Company diligence

Medida AI

Built from public sources only. Chips such as E05 link to the underlying evidence record. Company marketing statements are not independent verification, and no management contact has occurred.

Business description

Medida AI offers smartphone-based measurement for renovation installers, initially windows and doors, with a Tel Aviv / New York footprint.E06Screen PDF §01
The company is expanding product scope beyond doors and windows, according to its 2025 reporting.E04 E05Screen PDF §02

Founders

Founded by Jonathan (Yonatan) Gilat and Benzi Ronen.Screen PDF §01 (founding geography row; underlying source record not itemised in the screen)
Founder geography is a strong thematic fit with the published mandate of global B2B technology businesses with European and MENA founder backgrounds.E01Screen PDF §01
No discussions with Medida management were conducted for this research.Screen PDF Research integrity statement

Product & customer workflows

Problem: manual site measurements can delay a renovation sale or installation and create rework if a door or window does not fit.E04 E05Screen PDF §02
Proposed solution: replace selected site visits with remotely generated measurements and 3D visualizations.E04 E05Screen PDF §02
Benefits to be proven: fewer field visits, shorter cycle time, less installation rework, potentially better conversion.Screen PDF §02

Commercial thesis

Mandate alignment tests

Mandate alignment tests
Test	Public evidence	Preliminary view
Founding geography	Founded by Jonathan (Yonatan) Gilat and Benzi Ronen; Tel Aviv / New York footprint.	Strong thematic fit
Revenue threshold	Revenue reported at an annual rate in Dec 2025; unverified recurring share.E05 E01	Potentially at threshold; not verified ARR
Product	Smartphone-based measurement for renovation installers, initially windows and doors.	Direct vertical workflow benefit
Commercial demand	Website says trusted by 50+ remodeling companies.E06	Promising claim; paid adoption unknown
Unit economics	No verifiable CAC, churn, margins, pricing or customer cohorts.	Critical unresolved
Entry / ownership	No current round terms, security structure or cap table.	Not currently actionable

Moat hypothesis

Medida may win through installation-grade precision, manufacturer-specific datasets, channel relationships and workflow integrations, not simply a generic computer-vision model. These are hypotheses requiring validated win-loss data, customer interviews and product testing.

Competitive landscape

Published positioning; plans are not an apples-to-apples benchmark

Direct and adjacent substitutes
Provider	Published positioning	Differentiation / risk
HOVERE10	Phone-based interior/exterior measurement, 3D modeling and contractor workflows; advertised subscriptions plus per-project pricing.	Material platform and pricing competitor, increasingly end-to-end.
magicplanE11	Floor plans, field measurements, reporting, collaboration and project-priced plans.	Adjacent contractor workflow with broad functionality.
Canvas / Twindo	LiDAR capture and professional-grade CAD/BIM model conversion; paid as-built output.	Overlap in measurement; distinct device and end-output needs.

Gate status: PARTIAL PUBLIC Win-loss data is required. Canvas / Twindo has no discrete evidence record in the seed data.

Revenue quality

Taxonomy required before any SaaS multiple

Revenue taxonomy
Revenue stream	Diligence issue	Valuation implication
Recurring subscription	Paid seat, branch, platform or enterprise license; contracted renewal rights.	Potentially merits a software-led multiple if retention and margin are verified.
Metered measurement / transaction	Volume-dependent paid scans or completed projects.	Model cohort frequency, contraction and variable unit costs.
Implementation / services	Hardware, onboarding, model correction, bespoke integration.	Lower gross margin and revenue quality; avoid ARR treatment.
Channel / partner fees	Installer, manufacturer or distributor revenue sharing.	Check customer concentration and take-rate sustainability.

