KoatPro
AI that measures, calculates, and recommends, with every number checkable by the contractor
Category
AI Product Design
Client
EnKoat
Duration
6 Weeks
Tool
Figma, Figma Make
Project overview
An estimating app built around one site visit
KoatPro is a tablet app for contractors who apply EnKoat's IntelliKoat™ roof coatings. A contractor enters an address, reviews an AI-measured roof, calculates coating quantities and costs, and presents three priced options to the property owner on-site.
Outcomes
One visit instead of several days
The outcome KoatPro is designed for is that a contractor leaves the roof with a priced, presented quote in one visit instead of several days. In testing, contractors reported a 40% reduction in task time.
40% less task time
building a quote took 40% less time than their current process
One place for every estimate
roof data, condition photos, materials, and quotes are available in one place, at any time
Answers on demand
the AI assistant quickly resolves questions about products and pricing and pulls up estimate data when needed
Quotes that reflect damage
the satellite roof view, combined with photos of any damage, produces a quote that accounts for everything found on-site
Impact on contractors: the quote, before and after
Problem
Coating quotes go cold before they're delivered
A coating quote that leaves the roof with the contractor loses momentum. Firms that respond to a lead within an hour are nearly seven times as likely to qualify it as those that wait even an hour longer (Harvard Business Review, 2011). Four problems shaped the brief:
Slow turnaround
measuring, math, and write-up happen separately, so owners wait days
Error-prone math:
one wrong rate or rounding step changes gallons, cost, and margin
Low trust
satellite data can be outdated, and contractors won't sign a number they can't check
Hard to explain
contractors pitch a technical coating from spec sheets and old quotes
My role
Concept to working prototype, end to end
I owned KoatPro end-to-end: the problem framing, scope decisions, information architecture, interaction design, design system, and the working prototype. I built the prototype in Figma Make, directing it screen by screen through written specifications, then reviewing each build for layout, typography, and calculation accuracy.
I also created the KoatPro logo and grounded the materials logic in EnKoat's IntelliKoat™ Technical Manual v5.0.
Research
Every tool stops partway
I combined product documentation, a competitor analysis, and secondary research on speed and AI trust, then layered in what I see daily on EnKoat's product side.
Product documentation
EnKoat's IntelliKoat™ Technical Manual v5.0 defined the coating system, coverage rates, and warranty terms; only some configurations had published rates.
Competitor
analysis
I compared five measuring and quoting tools to see where coating contractors still fall back on spreadsheets.
Secondary
research.
Lead response-time research, plus human-AI interaction guidelines from Microsoft Research and Google PAIR, applied screen by screen in sections 9 to 11.
Domain
context
Leading product and brand design at EnKoat gives me daily exposure to how the coating is explained, specified, and sold.
Competitor analysis
Key insights
Speed wins the job, but only if the numbers hold
INSIGHT 1
Speed decides the sale
INSIGHT 2
Rates are complex and incomplete
INSIGHT 3
AI is only useful if it's checkable
INSIGHT 4
No tool covers coatings end to end
Persona
Built for the contractor on the roof
Marcus Rivera runs Rivera Roofing Co., a commercial roofing contractor working on warehouses, schools, offices, and data centers across several states.
Ideation
Seven ideas, narrowed to one moment
I explored seven directions and judged each against three questions: does it reach the people who sell the coating, does AI genuinely help, and can it run on data EnKoat already has?
Three questions framed the design:
How might a contractor get from an address to a quote in minutes?
How might AI measurements be easy to check and correct?
How might an owner compare three coating options at a glance?
Three principles guided every screen: Speed (everything on-site, in one flow), Verifiability (every AI value labeled and editable), and Consistency (one set of inputs drives every number).
The estimate flow
From roof to signed-off quote in one visit
Five steps and a confirmation take the contractor from the roof report to the owner's decision, all on one tablet, without leaving the property.
STEP 1
STEP 2
STEP 3
STEP 4
STEP 5
STEP 6
Around the flow
Nothing slips between visits
These four screens keep a contractor's estimates moving: where to pick up, what's waiting on owners, and the defaults every new quote starts from.
Home page
Estimates page
Estimate detail page
Profile page
AI in KoatPro
The AI suggests, the contractor decides
Every AI output has its own purple styling, shows its confidence or its source, and can be corrected, following guidance that AI should be easy to correct and explain itself (Microsoft HAX) and help users calibrate trust (Google PAIR).
IMAGE 1
IMAGE 2
IMAGE 3
AI inside the estimate flow
Roof Report: AI measures the roof; the contractor verifies.
Condition Check: AI suggests a tier and explains why; the contractor decides.
Materials: AI supplies one input; the pricing math stays rule-based.
Quote Builder: AI flags a recommendation; all three options stay on the table.
High-fidelity design
Built to read on a sunny rooftop
The visual system is light and calm, so data stays readable outdoors:
Neutral surfaces
keep the focus on roofs, numbers, and options
One brand blue
marks every action, so the next step is always obvious
A separate purple for AI
A separate purple for AI
Inter with tabular figures
lines up gallons, mils, and prices for easy comparison
An 8pt spacing grid
keeps dense screens consistent and scannable
44px minimum touch targets
work for gloved hands on a rooftop
Future scope
The biggest opportunity is AI damage detection, flagging likely cracks, blistering, or ponding in condition photos, labeled and editable like every other AI value, since contractors already value photos feeding the quote. Beyond that, a phone companion would support quick capture on the roof, offline mode would keep quoting possible where signal is weak, and a shareable proposal would let owners review and sign later. Over time, comparing gallons ordered with gallons used would tighten estimates and inform pricing, and an office dashboard would help teams manage estimates across crews and catch quotes before they go cold.
