Contact me

Opendoor Trade Partner Work

How I increased work volumes to get homes on the market faster

@nandi

@prianca

Why Opendoor needs trade partners

Opendoor buys and sells homes as is, which means acquired homes usually need repairs, renovations, and maintenance done before being put on the market.

Local trade partners are needed to handle jobs like plumbing, painting, cleaning, and even photography.

Project timeframe: 3 months

Role: Home Operations Designer

Why Opendoor needs trade partners

The business problem

I interviewed coordinators who assign work to trade partners and discovered they rely on a small percentage of trade partners to a majority of the work. This led to three key issues:

Uneven trade partner utilization

Some partners had too much work, while some were assigned nothing.

Uneven trade partner utilization

Some partners had too much work, while some were assigned nothing.

Inability to scale operations

Over-utilized partners were bottlenecking the entire process.

Inability to scale operations

Over-utilized partners were bottlenecking the entire process.

Homes sit in inventory longer

The longer it takes to get houses market-ready, the more it costs the business.

How might we…

maximize trade partner utilization to list houses on the market faster, reduce holding costs, and scale operations to handle larger inventory?

Strategy and ideas

I facilitated a design thinking workshop with stakeholders and product partners to brainstorm ideas, and these were top 3 ideas:

Rank our trade partners

Promote new trade partners

Give visiblity on work load

Rank our trade partners

Come up with a score to show trade partners best suited for each job based on several criteria such as performance, capabilities, and location.

Promote new trade partners

Show new trade partners higher on the list with a "NEW" badge to encourage coordinators to assign work to them and assess their work quality.

Give visibility on work load

Display work capacity comparing the current volume of assigned work orders vs. their peak capacity.

Strategy and ideas

I facilitated a design thinking workshop with stakeholders and product partners to brainstorm ideas, and these were top 3 ideas:

Rank our trade partners

Promote new trade partners

Give visiblity on work load

Rank our trade partners

Come up with a score to show trade partners best suited for each job based on several criteria such as performance, capabilities, and location.

Promote new trade partners

Show new trade partners higher on the list with a "NEW" badge to encourage coordinators to assign work to them and assess their work quality.

Give visibility on work load

Display work capacity comparing the current volume of assigned work orders vs. their peak capacity.

DESIGNS

Update trade partner data

There were two phases of the design work. The first phase was addressing an issue where inaccurate trade partner data was being logged, so coordinators couldn't assign work appropriately.

Trade Partner portal

I redesigned the trade partner portal so operators could update information on our trade partners, such as their locations, service areas, capabilities, etc.

Service area selection

I created a map interface so operators could easily add zip codes to a trade partner's service area through checkbox or click-and-drag selection instead of manually typing in numbers that could easily lead to errors.

DESIGNS

Trade partner matching

Next, I designed an automated system to match trade partners to work orders systematically. I worked with my Product Manager to define a “match score” based on the following factors:

1 Work Capacity

2 Service Area

3 Capabilities

Example of a great match!

This design shows a trade partner who fits all area and capability requirements for the work order and is not being utilized. Therefore, it rises to the top of the matching list.

Example of a poor match …

This design shows a trade partner who fits all the requirements for the work, but they have way too many open work orders, which brings their score down for the time being.

PILOT TESTING

Immediate issues arose

Usability testing on design prototypes received positive feedback, so we implemented the designs and ran a pilot test in Tampa, FL.

Before the pilot test, there were 15 trade partners being assigned work. After the pilot test, there were 32 trade partners being assigned work, which was an amazing increase. However, there were new issues:

Work was TOO evenly distributed

Trade partners who used to receive a lot of work were upset when they got less business.

Work was TOO evenly distributed

Trade partners who used to receive a lot of work were upset when they got less business.

Can't select favorite partners

Some favorite, reliable trade partners simply weren't being recommended by the system and couldn't be selected.

DESIGN ITERATIONS

Ensuring work for trusted partners

In response to feedback that the system worked too well distributing data, we added a trade partner classification to balance assignments a little.

Expert

History of quality, reliable work

Identified as a partner we want a strong relationship with

Earns "Expert" badge to indicate they're to be assigned regular work.

Rookie

Newly onboarded trade partner

Work quality and skill to be assessed

Need to be given an assignment ASAP so they see the value of becoming an Opendoor partner.

Project impact: 20% increase in work capacity

There was steady increase in trade partner utilization across every market.

We were seeing stronger partner relations as more trade partners were getting business, and the matching system opened the door to increased automation in the future.

Screenshot from an analysis report