How QuoteRunner is measured

50 simulated small-business owners, each with known true facts (and 1–2 facts they don't know), talk to the same agent this site runs. Every field it records is checked against the truth.

Showing: run 2 · 2026-11-08 · 50 conversations · agent gpt-5-nano (reasoning low) · commit cd5506f-dirty · “first full run”

98.9% field accuracy
0 invented values, in 0 of 50 conversations
78% conversations finished
10.8 questions on average, vs a 25-field form
$0.0040 agent cost per conversation (+$0.0013 for the simulated customer)

What the numbers mean

Sample conversations

Accuracy by field

FieldCorrectAccuracy
Deep fryers12/13 92%
Offers delivery12/13 92%
Seats12/13 92%
Serves alcohol12/13 92%
Annual revenue49/50 98%
Industry49/50 98%
Alcohol share of sales (%)6/6 100%
Business name50/50 100%
Pool on premises13/13 100%
Licensed capacity (children)13/13 100%
Employees50/50 100%
Business vehicles50/50 100%
Claims in last 3 years50/50 100%
Tallest roof worked on (ft)12/12 100%
Residential or commercial12/12 100%
State50/50 100%
State licensed13/13 100%
Work subcontracted (%)6/6 100%
Transports children13/13 100%
Uses subcontractors12/12 100%
Torch-down / hot work12/12 100%
Years in business50/50 100%

By industry

daycare100.0%
other100.0%
restaurant96.9%
roofing99.3%

By speaking style

changes answer100.0%
many facts93.5%
rambling100.0%
state names100.0%
terse100.0%
typos100.0%
vague numbers98.7%

Every conversation in run 2

#IndustryStyleAccuracy QuestionsFinishedProblems
1 daycare rambling 100% 3 yes
2 restaurant typos 100% 4 yes
3 other many facts 100% 25 no
4 roofing vague numbers 92% 5 yes wrong: Annual revenue
5 daycare state names 100% 10 yes
6 restaurant changes answer 100% 9 yes
7 other terse 100% 23 yes
8 roofing rambling 100% 4 yes
9 daycare typos 100% 8 yes
10 restaurant many facts 100% 3 yes
11 other vague numbers 100% 25 no
12 roofing state names 100% 9 yes
13 daycare changes answer 100% 10 yes
14 restaurant terse 100% 8 yes
15 other rambling 100% 4 yes
16 roofing typos 100% 3 yes
17 daycare many facts 100% 3 yes
18 restaurant vague numbers 100% 11 yes
19 other state names 100% 25 no
20 roofing changes answer 100% 9 yes
21 daycare terse 100% 6 yes
22 restaurant rambling 100% 3 yes
23 other typos 100% 2 yes
24 roofing many facts 100% 3 yes
25 daycare vague numbers 100% 8 yes
26 restaurant state names 100% 25 no
27 other changes answer 100% 25 no
28 roofing terse 100% 9 yes
29 daycare rambling 100% 10 yes
30 restaurant typos 100% 5 yes
31 other many facts 100% 25 no
32 roofing vague numbers 100% 9 yes
33 daycare state names 100% 25 no
34 restaurant changes answer 100% 4 yes
35 other terse 100% 3 yes
36 roofing rambling 100% 8 yes
37 daycare typos 100% 7 yes
38 restaurant many facts 58% 25 no wrong: Industry missed: Seats, Serves alcohol, Deep fryers, Offers delivery
39 other vague numbers 100% 5 yes
40 roofing state names 100% 25 no
41 daycare changes answer 100% 7 yes
42 restaurant terse 100% 9 yes
43 other rambling 100% 25 no
44 roofing typos 100% 4 yes
45 daycare many facts 100% 7 yes
46 restaurant vague numbers 100% 7 yes
47 other state names 100% 25 no
48 roofing changes answer 100% 7 yes
49 daycare terse 100% 10 yes
50 restaurant rambling 100% 4 yes

Run history

RunDateNotePersonasAccuracy InventedFinishedAvg questions
4 2026-11-08 repeat of run 3, same code, to measure run-to-run variance 50 99.3% 0 90% 7.5
3 2026-11-08 after summary fixes (state names, hide industry other, classification hint) 50 99.8% 1 98% 6.8
2 2026-11-08 first full run 50 98.9% 0 78% 10.8
1 2026-11-08 smoke test 3 100.0% 0 100% 5.0