Contract review
Choosing Third-Party Paper Review Software When Most Contracts Arrive on the Other Side's Draft
Some legal teams spend most of their week starting contracts: their own NDA, their own sales agreement, their own vendor form. Others spend it reading drafts somebody else wrote. A procurement-heavy manufacturer, a mid-sized IT services company selling into large enterprises, a startup signing customer MSAs written by the customer's legal team: for these teams, the counterparty's draft is the normal case, not the exception. That difference should drive which software you buy. Much of what is marketed as contract automation helps teams that start documents. If your intake is mostly third-party paper, you need a tool whose strength is reading, comparing and marking up wording you did not write. This page is about choosing that tool for Indian contracts. ## First, count where your contracts come from Before looking at any software, take the last 50 contracts your team handled and sort them into two piles: drafted on our paper, drafted on theirs. Then sort the "theirs" pile by who sent it: large customers, large suppliers, government or PSU tenders, smaller counterparties. If more than half came on the other side's draft, the rest of this page applies to you directly. If most came from large customers and PSUs, note that their drafts are often non-negotiable on many points, which changes what you need: less redlining, more accurate risk marking so the business can decide whether to accept. ## Why template automation does little for this work Template and workflow tools are good at three things: generating a document from a form, routing it for approval, and sending it for signature. On third-party paper, the first is irrelevant, because the document already exists. The second and third still help, but they are not where the hours go. The hours go into reading. A reviewer has to work out what the draft says about liability, indemnity, termination, payment and data, compare each position with what the company will accept, spot what is missing, and write a redline the other side will take seriously. A tool earns its fee on third-party paper only if it shortens that reading and makes it more reliable. ## Capabilities that matter, ranked for counterparty drafts | Rank | Capability | What to test | |---|---|---| | 1 | Marking positions against your interests | Does it say which side a clause favours, not just summarise it? | | 2 | Missing-clause detection | Does it notice that a draft has no limitation of liability or no data protection clause at all? | | 3 | Your playbook, not a generic one | Can you set your positions (cap at fees paid, 30-day payment, Indian seat) and have the tool check against them? | | 4 | Usable redline output | Does it produce tracked edits or a clean list of proposed changes you can send, in Word? | | 5 | Handles real files | Scanned PDFs, documents with heavy tracked changes, long annexures | | 6 | Key dates and obligations | Notice periods, renewal dates, delivery milestones pulled out for the record | | 7 | Record after signature | The reviewed draft, the final version and the decisions kept together | The first three separate review tools from document stores. A tool that summarises clauses neutrally is a reading aid, not a reviewer; you still have to decide whether "Supplier's aggregate liability shall not exceed the fees paid in the preceding three months" is acceptable, and a good tool should already tell you it is below most buyers' positions. ## Indian law points a review tool should catch Generic review engines trained on US and UK paper miss points that matter in India. Ask any vendor to show how its tool handles these on an Indian draft: Non-compete and non-solicit clauses. Section 27 of the Indian Contract Act, 1872 provides that every agreement restraining anyone from exercising a lawful profession, trade or business is void to that extent, with an exception for a seller of goodwill agreeing to reasonable local limits. Courts have generally upheld restraints that operate during the term of a contract while treating post-term restraints as void. A good tool flags a two-year post-term non-compete in a services agreement as an enforceability problem, not just as a commercial term to trade. Arbitration seat versus venue. Section 20 of the Arbitration and Conciliation Act, 1996 lets the parties agree on the place of arbitration, and Indian courts have read that place as the seat, which decides which courts supervise the arbitration. A counterparty draft that names a foreign seat, or names only a "venue" and leaves the seat unclear, deserves a flag. The tool should distinguish the two, because a clause saying "venue: Singapore" with Indian governing law creates exactly the confusion that leads to litigation about where to litigate. Stamp duty. Stamp duty is a state subject and the rate depends on the instrument and the state. An unstamped or under-stamped agreement is not admissible in evidence until duty and penalty are paid. In December 2023 a seven-judge bench of the Supreme Court held that non-stamping of an arbitration agreement is a curable defect that does not make it void, as summarised by Fox Mandal. That softens the worst outcome but does not remove the cost and delay. A review tool should at least prompt for the stamping state and instrument type. For the practical side, see which agreements need stamp paper in India. Data clauses. Counterparty drafts from overseas customers often carry GDPR-shaped data terms with no reference to India's Digital Personal Data Protection Act, 2023. A tool reviewing Indian paper should notice when personal data is processed and the draft is silent on Indian law. Tax and payment mechanics. GST on the invoice, TDS deductions and gross-up clauses are frequent sources of later disputes in services contracts. Flagging a missing gross-up or a "fees are inclusive of all taxes" line in a supplier draft saves an argument at the first invoice. ## Four kinds of help, compared honestly | Option | Strengths | Weak spots | Fits when | |---|---|---|---| | Dedicated AI contract review tools | Fast first pass, playbook checks, redline suggestions | Varying depth on Indian law; some sit apart from where contracts are stored | Volume of inbound drafts is high and positions are stable | | CLM platforms with review built in | Review, storage, dates and signatures in one record | Review quality varies widely between platforms; setup takes longer | You also need a contract repository and renewal tracking | | Outside counsel | Judgment on unusual and high-value deals; accountability | Cost per draft and turnaround for routine paper | Low volume, high stakes, or a deal outside your playbook | | General AI chatbots | Available now, low cost, good at summarising | No stored positions, no record, data terms vary by plan, inconsistent redlines | Occasional low-risk reads, with care about what you paste | Most teams end up combining two: a tool for the first pass on routine inbound paper, and counsel for the deals that justify it. The software decision is about which tool takes that first pass. ## Build a five-draft test pack before you speak to vendors Pick five counterparty drafts your team has already negotiated, so you know the answers: 1. A supplier MSA with a liability cap lower than your position. 