bolt EVVisor
Pilot proposal
One API connection. The car's data in your tariff.
Rating factors from the EV's own data, for individual drivers and corporate fleets. With the driver's consent, no hardware, straight into your rating engine.
open_in_newLive demo/insurer
1 / 13
Turkey · EVs and kasko
The sector is already asking for this.
“…a system that charges a higher premium to the risky driver and a lower premium to the safe driver.”
10.1% → 17.5%
electric share of new car sales, 2024 → 2025 (ODMD)
18.1%
of cars registered Jan–Aug 2026 are electric (TÜİK)
2.6%
of the 18.0 M cars on the road are electric (TÜİK, Aug 2026)
2.3 M
EVs on the road in 2030, EPDK mid case (Mar 2026: 412k; 2035: 5.6 M)
ⓘ TSB 2025 Sector Report p. 62 · ODMD press bulletins, 6 Jan 2025 and 6 Jan 2026 (battery-electric) · TÜİK motor vehicles, 16 Sep 2026 · EPDK EV and charging projection, Board Decision 14510, 22 Apr 2026
shield_lock KVKK · explicit consent · no location shared
2 / 13
One API
What one API connection gives you.
Per driver, with consent
- Driving Index 0–100
- 6 factors
- Data quality
- Real km
- Tyre and vehicle alerts
- Sentry Mode use
- Battery and charging
- FSD km
Dashed: read from the car today; in the next API version
Rating factors
Driving Index, real km, seat belt use
Rating + claims
Sentry use: theft risk, a record for the claim file
FSD share
FSD km / total km, once FSD is available
Prevention
Low tyre, service warnings
Sum insured
Battery health for sum insured and total loss
Fleets
Fleet risk, no surveillance
shield_lock KVKK · explicit consent · no location shared
3 / 13
Insurer · rating factors
We measure the rating factor. You set its weight.
- 111 items: rating, claims, prevention, sum insured
- 2Status per item: today or with telemetry
- 3Your actuarial decision
123shield_lock KVKK · explicit consent · no location shared
4 / 13
Insurer · tariff simulation
Your coefficients on a sample portfolio.
- 1You set a coefficient per index band
- 2Each band's effect on premium, instantly
- 3Sample tariff; no loss outcome is simulated
123shield_lock KVKK · explicit consent · no location shared
5 / 13
Fleet admin
Corporate fleets without surveillance.
- 1Connection and alerts per car
- 2Fleet totals: the same figures reach the insurer by API
- 3Individual scores need consent
123shield_lock KVKK · explicit consent · no location shared
6 / 13
Insurer panel · from the API
The API data, in a ready panel.
- 1Portfolio index distribution
- 2Consented drivers only
- 3Method, CSV and PDF in one click
123shield_lock KVKK · explicit consent · no location shared
7 / 13
Trend 2 · Insurance moves to data
What do the benchmarks measure?
| Measures | Number | |
|---|---|---|
Tesla Real-Time Insurance (US) Premium recalculated monthly from the Safety Score (0–100); FSD km score 100. Turkey's equivalent: an annual policy whose technical premium is recalculated monthly. | Price rule | Re-rated monthly |
Lemonade (US) Fixed monthly base + a rate per distance. Since 26 Jan 2026 (Arizona) the per-distance rate halves on FSD km; the base stays, so the total falls far less than 50%. | Per-km rate | −50% on FSD km |
IVASS (Italy) Insurers with ≥50% black-box policies vs none, 2019. Correlation, selection included. | Loss ratio | ≈−15 pts |
Aksigorta (Turkey, since 2020) Fiat Yol Arkadaşım Connect Kasko: an add-on device with subscription gives a driving score and days of use. At renewal, score >75 → 8.5–10% off; up to 250 TL more by days of use. | Tariff discount | 8.5–10% |
Randomised field experiment (US) 1,449 drivers vs a control group. Measured: hard braking −16–21%. CMT, the trial's telematics partner, estimates this at −7.4% loss cost: a model estimate, not observed claims. | Loss cost, estimate | −7.4% |
info Only IVASS measures a loss ratio, and it is a correlation. Our scenario uses the most cautious figure, the trial's −7.4%.
