember her at all
— Interviewee
 

SYNTHESIS / CRYSTALLIZING THE USER PROBLEM

LinkedIn users with large and growing networks need a way to organize their connections in order to better remember and leverage them.
— User Problem Statement
We believe that by providing LinkedIn users with ways to view, add, search and organize their connections by customizable contextual information and shared attributes, we will make it easier for these users to both understand the context behind their connections and to look up specific contacts using a variety of naturally memorable attributes.
— User Solution Statement
We will know this to be true when LinkedIn users with many connections report fewer issues finding, remembering, and associating contacts.
— User Solution Hypothesis

SYNTHESIS / CRYSTALLIZING THE BUSINESS PROBLEM

By providing these additional contact management features to professionals actively using LinkedIn, we believe they will value the platform more and engage each other more consistently through the platform.
— Business Solution Statement
We will know this to be true when LinkedIn users with many connections spend more time per month managing their network and communicating with other members via the platform’s “InMail” direct messaging and other content sharing tools.
— Business Solution Hypothesis

SYNTHESIS / PROTO-PERSONAE

We created two proto-personae to further synthesize our design challenges and to keep our target users in mind as we navigated subsequent parts of the project.

 
user.png
 

PERSONA 1: OPPORTUNITY SEEKER

NEEDS

  • Remember who I am connected with on Linkedin

  • Organize contacts to help with my job search

GOALS

  • Find a job

  • Get referrals from Linkedin connections

PAIN POINTS

  • I forget how I met someone who’s part of my Linkedin network

  • People invite me to connect on LinkedIn, but I don’t know why they want to connect with me

BEHAVIOR

  • Go to networking events

  • Message hiring managers

  • Message connections for referrals, informational interviews, etc.

  • Search, save, and apply for jobs

  • Moderately comfortable with LinkedIn’s website and app

 
user-2.png
 

PERSONA 2: NETWORK SUSTAINER

NEEDS

  • Organize contacts to help me remember who they are

  • Look up people from my Linkedin connections based on the context of how we met

GOALS

  • Remember contacts I haven’t communicated or worked with in a long time

  • Reconnect with people effectively

PAIN POINTS

  • I can’t find a person that I’ve worked with because I can’t remember their name, but I know that they are part of my Linkedin network

  • People invite me to connect on LinkedIn, but I don’t know why they want to connect with me

BEHAVIOR

  • Have many connections on Linkedin

  • Browse content on newsfeed

  • Use Linkedin messaging to reconnect to former coworkers and schoolmates

  • Moderately comfortable with LinkedIn’s website and app


DESIGN / DESIGN STUDIO

We began with a design studio, rapidly sketching ideas and iterating on them. Notably, I thought to use existing profile information like experience and education to create automatic groups and enable a form of “smart” browsing. Though this feature evolved through various iterations, it provided valuable “out-of-the-box” functionality to our final prototype.

The beginnings of the smart tagging feature

The beginnings of the smart tagging feature

Design Studio - Luca Iaconi-Stewart - 05 shadow.jpg

DESIGN / SKETCHING

We then created our own sketches based on everything we’d discovered up to that point. Afterward, we reconvened and discussed what we liked best about each of our ideas before creating a master paper prototype for usability testing.

DIVERGING: SKETCHING OUR IDEAS

Various forms of context

Various forms of context

Adding tags to a connection

Adding tags to a connection

A refinement of the smart tags feature

A refinement of the smart tags feature

 

CONVERGING: SYNTHESIS FOR PAPER PROTOTYPE

Adding tags to new connection

Adding tags to new connection

Browsing tags

Browsing tags

Context summary (Camila’s sketch)

Context summary (Camila’s sketch)

Tags in search results (Camila’s sketch)

Tags in search results (Camila’s sketch)

 

DECISION POINT: SHARED TAGS

Should tags be public? We debated at length whether to add this functionality.

 
 

PROS

If the user initiating the connection were able to pre-fill the custom tag and share it with the recipient, it would halve the total amount of work required to categorize the connection, while encouraging organization on both ends.

