Operator Testing

icon picker
H - Finding Homes on Zillow

High Complexity Travel: Find homes for sale given criteria

Prompt: Find me 3 home for sale listings under $1M with at least 3 bedrooms and a backyard
Personalization Setting: None
Parameter
Description
UX - Information Exchange
Slightly inconsistent execution over 2 runs. The first instance, it only asked about location. The second run, it inquired about location and any other filtering criteria important for the search. I replied with a minimum area requirement of 1500 sq ft.
UX - Ease of Use
Relatively simple to use. The operator experience is built using an embedded browser – so there is no way for it to be used on the local browser (with any saved logins, cookies, etc,) – this hints at the fact that this is a “testing only” capability and real adoption will come from the CUA model APIs.
Clarification & Higher-Order Thinking
It performs a google search for “homes for sale in oakland, ca” and clicks on the first link (zillow). It does show some planning by interacting with the user to confirm all the filters that need to be set, using a “confirm” button that sends the response as “yes, please proceed”. One follow up question I did ask was trying to understand how it categorized a listing as having a backyard (since that's not an explicit filter) – it gave a reasonable response saying it used the description along with some listing photos to determine.
Task Decomposition & Modularity
It does a better job of understanding user requirements and planning compared to the low and medium level complexity tasks.
Application Identification
It does a google search initially, then navigates to zillow and completes the rest of the execution on the website.
Personalization
No long-term personalization taken into account. I did ask a follow-up question on how did it filter listings with a backyard (since there is no explicit filter) – response was quite accurate (looking at listing photos and )
Risk Management, Intervention & Handover
Addition of the confirm button is a noticeable improvement.
Browser Navigation & Integration
Performs well in this area – no browser actuation errors as such. Used the dropdown for all 3 filters and set values accordingly. Impressive here.
Exception Handling & Reliability
No exceptions encountered. Reliability is slightly low since 2 different behaviors are noticed in 2 executions. Once again, no app specific personalization offered.
Time to Completion
Relatively faster speed of execution. Accurate links to listings provided as well.
Consistency
Credential Management
None encountered. As noted in the personalization section, the model does not ask the user to link any accounts for a better experience.
There are no rows in this table
Want to print your doc?
This is not the way.
Try clicking the ⋯ next to your doc name or using a keyboard shortcut (
CtrlP
) instead.