2. A customer agreement with a post-term non-compete or non-solicit wider than you accept. 3. A services contract with a foreign seat or a seat-versus-venue muddle. 4. An agreement silent on personal data although data is processed. 5. A long draft with heavy tracked changes or a scanned signature page. Run each candidate tool on all five. Count the issues it catches, the false alarms, and the minutes your reviewer spends turning the output into something you would send. If you want the review discipline that sits behind this, our guide on how to review third-party paper safely walks through the steps a human reviewer follows, which is a good benchmark for what the tool should replicate. ## When the draft cannot be negotiated at all Large enterprise customers, banks and public sector buyers often send paper with a short list of negotiable points and a long list that is fixed. Redlining every unfavourable clause in such a draft wastes the reviewer's time and annoys the counterparty. What the business needs instead is a clear risk note: which fixed terms carry real exposure, how large it could be, and what the company can do outside the contract to manage it, such as insurance, internal controls on the specific obligation, or pricing. A review tool helps here if it can rank issues by severity and pull out obligations, deadlines and monetary terms, so the note writes itself from the output. It helps much less if all it can do is propose redlines. When you test tools, include one non-negotiable draft from a large customer and ask each tool for a severity-ranked list rather than a markup. The difference between tools on that task is often wider than on ordinary redlining, because severity ranking needs a view of what matters to your business, which is exactly what a generic engine lacks. ## Questions to ask each vendor about third-party paper - Can we load our own positions, and does the tool cite which position a flag relies on?
- How does it handle Indian law points such as Section 27, arbitration seat and stamp duty?
- What happens with scanned PDFs and documents with tracked changes?
- Is the output a Word redline, a comment list, or both?
- Does our contract text train any model, yours or a provider's?
- Where is our data stored, and can we export everything if we leave? ## Where Adira fits for teams reviewing the other side's draft Adira is an AI contract lifecycle tool built India-first, which is relevant for the Indian law points above. Opening a counterparty draft in Adira shows positions marked favourable, neutral or unfavourable to your company, red flags, vulnerabilities with severity, missing clauses, key dates, obligations and monetary terms, along with compliance items it checks for, including Companies Act points, Section 42 and PAS-3, SEBI, FEMA and stamp duty. A company persona built from your own documents sets what "favourable" means for you. Changes can be made in plain English at clause level, for example "cap liability at fees paid in the previous twelve months", and Adira returns a diff; documents export to Word and PDF. Adira's own measured result, from the 18 law firms that use it, is contract review and due diligence that took 10 to 12 days manually reduced to under 2 hours, with only 2 to 3 rechecks needed. Treat that as a reason to test it on your five-draft pack during the 7-day free trial, not as a forecast. Limits to weigh: Adira does not replace judgment on unusual deals; it is not a procurement system; and while it is architected for UK, Irish, Singapore and Australian law, its depth is greatest in India, so foreign-law drafts deserve a closer check. Adira states that it does not train on customer contract corpora. ## Buy for reading, then prove it on five drafts If most of your contracts arrive on someone else's draft, buy for reading, not for generating. Rank playbook-based position marking and missing-clause detection above template features, insist on Indian law handling for non-competes, arbitration seat and stamping, and decide on the evidence of five drafts you already know inside out.
Frequently asked questions
- Why do contract template tools not help with third-party paper?
- Template tools speed up documents you start. When a supplier or customer sends its own draft, there is no template to fill; the work is reading someone else's wording against your positions. Only review features, such as playbook comparison, unfavourable-term flags and missing-clause checks, apply to that work.
- Should a review tool flag non-compete clauses in Indian contracts?
- Yes. Section 27 of the Indian Contract Act, 1872 makes agreements in restraint of a lawful profession, trade or business void to that extent, with a statutory exception for the sale of goodwill. A tool reviewing Indian paper should mark post-term non-competes as a likely enforceability issue, not merely as a negotiation point.
- Does an unstamped agreement become void in India?
- In December 2023 a seven-judge bench of the Supreme Court held that failure to stamp an arbitration agreement is a curable defect and does not make it void, while unstamped instruments remain inadmissible in evidence until duty and any penalty are paid. Review software should still prompt you to check stamp duty before signing, because curing it later costs time and money.
- Can we use a general AI chatbot to review counterparty drafts?
- For a quick read of a low-risk draft, some teams do. The limits are that it does not know your positions unless you paste them every time, keeps no record of what was reviewed, may not be on terms that keep your text out of training, and does not produce a structured redline. Check the plan's data terms before pasting any contract.
- How many test drafts do we need to compare review tools fairly?
- Five well-chosen counterparty drafts with known issues are enough to separate tools, as long as each one carries at least one trap your team has already found. Run every tool on the same five and count catches and false alarms.
Sources
See how Adira drafts in your voice and reads contracts from your side.
Explore the showroomWorking through a contract like this? Weave is Adira’s free tool to read, mark up, and connect any contract in your browser — no account needed.
Try Weave — free