Settl: Pricing a Bed in Seconds

Overview

Settl is an Indian co-living company with more than 3,400 apartments. Its sales agents priced each bed by hand, so prices varied from one property to the next, and agents had trouble balancing a competitive rate against profit targets. As an analyst on a 180 Degrees Consulting team, I helped build a full-stack calculator that gives agents ready-made price options. It cut manual calculation time by 50% for more than 30 agents.

An agent picks a property or micromarket, filters rooms, and enters a move-in date and tenure. The Flask backend looks up the property’s gross margin, room category and tenure from cost data in SQL, and returns 2 or 3 price options tied to profit levels: 15% and 10% for properties under a 40% gross margin, plus a 5% option above it. Preferred rooms (balcony, attached washroom, high floor) add 2 points to every tier.

Property margin under 40%

15%10%none

Option 1Option 2Option 3

Property margin over 40%

15%10%5%

Option 1Option 2Option 3

Preferred room? +2 points on every tier.

A tradeoff I made, and why

Fixed tiers instead of one “optimal” price. The tool could have calculated a single best price per bed. Instead it returns 2 or 3 options tied to set profit levels. That gives up some precision, but agents keep room to negotiate, and even the lowest option is still profitable. Under a 40% margin, the 5% option disappears, so a thin-margin property can’t be discounted below its floor.

Rules a sales lead can read. Every price comes from a short rule: margin band, room type, tenure, plus 2 points for a preferred room. An agent or manager can check a number by hand, which matters more for adoption than a cleverer model nobody trusts.

What I learned

• Usability checks before code save rework. Reviewing the Figma screens against Nielsen’s heuristics caught problems while they were still cheap to fix.
• A simple rule people trust beats a smart one they don’t. Sales agents adopt a tool they can explain to a customer.