Bus wait, bunching and bus-lane audits built from the fleet's own GPS data
TL;DR Busgapindex takes the GPS and ticketing data a city bus operator already collects and turns it into a monthly report on what riders actually live with - how long they wait, where buses bunch and which stretches need a bus lane or a moved stop. State transport undertakings and city SPVs pay per route network.
Bus riders are the people a city's traffic plans tend to forget. In Mumbai, 77% of public transport users surveyed lived in informal settlements or slum-rehabilitation housing[1]. In Kochi, two-thirds of bus users had no car, and what they cared about most was frequency and punctuality, which most routes studied failed on[2].
In some places it's getting worse. In Pune and Pimpri-Chinchwad, the share of people with frequent bus service within a five-minute walk fell from 66% in 2019 to 21% in 2026[3]. Operators report kilometres run and buses on road. Hardly any report the rider's wait, and that's the number that decides whether people stay on the bus.
Busgapindex plugs into the bus operator's tracking and ticketing feeds, and no new hardware goes on buses or roads. From each bus's position every few seconds, it rebuilds what happened at every stop - the gap between buses, how often two showed up together and how long a rider arriving at a random moment actually waited.
Every month the client gets a route-by-route scorecard and a short list of fixes. Some are operational, like holding a bus that runs early to even out gaps, which is the standard remedy for bunching[4]. Others are physical. Say a 2 km stretch where buses lose ten minutes in the evening peak, ranked as a case for a bus lane, signal priority or a moved stop. A re-audit then shows whether the change worked.
The original version of this idea was AI that optimises traffic flow to make mobility fairer. Optimising car flow helps car users first, and one big tech firm already gives cities free signal timing advice aimed at fewer vehicle stops[5]. The fairness gap is on the bus, and the data to measure it already exists.
The users are planning and operations teams at state transport undertakings and city bus companies. City SPVs and unified metropolitan transport authorities are a second buyer, because they decide where bus lanes and stops go. I found no public Indian price for bus reliability analytics, so an annual fee per route network with a paid pilot first is my assumption to test.
Ticketing is going digital. Bengaluru's BMTC started accepting the National Common Mobility Card in September 2026, and Chennai and Delhi buses have gone the same way[6]. Tap data adds boardings to the GPS trail. Delhi's transport department already publishes static and real-time bus data with IIIT-Delhi[7], which proves the feeds can be opened up.
Nobody publishes a figure for Indian spending on bus performance analytics. Abroad, one transit-software firm says more than 200 transit agencies in 10 countries use its platform, which includes spotting bunching early[8]. As an adjacent number, India's bus market was put at about US$20 billion in 2021, roughly ₹1.9 lakh Cr[9]. That's operator revenue though, not your market.
Transit-software firms sell real-time operations and on-time-performance tools to US and European agencies[8]. In India, one bus-ticketing app describes its work as digitising bus commutes[9], and Delhi's open data is used mostly by app builders and researchers[7]. Big tech's signal programme tunes signals for vehicles[5]. None of them leads with the rider's wait as an audited public number.
| Difficulty | Moderate (40/100) |
| Starting capital | ₹30–60 L |
| Time to launch | 4–6 months |
The analysis is well understood. The hard part is getting raw GPS feeds out of vendor systems that operators don't fully control, cleaning up patchy data and then selling to public bodies that buy through tenders, change officers often and pay late.
No special licence is needed to sell analytics to a bus operator. The baseline plus a few contracts covers it.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
Don't trust AI-filled data as ground truth. Where the GPS is patchy, say so in the report, and check measured waits against manual stop counts.
Want to work out the AI layer for Busgapindex? Talk to Yash →
Here's what can go south -
I think this is a good fit for a founder with transit-planning or data-engineering experience who can work patiently with public operators. If you want a quick SaaS sale then this isn't it. Prove one thing first - on 20 routes in one city, your measured waits match manual stop counts within 15%.
Plan for roughly ₹30–60 L, mostly for data engineers, a paid pilot and the long public sales cycle.
No sector licence. You'll need data-sharing agreements with operators, DPDP compliance for staff IDs in the feeds and usually a tender or GeM route.
Yes, any city with GPS on its buses has the same data. Check local open data terms and data protection rules for staff records.
The free Busgapindex validation kit has the line-by-line capital breakdown, the effort scoring, the data-sharing and procurement checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for transport operators and city agencies, and 5 assumptions to test. Download the kit →
Thinking about building Busgapindex or something close to it? The make-or-break bit is getting raw bus data out of an operator and turning one finding into a change on the road, and that's the kind of thing I help founders with. If you need a hand, you know where to find me. Talk to Yash →
For illustration only. Names are fictional and ideas hypothetical; sourced facts are real but may be out of date. Nothing here is legal, financial or professional advice, and Yash Khatri is not personally liable for any use of it. Read the full disclaimer.