The index of completed tasks.
A real browser agent attempts real customer tasks on live airline sites. A deterministic harness verifies every outcome. Sites are ranked by the tasks the agent finished.
- Measured airlines
- 7
- Task completion
- 10%
- Honest gates
- 3
Route 1 of 8: King Khalid International Airport to London Heathrow Airport
Three gates. One completed journey.
A site can be discovered and understood and still fail before the task is complete. Last Mile tests every gate with a real browser agent.
-
Discoverability
Can agents find you?
Surfaced and reached
-
Understanding
Can agents read you?
Structured path · legible inputs and errors
-
Interaction
Can agents operate you?
Unpaid task completion · consent wall passed
Every gate receives one verified outcome
- Fails
- Struggles
- Completes
These are the winners: the sites where agents finish the job.
Ranked by the tasks a real browser agent completed, not by how clean the static markup reads. Each row keeps its Layer-1 scan alongside — how the scan and the run compare is on the methodology page.
Read the Index. Apply the playbook. Test it in the Playground.
The public ranking of measured sites, task by task, with each site's static scan kept in view next to the agent's verdict.
Playbook Turn friction into best practicesEvery friction point an agent hits maps to a best practice — the tiered changes that move a site up the maturity model.
Playground Watch the levers move the scoreFriction Airways is one airline in three readiness profiles: same tasks, same scorer. Flip the levers and completion swings from none of the tasks to all of them.