Company / public statements versus what must be proven

Claims versus what must be proven
Statement	Confidence	What we must prove
$3m revenue rate in Dec 2025E05	Reported by press, no audited accounts	Is this contracted ARR, annualized scan volume, implementation fees or a blend?
More than 50 remodeling-company logosE06	Company website, self-reported	How many are paying now, at which ACV, with what retention and product depth?
99.9% accuracy, up to 1/8 inchE07	Company product claim, not independently established	Blind installation-grade tests by product, hardware, lighting and geometry.
33%+ fewer cancellationsE08	Company product claim, no causal evidence found	Controlled pilot, definition of cancellation, baseline, confidence interval.

Harvard JCHS forecast (Oct 22, 2026) that U.S. owner-occupied home improvement and repair spending stays broadly flat through late 2027. This is context for customers' spending exposure, not Medida's software TAM. No credible bottom-up count, account segmentation or validated pricing data is yet available.E09

Critical test

If the product depends on substantial manual human review, or usage collapses when remodeling activity cools, it should be underwritten with lower gross margins and potentially a lower exit multiple than a high-retention SaaS product.

Customer economics

No verifiable customer-level economics exist in the research package

LTV/CAC
Not available
CAC and payback
Not available
Net revenue retention (NRR)
Not available
Gross revenue retention (GRR)
Not available
Average contract value
Not available
Paying customer count
Not available
Customer concentration
Not available
Pricing
Not available

Proposed customer diligence sample

10 interviews: four active high-usage customers, two newly onboarded, two churned / downgraded, and two qualified prospects that selected a competitor.
Reconcile site-visit savings by measuring average trip cost, repeat visits, installation error rate and time to installation for sampled work orders.
Request monthly revenue by customer and product type, scan volumes, gross margin per scan, support interventions and customer-specific claims / refunds.

Technology & product accuracy

Company states 99.9% measurement accuracy up to 1/8 inch. This is a marketing claim: no independent blind test or sample size is available.E07
Company states 33%+ fewer cancellations. Definition, causality and sample are unknown.E08
Required evidence: independent blind validation across hundreds of jobs, device types, materials, lighting and field error, plus installation callback outcomes.E07Screen PDF §05
Reject if physical error / liability makes the product non-scalable.Screen PDF §05

Gate status: COMPANY CLAIM ONLY

Management diligence requirements

Requests, not findings

No management contact has occurred; all items below are requests, not findings.
Last 18 months of actuals by recurring, metered and services revenue, with contracts and bank / invoice reconciliation.
Latest 12 months actual revenue, margin and gross cash burn, reconciled to bank.
Monthly revenue by customer and product type; signed paid contracts; pilots vs paying users; logo permissions.
Cloud inference cost, human QA, scanning support and warranty liability.
Fully diluted capitalization, SAFEs, preferences, option pool, debt and investor consents; proposed security and live terms.
CAC, payback, channel partners, and split of outbound, referral and OEM.
Patent position, training-data consent, warranty / indemnity and security review.

Investment risks

Each risk maps to a diligence gate

Investment risks and gates
Risk	Detail	Gate / status
Revenue misclassification	Reported '$3m' may blend subscription, per-scan, services and implementation revenue; ARR cannot be claimed.	P0 Revenue classification UNVERIFIED
Stale entry basis	Underwriting on a Dec-2025 run-rate without 2026 trading.	P0 2026 current trading UNVERIFIED
Measurement accuracy and liability	Fundamental product risk and potential rework liability if accuracy claims do not hold.	P0 Measurement accuracy COMPANY CLAIM ONLY
Overstated demand	'50+' may overstate monetized demand: logos may be pilots, trials or historical.	P0 Customer / logo quality COMPANY CLAIM ONLY
Margin quality	No SaaS multiple without software-grade margin; human review or services would lower it.	P0 Gross margin and unit costs UNVERIFIED
Commoditization	High risk of commoditization against HOVER, magicplan and LiDAR/CAD players.	P1 Competitive durability PARTIAL PUBLIC
Remodeling cyclicality	Flat remodeling spend to late 2027; the forecast must stress cyclicality.E09	P2 2027 demand risk MACRO HEADWIND
Price and availability	No evidence that any specific entry price is available; the illustrative price cannot be used as a term sheet.	P1 Cap table and financing UNVERIFIED