In 2024, Linksys overhauled its router setup experience for its mesh routers

A router is the piece of hardware that connects to the internet and routes internet data through your home.

A “mesh” router is when multiple units work together to create better WiFi coverage.

Why?

The data was painting a dark future. Sales were dropping year over year, partly due to an increasingly competitive market.*

*Everybody’s working with the same hardware and chipsets. In a sea of like offerings, differentiation would have to be created in experience, software, and brand.

Another hard truth to face was our product experience wasn’t standing up to the competition, a sentiment echoed throughout customer feedback channels.*

*Other customer feedback channels included the AppBot plugin, where all app reviews are posted to a Slack channel. Additionally, a design team member was tasked with counting Amazon and Best Buy product reviews 3 stars and below, and tallying them by topic.

A $1,000,000 problem

With urgency to cut costs, leadership turned a critical eye towards expenses and found that support calls were racking up a whopping $1,000,000 bill* annually, and steadily climbing.

*Support calls, fulfilled by a call center in the Philippines, can be calculated at 18 cents per minute.

Our support call analytics, which tracks call volume by topic, identified setup as a top-ranking issue.

While not the leading call driver, setup-related issues were strongly linked to product returns—an expensive problem—and the most critical customer reviews.

The objective was clear:
reduce setup friction and failure rates

Current Setup Audit

Linksys has a diverse product family. Mesh products come in single or multi-pack options, designed to be modular.

A single app is used to set up all products.

There is a local and web-based interface, but it’s not used for setup, only post-setup management.

The current strategy is to direct users straight to the app, to receive step by step instructions.

Users select the setup path and identify their product

Follow instructions to connect the router to the modem

Wait for phone to perform Bluetooth scan and find hardware

After setting up the main router, users proceed to add additional units, called “nodes.”

Allow app to find and add to Mesh network

The flow is repeated for the 3rd node.

Feedback: node is placed too far from other nodes

With setup complete, we have a mesh network with 3 Nodes, reporting to the app.

Another piece of the audit was going through the Setup documentation file, examining the logic and validating its implementation.

Below: the existing documentation file built in Overflow

SETUP PROS

Ability to dispense information at each step (helpful for less tech-savvy users)

Feedback loops for every step in the process with intelligent error handling

(eg: internet check failure due to no WAN port connection)

Ability to guide node placement

Capture user email for marketing purposes

Flexibility: any setup changes can be pushed to the app

CONS

Step by step method is lengthy

Some customers annoyed that they’re required to download an app and create an account to use product

Bluetooth, the chief communication method between nodes, was problematic (permissions, bugs)

Because the app was built to set up all Linksys products, images and messaging had to be generic and as a result vague

Hardware felt frustratingly slow to boot up

Lots of waiting… bluetooth scans, to add nodes to the network, to create an account, for hardware to bootup, firmware update, etc.

Conclusions

The decision to force users to use an app for setup seemed more for the benefit of the company vs the user. It allowed Linksys to have a one size fits all solution for setting up all of its products, capture user contact info, and have ongoing engagement with customers.

The step by step process felt long and tedious, punctured with long periods of waiting that felt outdated for today’s expectations of consumer electronics.

Ideation

With logic investigations still fresh, I huddled with other department members to propose and discuss improvements.

Much of the logic had not been revisited in years, and the list of potential improvements began to pile up.

For larger PDF view, click on image

Do we want to continue forcing users to create an account?

Why do we have to use the app for setup? Is there any value in bringing back the web interface?

Do customers need to customize their WiFi?

Can it be done later?

Why is the Firmware update in setup? Why not after?

Can child nodes be added AFTER users come to the dashboard? Why or why not?

Why are all these OS permissions required?

Why does scanning for unconfigured nodes with Bluetooth take 5 minutes? Can we get it shorter?

This led to a pared down version, pictured below

Do users have to see the end user license agreement/terms here?

Despite several improvements, the same time-expensive risks were still present.

Our chief app developer communicated concerns with Bluetooth, the primary method of hardware communication, reporting complicated bugs and edge cases. The 2 minute golden path wait was long enough - users quickly grew frustrated if they had to try again.

Plus, the experience still felt cumbersome. So many app screens. And waiting. Hardly the stand-out experience to get a competitive edge back.

Executives reviewed the changes and pushed for more improvement.

Average question:
How do we reduce setup time?

Average result:
Same strategy, minute improvements

Incredible question:
How can we setup this product in 2 minutes?

Incredible result:
Requires bolder moves…

What is the literal bare minimum for someone to be using our product?

What network customization is essential, if at all?

(eg: changing WiFi name)

What other methods and technologies can our hardware use to communicate?

For a larger PDF view, click on image

What’s funny is this method was similar to the earliest routers, circa 1990s.

Their setup was no-frills because there were no smartphone-driven frills to be had.

But who cares? What’s truer today more than ever is today’s consumer wanting simple.

Moving forward with Option 1

For a larger PDF view, click on the image

I began designing the printed Quick Start guide—2-minute edition—to replace 100 app screens.

When addressing edge cases and common errors, I made sure support data or analytics could justify the content:

eg: if 4% of setup related calls are about modem resets, and it takes a support representative 20 minutes minimum to walk users through it, it’s worth adding to the start guide

!

Some users leave their old modem plugged in

!

Some users need to reset their modem to install a new router

!

Connected ethernet cord in LAN port instead of WAN port

!

Troubleshooting if nodes can’t be found. (Placed too far away, haven’t finished booting up)

Exploring options printed directly on the box

I tested the guide with prototypes in several user’s homes to drive refinements.

Final design:

Printed bi-fold format allowing users to carry the guide with them around the house during setup
More real estate to fit trouble-shooting & light behavior feedback loop
Easier storage to keep setup instructions
Colors enhance instructions

3-step version for products with nodes

With the quick start guide in a good place, we shifted focus to the interface.

The strategy was to replace the old app with a new, browser-based interface.

Several reasons lead to this decision:
No requiring users to download an app (a common complaint)
No more risky bluetooth bugs
No more expensive maintenance of the iOS and android versions of the app
Increased privacy by eliminating the ability to log in to a home network remotely