CONS

We realized there would be significant technological hurdles to successful implementation: What would happen when the initiator used a slightly different tag spelling than an existing tag the recipient already had? This would likely lead to organizational chaos on both ends. Additionally, what would happen when one party wanted to modify the tag?

 
 

In the end, we decided that in the interests of time and complexity, tags should be kept private.

 

ITERATIONS / USABILITY TESTING

We conducted three usability tests with the paper prototype, asking users to complete four separate tasks:

  • Adding a new connection and entering contextual information

  • Batch adding connections to a new group

  • Finding a connection without remembering their name

  • Looking up contextual information about a connection

IMG_7797 - blurred.jpg
IMG_7801 - blurred.jpg
IMG_7807 - blurred.jpg

DESIGN / HI-FI DIGITAL PROTOTYPE

Screen Shot 2019-11-18 at 3.38.50 AM.png

We took all our usability testing results into account and created a high-fidelity prototype in Sketch.

To better organize our workflow, Camila created a style guide for the existing brand and design aesthetic, while I created a version of the user flows to identify the number of distinct app screens we would need to build or modify.

P3 - User Flows@2x.png

ITERATIONS / ADD TAG

The primary source of confusion involved adding a tag at the time of connection, which was unsurprising given that it was a completely new and unfamiliar feature. We cleaned up the copy and changed the flow to better inform users of the value of tagging their connections.

Paper prototype

Paper prototype

Initial digital prototypes

Initial digital prototypes

Add tags - Individual B2 - shadow.jpg
Add tags - Individual B3 - shadow.jpg
 
Final digital prototypes

Final digital prototypes

07c Add tags - Individual - Enter Tag Name@2x - shadow.jpg
08 Add tags - Individual - Confirm Tag Addition@2x - shadow.jpg
09 Tags Added@2x - shadow.jpg

ITERATIONS / CONTACT CARD

The other source of confusion was the “contact card” functionality, which featured highlights about the relationship; some users were confused about its role and the fact that it was separate from the profile page. We decided to integrate it directly into the profile as an expandable section to make its function more intuitive.

Should we make an independent contact card?

Should we make an independent contact card?

…or just keep highlights on the profile?

…or just keep highlights on the profile?

Relationship highlights - collapsed

Relationship highlights - collapsed

Relationship highlights - expanded

Relationship highlights - expanded


ITERATIONS / MANAGE MY NETWORK

Users also expressed distaste for having to tap “My Network”>“Manage My Network” to access key connection management functionality, so we removed the “Manage My Network” screen entirely and nested all functions (Connections, Tags, Groups, Pages, etc.) directly within the “My Network” tab.

Two screens were required before to access connection management features

Two screens were required before to access connection management features

We brought functionality up a level onto a single screen

We brought functionality up a level onto a single screen


 
 
 

THE FINAL PROTOTYPE

We created an improved LinkedIn experience that encourages the user to tag their connections, while also providing useful default tags for those who don’t want to spend time organizing.

The interface is similar enough to feel familiar, while addressing key organizational and contextual shortcomings of the current app.

 
 

TAKEAWAYS

  • Group projects are wonderful. As someone who is used to doing solo work, I cannot stress enough how much I loved brainstorming and iterating on each other’s ideas. I felt we created much better solutions when we leveraged our diverse backgrounds and mental frameworks. I also liked the group accountability, and the ability to “multi-thread” our research and design processes.

  • Test early! I was grateful to have conducted paper usability tests, as we caught a variety of issues before we’d taken the time to bake them into the digital prototype.

  • Set time limits on certain activities. As much as I loved our group discussions, at times we allowed them to go on unchecked (e.g. affinity mapping and sketch synthesis). Some of our time would have been better spent on things like usability testing and iterations.

FURTHER EXPLORATION

  • Nested tag functionality — what if a user wants to get even more specific and create sub-tags (e.g. a cohort at General Assembly)?

  • Shared tags — can we share tags publicly as a way of encouraging organization platform-wide?

  • News feed customization — can we leverage tags to tailor what a user sees in their news feed and cut down on unwanted platform “noise”?