Every managing partner asks the same question within ten minutes of a software conversation: should we build this ourselves, or buy something? The honest answer is that it depends on five things, none of which are price. Anyone who answers without asking about those five things is selling you something — and that includes us, on the days we forget.
This post is the framework we use, written for legal leaders weighing matter management software in 2026. It is opinionated. It assumes you already know the symptoms — files in shared drives, conflicts checked twice or not at all, billing realisation no one can explain. The question now is what to do about it.
The five questions that actually decide it
Forget total cost of ownership spreadsheets for a moment. Run your firm through these questions first.
- Do your processes match the defaults of a packaged product? Not "can they be made to match" — do they already match, today, without partner re-education. If yes, buy. If you'd be running training sessions to teach partners how the product wants them to work, you're building anyway. You're just paying someone else to make the choices.
- How specific is your partnership deed? Originating-partner cuts, supervisory-partner overrides, practice-area pools, write-off authority — every firm's deed reads slightly differently. Packaged products encode one or two allocation models. If yours doesn't fit, you'll be exporting to Excel forever.
- Are you single-practice or multi-practice? Litigation, corporate, IP, tax — each has different matter shapes. Packaged products are usually strong in one and weak in two. Multi-practice firms feel this within a quarter.
- What's your appetite for owning the system long-term? Custom software requires a retainer relationship with whoever builds it. SaaS requires a renewal cycle and a vendor relationship. Both cost money; they cost different kinds of attention.
- Where are your conflict and confidentiality risks? A litigation-heavy firm with frequent group-company conflicts has different needs than an IP boutique. The bar council and your largest client both have opinions. If those opinions don't match a foreign-built product's defaults, you're customising regardless.
If four of five answers point to "we look like everyone else," buy. If three or more answers point to "we don't," seriously consider building — or, more often, building on top of something bought.
The case for buy
We say this honestly because we lose work by saying it: most small Indian law firms should buy.
A packaged product will be running this afternoon. It will have a billing module that's been pressure-tested by hundreds of firms. It will have integrations to email, calendar, and (sometimes) e-filing portals. It will have a support team, a roadmap, and a price you can predict. None of that is trivial. None of that is what a custom build gives you in week one.
If you are under ten fee-earners, single-practice, and your partners can describe their workflow in one paragraph that matches the product's marketing page, you are the customer that packaged matter management software was built for. Buy. Use the eighteen months and the eight to twelve lakhs you didn't spend on a custom build to actually grow the firm.
The Xakia and LawVu and Clio products do real work. They are good. The honest test of whether they fit you isn't a feature checklist — it's whether two of your partners can shadow your senior associate using the trial version for a week without filing twenty workaround tickets. If they can, you have your answer.
The case for build
The case for build is narrower than vendors admit and broader than packaged-product blogs admit. It usually applies when one of these is true:
- Your conflict logic is the firm's competitive moat. Some Indian firms have spent twenty years refining how they think about ultimate beneficial owners across group companies. Encoding that in software is real intellectual property. A packaged conflict checker will dilute it.
- Your partner allocation rules are unusual. Originating-partner-plus-working-partner-plus-practice-pool with a supervisor override is not unusual to Indian firms; it is unusual to the products being sold to you. Forcing your deed into a foreign data model is how partners stop trusting the system.
- You have ten or more fee-earners and grow ten percent a year. Per-seat SaaS pricing scales linearly; a custom build amortises. The break-even on a six-week core build is usually month eighteen at fifteen seats.
- You have integrations no packaged product covers. State-specific e-stamping APIs, district court e-filing portals, your specific GST consultant's import format. Each is a small piece of work; they add up.
- Year two matters more than year one. Custom software in year two is a managed retainer that adapts to how the firm has changed. Packaged software in year two is a renewal email and a dashboard that hasn't moved since the demo.
Note what's not on the list: "we want to be modern" or "our consultant said we should." Build for the reasons above, or don't build.
We've written separately about what actually works in custom software for Indian law firms, including the four things any matter management system has to do before partners trust it.