ⓘ tesla.com/support/insurance · Lemonade, Jan 2026 · IVASS, Bollettino IPER 2020 Q4 · aksigorta.com.tr · AA, 2 Oct 2020 · Accident Analysis & Prevention 225 (2026)
shield_lock KVKK · explicit consent · no location shared
8 / 13
FSD · the pricing shift
An FSD km is a separately priced risk.
Distance driven without a major collision
FSD (Supervised)
9.16M km
US average, all cars
1.12M km
≈8× more km between major collisions on FSD (tesla.com/fsd/safety, read 27 Sep 2026: last 12 months, North America, 5.69M vs 699k miles; US average estimated by Tesla)
−50%
Lemonade · per-km rate on FSD km only; the fixed monthly base stays
100
Tesla · every FSD km scores 100 in the Safety Score
−10%
Tesla · FSD discount, up to
boltFSD km is readable on HW4 cars today. FSD in Turkey: not yet.
ⓘ Tesla-reported: a collision counts if FSD was active within 5 s before impact; FSD miles are highway-heavy. · tesla.com/fsd/safety (read 27 Sep 2026; rolling 12 months, updated quarterly) · Lemonade (Jan 2026) · tesla.com/support/insurance (Safety Score v3.0) · RDW (2026) · developer.tesla.com
shield_lock KVKK · explicit consent · no location shared
9 / 13
Business case
Sector + Tesla data: what changes?
72.9%
Kasko gross loss ratio, sector, Jan–Jun 2026: TSB's latest; the August release is premium only
67.5%
EV segment with Tesla data, behaviour effect only, before discount
−5.4
points
72.9% × (1 − 7.4%) = 67.5%: the most cautious effect below. Gross, DERK excluded.
What Tesla data measures, and the published effect of each
Driving behaviour
−7.4%
loss cost: CMT's estimate from measured safer driving, 1,449 US drivers
Randomised trial, estimate · used here
Safety systems
−13% / −39%
property-damage liability claim frequency: front AEB alone / full ADAS bundle, US
Same-model comparison · HLDI, Dec 2025
Telematics on the policy
≈−15 pts
loss ratio at Italian insurers with ≥50% black-box policies, 2019
Correlation · IVASS
FSD km
≈8×
more km per major collision on FSD than the US average
Tesla-reported · tesla.com, Sep 2026
ⓘ TSB 'Hasar Prim Oranları' Jan–Jun 2026 (all kasko, quarterly) · IVASS IPER 2020 Q4 · HLDI 42-30 (Dec 2025) · AA&P 225 (2026) · tesla.com/fsd/safety (27 Sep 2026)
shield_lock KVKK · explicit consent · no location shared
10 / 13
Driver app · KVKK
The driver shares the data.
- 1Explicit consent, KVKK Art. 5
- 2Never location or routes
- 3Withdraw any time
123shield_lock KVKK · explicit consent · no location shared
11 / 13
Pilot proposal
7 cars, 90 days: a rating factor or not?
01
Weeks 0–2
Consent, connection, key pairing
02
Weeks 2–6
Index v1, first risk report
03
Weeks 6–12
Regular reports, pricing review
Success: 7 cars scored, and the index reviewed as a candidate rating factor.
What you work with during the pilot (live, sample data today):
- verified_user Portfolio panelIndex distribution, consenting drivers only, CSV/PDF
- api Rating factors and API11 items per driver, each with its use and status
- calculate Loss-ratio calculatorStart from your own kasko loss ratio and participation
- functions MethodologyEvery index recomputable by your actuaries
arrow_forward Next step: a working session with kasko pricing.
shield_lock KVKK · explicit consent · no location shared
12 / 13
Live product
One data source: driver, fleet, insurer. Click through.
- 01open_in_newspeedMy real, connected carevvisor.com/dashboard/my-tesla
- 02open_in_newlocal_shippingFleet boardevvisor.com/insurer/fleet
- 03open_in_newapiRating factors and APIevvisor.com/insurer/pricing
- 04open_in_newverified_userInsurer portfolioevvisor.com/insurer
Green: real data. Blue: demo pages with sample data, marked on every screen.
calculate Any calculation you need, we run on your data.
bolt Thank you.
shield_lock KVKK · explicit consent · no location shared
13 / 13
