Most teams targeting newly registered businesses segment by industry and location, then send the same message to everyone. The variable that predicts response better than either is days since registration — because a new business makes its purchase decisions in a rough sequence, and what it is deciding at day 10 is not what it is deciding at day 120.
This playbook sets out four cohorts, what each is actually working on, and how to sequence against it. The filters that produce these cohorts are covered in what new GST registration leads actually are; the rules governing how you may contact them are in the TRAI and DND guide, and you should read that before running any of this.
Four cohorts
| Cohort | Days since registration | Preoccupation | What converts |
|---|---|---|---|
| Setup | 0–30 | Compliance and banking basics | Speed, specificity, low commitment |
| Systems | 31–90 | Operations and tooling | Demonstrated fit, peer proof |
| Scaling | 91–180 | Growth constraints | ROI framing, case evidence |
| Established | 180+ | Optimisation and switching | Displacement of an incumbent |
Setup: days 0–30
The business has a GSTIN and a list of things it has not yet done. Current account. First invoice format. Whether the CA it used for registration is the CA it keeps. Whether it needs to file this month.
- Sells well: accounting and filing services, current accounts, basic invoicing, seals and stationery, compliance calendars.
- Sells badly: anything requiring a budget process, a demo cycle, or an annual commitment.
- Message: one specific, imminent obligation and an offer to handle it. Not a product tour.
- Reality: this is the most contested cohort in the market. If your lag is three weeks, you arrive as the fourth caller. Measure your lag before you plan a campaign here — the method is in what "daily" and "fresh" mean.
Systems: days 31–90
Manual processes have started to hurt. The spreadsheet is failing. The business is receptive to tools in a way it was not at day 5, when it had no operational history to be annoyed by.
- Sells well: accounting and billing software, inventory tools, payment infrastructure, insurance, logistics accounts, payroll for the first hires.
- Message: a concrete problem they are having this month, and evidence a similar business solved it.
- Why this cohort is underworked: everyone piles into days 0–30 and gives up by day 40. The competitive intensity here is considerably lower and the buyer is better informed.
Scaling: days 91–180
There is trading history. Constraints are visible — working capital, capacity, hiring, distribution.
- Sells well: working capital and business lending, larger software, staffing, marketing services, expanded logistics.
- Message: ROI with numbers, and evidence from a comparable business.
- Note for lenders: this is the earliest cohort where trading history exists at all. Targeting a lending product at days 0–30 is targeting businesses with nothing to underwrite — see GST data for DSA and lending teams.
Established: 180+ days
No longer new. Systems exist and are entrenched. Any sale is now a displacement.
- Sells well: better terms, better pricing, better service against a named incumbent.
- Message: a specific comparison against what they already use.
- Reality: treat these as ordinary competitive prospects. The "new business" angle is spent and using it signals you have not looked.
Segmenting inside a cohort
Two fields do most of the work after registration age:
Constitution of business. A proprietorship decides in one conversation with one person and buys on the call. A private limited company has a process, more stakeholders, and a longer cycle — but a larger deal. These need different sequences, different collateral and different rep seniority. Sending the same email to both wastes the better opportunity.
Nature of business activity. Manufacturing, trading, services and export have genuinely different operational problems. A message about GST input credit on capital purchases means something to a manufacturer and nothing to a consultancy.
Layer geography last, and treat it with the scepticism it deserves — registered addresses are often residences or CA offices, as covered in state-wise GST registration data.
A sequence that respects the rules
For a Systems-cohort prospect, five touches across three weeks:
- Day 0 — Email. Subject names their business activity and one specific problem. Three sentences. One question. No attachment, no deck.
- Day 3 — Call. If you have a number and it is not on a registered preference, and your calling entity is compliant. Purpose: confirm relevance, not to pitch. Ninety seconds.
- Day 7 — Email. One piece of evidence — a comparable business, a short worked example. Still no deck.
- Day 14 — Call. Final attempt.
- Day 21 — Email. Explicit close-out: "I'll stop here unless you'd like me to follow up." This one gets more replies than steps 2 through 4 combined.
Then stop and suppress. Two hard rules:
- Any opt-out is permanent and applies across your whole organisation, not just this campaign. Maintain one suppression list, check it before every send, and honour it at the account level.
- Registered preferences and consent requirements govern calls and SMS, and B2B is not a blanket exemption — a proprietor's mobile is a personal number regardless of what you use it for. Read the TRAI and DND guide before dialling anything.
The registration-age filter this whole playbook depends on is a standard field in GST-based feeds, including FinScreener (built by the team publishing this site — see our disclosure). MCA-derived products such as Tofler and Probe42 offer the equivalent on incorporation date, for the company and LLP subset.
Measuring by cohort
Track these separately for each cohort, or you will average away the finding:
| Metric | What it tells you |
|---|---|
| Contact rate | Data quality, not message quality |
| Response rate by cohort | Whether your timing hypothesis holds |
| Meeting rate by constitution | Whether you are approaching each entity type correctly |
| Cost per meeting | The only figure comparable across channels |
| Opt-out rate | Your early warning on targeting and tone |
Opt-out rate is the one to watch weekly. A rising rate means you are contacting people for whom you are irrelevant, and it damages deliverability and reputation long before it shows up in pipeline.
What not to do
- Do not treat registration as intent. It is a timing signal. Nothing more.
- Do not buy volume you cannot work. 500 records worked properly beat 50,000 blasted, and the second approach damages your sending reputation for months.
- Do not use registration data as a credit signal. It says nothing about revenue or ability to repay.
- Do not re-contact across campaigns without checking suppression. The complaint that ends a channel usually comes from someone contacted twice after opting out.
- Do not claim official endorsement. Nothing about holding public registration data associates you with a government body, and implying otherwise is both false and a fast route to a complaint.
Common questions
Which cohort has the best conversion? It depends entirely on what you sell. Compliance and banking peak in the Setup cohort; software in Systems; lending in Scaling. Match your product to the decision the business is actually making.
How long should I keep working a record? Through one sequence, then suppress and revisit at a natural trigger — a new cohort boundary, a status change, a new location. Continuous contact of a non-responder produces complaints, not sales.
Is email or phone better here? Email scales and carries lower regulatory friction. Phone converts better when the number is right and the call is short. Most teams over-invest in dialling bad numbers — fix contact quality first with the data quality checklist.
Can I message on WhatsApp? Business messaging platforms have their own consent and template requirements layered on top of telecom rules. Do not treat it as an unregulated channel because it feels informal.