The hybrid most firms end up with
Here is the option packaged-product comparisons rarely surface and that custom-software studios rarely volunteer: most mid-sized Indian firms are best served by a hybrid.
Buy the boring core. Build the partnership-specific surface.
Concretely, that often looks like this. A packaged document management system or billing engine — boring, reliable, probably foreign — runs underneath. On top of it, a custom matter-management surface holds your conflict logic, your partner allocation rules, your practice-area-specific matter shapes, and the partner reporting that the managing committee actually argues about. The two talk to each other through APIs you own.
Why this works: you get the long-tail reliability of a packaged product where reliability is what you need, and the firm-specific shape where shape is what makes you different. You also get a much smaller custom build — eight to ten weeks instead of six months — because you're not rebuilding billing infrastructure that already exists.
The trap to avoid: a hybrid that nobody owns end-to-end. If your packaged vendor blames the custom layer and the custom builder blames the packaged vendor, you've built a finger-pointing machine. The cure is one accountable partner — usually the custom builder — who owns the integration and shows up when something breaks.
When firms ask us to scope a custom matter-management surface or a partner-reporting layer, this is the shape we usually arrive at after the first discovery call.
What "build" actually costs in 2026
This is the part most legal-tech blogs are vague about, so:
A first-version custom matter-management build covering matter records, conflict logic, a document vault, time entry, and partner-allocation reporting runs in the range a fixed-scope studio will quote in a week. Six to ten weeks of build, fixed scope, fixed price. After that, a monthly retainer for what's next — e-filing, the second-office rollout, the partner dashboard your managing committee will inevitably ask for.
A packaged product runs roughly fifty to a hundred and fifty US dollars per fee-earner per month, plus implementation. At fifteen fee-earners over five years, that's a meaningful number — comparable to a custom build plus retainer, sometimes higher, sometimes lower, depending on seat growth.
The price difference is rarely the deciding factor. The deciding factor is who owns the system in year three when your firm looks different than it does today. With a packaged product, the vendor owns it and you adapt. With a custom build on a real retainer, you own it and the system adapts. Both are legitimate choices. They are different choices.
We've written separately about what an honest software retainer looks like — the year-two question matters more for legal tech than for any other vertical we work in.
A decision flow
If you want a single decision tree, here it is:
- Under ten fee-earners and single-practice: buy. Pick a product that has been used by a firm shaped like yours and trial it for a month with your busiest associate, not your partners.
- Ten to thirty fee-earners, multi-practice, partnership-deed-specific: hybrid. Buy a billing or DMS core, build the matter-management and partner-reporting surface on top.
- Thirty-plus fee-earners, multi-office, with deep workflow IP: full custom is usually defensible, especially if you already have a tech lead in-house who can act as the managing-committee's translator to your build partner.
Whatever you decide, the rule that matters more than build vs buy: ship a thin slice into production within ten weeks. Most legal-tech projects fail not because the wrong commercial model was chosen, but because no version of the system was used in real client work for the first six months. Software that lives in a UAT environment doesn't survive the first partner who pushes back. Software that's already booking real time entries and producing real bills survives because the alternative is going back to the spreadsheet.
For more on the failure modes that kill legal-tech builds — Big Bang launches, software that treats partners as users, skipping the boring numbers — we've covered them in the law firm field guide. For a working example of what a custom matter-management surface looks like in production, Halverson Chambers is the build we point to most often.
What we do
Tanvora Labs builds custom matter management software, document vaults, partner reporting layers, and the integrations that make a packaged core fit an Indian law firm's actual deed. Six-week core builds, fixed scope, fixed price, monthly retainer for what's next.
If you're at the stage of arguing build vs buy with your managing committee, book a thirty-minute discovery call. We'll listen for the bottleneck and tell you honestly which side of the line your firm sits on. The Bar Council of India publishes guidance on confidentiality and electronic record-keeping that's worth reading before any vendor conversation — it tends to clarify which compromises are acceptable and which aren't.
Sometimes the answer is buy, and we say so. Sometimes it's build. More often than either, it's a smaller custom layer over a boring core, and the partnership keeps the parts of itself that made the firm worth building software for in the first place.