One property-tax register and notice desk for chains with sites in many cities
TL;DR Levyline tracks property tax for businesses with dozens or hundreds of sites, like retail chains, bank branches, warehouses and clinics. It keeps one register of every property's assessment, dues, rebates and notices, files returns on time, and prepares replies to show-cause notices. Companies pay per site per year, with a share of any recovered excess.
Property tax is the main tax Indian cities collect, at an average of 59.1% of municipal corporations' tax revenue between 2020-21 and 2022-23[1]. And each city does it differently. In an RBI survey, 45% of corporations used annual rental value and 21% used capital value, and a quarter relied on civil courts to settle disputes[1].
For a company with 300 sites, that means 300 deadlines, rebate windows and notice formats. Bengaluru alone runs self-assessment returns, revised returns for past years, show-cause notices with online objections, a rebate window that closes in April and penalties from late November[2].
Store managers forward notices late, rebates lapse and wrong classifications go unchallenged.
Levyline builds a register of every site the client owns or leases with a tax obligation - the municipal ID, the assessment method, the classification used, the last demand and payment, and any open notice. It pulls dues from city portals where it can and reads paper notices where it can't. A calendar flags rebate and penalty dates per city.
When a notice or revised demand arrives, Levyline checks it against the property's recorded area, use and zone, and drafts a reply or objection for the client's approval. If the case reaches a tribunal or court then it hands the file to a local lawyer.
The seed proposed AI that finds overpayment and files automated appeals. The RBI report shows weak coverage and old valuations in many cities[1] though, so Indian property tax is more often under-collected than inflated, and blanket appeals would fail. The money for a multi-site business sits in missed rebates, penalties, duplicate demands and misclassification, and a register plus a notice desk catches those.
The finance or real-estate head of a multi-site business pays. My assumed pricing is ₹3,000–8,000 per site a year, plus 15–25% of any refund or demand reduction Levyline secures. A chain with 200 sites would pay about ₹6–16 lakh a year (our estimate). Landlords with many leased-out properties are a second buyer.
Cities are moving filings and objections online. Bengaluru's portal now takes self-assessment, revised returns and objections to show-cause notices through its court case system[2]. Online data makes a central register cheaper to keep, and cities short of revenue have every reason to chase old dues[4]. The RBI also found that 37.7% of surveyed cities ran one-time settlement schemes in the past ten years[1], and each one is a deadline a multi-site firm can miss.
There's no public figure for what Indian businesses spend on managing property tax. The narrowest firm signal is from the RBI survey - methods vary across cities, coverage is between 40% and 80% in most of them, and only 9.4% exceed 80%[1]. In the US, one appeals firm says it has saved customers over $400M (about ₹3,869 Cr) since 2020[3]. That's an adjacent market with different rules.
Most multi-site firms use a local CA or the store manager in each city, which leaves no central view at all. Large property consultancies handle tax for some big landlords as part of wider services. One US appeals firm shows a pay-if-you-save model works for appeals[3], but its premise of overassessment doesn't translate directly.
The gap is a single register and notice service across Indian cities.
| Difficulty | Moderate (32/100) |
| Starting capital | ₹30–60 L |
| Time to launch | 3–5 months |
The software is simple. The work is in local knowledge, because each city's portal, notice format and objection process differs and many still need a visit. The first client's site list will be incomplete too, so onboarding means some detective work across leases and old receipts.
There's no licence for running a property tax register and notice desk. The limit is who can argue a case.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
Don't let AI decide whether to contest a demand or what to pay. Those are money and legal decisions, and the client and its lawyer make them.
Want to work out the AI layer for Levyline? Talk to Yash →
I like this for a founder with a CA or municipal-finance background who genuinely enjoys process work. If you're chasing an AI appeals engine then skip it. Prove first that a 100-site client's last two years show missed rebates, penalties or wrong demands worth more than your annual fee.
Plan for about ₹30–60 L, mostly for a small team of tax staff and the register software.
No sector licence. You can run the register and draft objections, but court and tribunal cases need an enrolled advocate.
Yes. US appeals firms do a version of this. Local tax law and appeal rules differ a lot, so learn them before you sign clients.
The free Levyline validation kit has the line-by-line capital breakdown, the effort scoring, the municipal and legal-practice checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for finance and real-estate heads, and 5 assumptions to test. Download the kit →
Thinking about building Levyline or something close to it? The make-or-break work is showing a first client recoverable money in its own records, and knowing where to dig for it. 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.