The promise, in one sentence, with every word doing work
Here it is: we freeze your trailing-90-day count of qualified leads from organic search on day one, and if we haven't beaten that number within 90 days, we keep working free until we do.
Most agency guarantees fall apart under a careful reading, so it's fair to put ours through the same test. Every phrase in that sentence is load-bearing.
- Trailing 90 days — your own recent past, not an industry benchmark and not a projection. The only number nobody can argue was invented to be easy.
- Qualified leads — defined by you, in your CRM, in writing, before we start. A real enquiry from someone who could plausibly buy: not a spam form fill, not a job application, not a supplier pitch.
- From organic search — the CRM's own source field, with paid, social, direct and referral excluded. One field, one system, one owner.
- Within 90 days — long enough for content and technical work to land, short enough that neither side can hide, and short enough that we can fund the downside.
- Keep working free — we don't refund and disappear. A refund gives you money back and leaves you exactly where you started, three months older. We'd rather owe you the outcome than the invoice.
How the baseline gets frozen, step by step
Every argument about a guarantee is really an argument about the baseline, which is why we do this before any billable work and why it takes a week rather than an afternoon. Six steps, all boring, all the reason nobody has to litigate anything in month four.
- Agree the source of truth. Your CRM, named. Not GA4, which counts sessions rather than people, and not a spreadsheet somebody maintains by hand.
- Write the definition of a qualified lead — one paragraph, with examples of what counts and what doesn't. If your sales team already disqualifies with reason codes, we use those rather than invent new ones.
- Fix the channel rule. How a lead gets tagged as organic: UTM conventions, hidden form fields, first or last touch. Whatever you already do, we adopt — changing it mid-measurement is how numbers become deniable.
- Pull the trailing 90 days and check it for distortion. A one-off PR spike, a festive quarter, a fortnight when the form was broken. We flag it, both sides see the raw data, and we agree in writing whether to adjust or accept it.
- Write the exclusions. Existing customers re-enquiring, brand demand from an offline campaign you're about to run, internal test submissions. If a TV spend lands in month two, we say so now rather than claim the lift later.
- Sign it, with the number on the page. The baseline goes into the contract as a figure with a date. If it isn't a number on a page before work starts, it isn't a baseline — it's a story told later by whoever is winning. Setting an SEO baseline walks through the same process if you want to run it with your current agency.
The four costs on our side of the ledger
Founders assume the cost of a guarantee is the refunds. It isn't, mostly. The expensive parts are structural, and they show up whether or not we ever miss.
- The capacity cap isn't scarcity marketing, it's arithmetic. Forty accounts each carrying a possible free quarter is a liability no boutique balance sheet survives — which is exactly why agencies with forty accounts don't offer this.
- Turning down a client who can't win is cheaper than winning them and then owing three free months while both sides get quietly angry.
| The cost | What it is | What it does to the business |
|---|---|---|
| Capacity | Three new clients a month, maximum | The biggest single cost. Every account carries a contingent free quarter, so growth is capped by risk we can hold — not by how many leads we get. |
| Sales friction | A long qualification call that often ends in no | We have to talk people out of signing when the guarantee couldn't be honoured. Expensive habit for a sales function, cheap one for a reputation. |
| The reserve | Cash set aside to fund a quarter of delivery with no fee against it | Working capital that can't go into hiring or tools. It sits there being useful only if we're wrong. |
| No escape hatch | We can't quietly cut hours on a struggling account | The month a normal agency reduces effort is the month we have to increase it. The incentive runs the right way for you and the wrong way for our margin. |
The qualification filter it forces on us
Because we carry the downside, we ask questions most agencies can skip. These aren't questions to find rich clients — they're questions to find winnable ones. If you're evaluating any guarantee-led agency, notice whether they ask you these. If they don't, they haven't priced their own promise.
- Where do your leads land, and can you export the last 12 months? No exportable history means no baseline, and no baseline means the guarantee is decoration.
- Who decides a lead is qualified? A name, not a department. Somebody has to arbitrate edge cases monthly without it becoming a negotiation.
- How many dev hours a month can you give us, and who approves a deploy? This predicts outcomes better than any audit. A great plan behind a six-week release train produces nothing in 90 days.
- How fast does someone call an inbound lead back? Response time moves conversion more than most on-page work. If leads sit for two days, we're measured on a number your sales team controls.
- Is anything already broken? A manual action, a half-finished migration, bought placements in the link profile. All fixable, none fixable inside a quarter while also growing leads.
- What else are you about to launch? Funding news, a TV campaign, a rebrand. Not disqualifying — but if brand demand is about to spike for reasons that aren't us, we'd rather say so now than take credit in month three.
Who we decline, and why it isn't snobbery
This is the part that costs us most, and it lands hardest on the businesses that most want a guarantee. Uncomfortable to publish, so here it is plainly.
- Brand-new domains with no baseline. If your trailing-90-day organic lead count is zero, one enquiry technically wins us the guarantee. That's absurd, and absurd in our favour, so we don't do it. New sites get a normal engagement with honest 6–12 month expectations.
- Businesses with no CRM and no lead tracking. We can help you build that, but we won't guarantee against a number that doesn't exist yet.
- Heavily seasonal businesses, where 90 days compares your peak to your trough or the reverse. Both directions are dishonest. Those get year-on-year measurement and a normal retainer.
- Sites behind a six-week release train. If a title-tag change needs a ticket, a sprint slot, QA and a release window, 90 days buys about one deployment. We'd be guaranteeing an outcome we can't reach the controls for.
- Anything mid-replatform. A migration in flight moves your numbers more than we will, in either direction. Better to run it properly as a project and start the clock after.
- Sites carrying a penalty or bought links. The first quarter is cleanup, and cleanup usually costs traffic before it returns any. Scoping that as recovery is honest; wrapping it in a lead guarantee isn't.
- Anyone who wants a specific keyword in position one. Not because it's hard, but because it isn't ours to promise — and a firm that promises it once will promise anything.
What a missed quarter actually costs us
Time for arithmetic, because vague risk-talk is how guarantees get sold without ever being funded.
Take an engagement at ₹75,000 a month. Miss the baseline at day 90 and the cost is the fee foregone plus every further month of delivery at zero. If it takes two extra months to clear the number, that's five months of writers, technical time, outreach and tool licences funded by three months of fee — on one account. Delivery cost is most of the fee, so the reserve isn't a rounding error.
Which is why the guarantee and the three-client cap are one decision, not two pieces of positioning. You can offer this at three clients a month, or you can offer it at forty and quietly rewrite the terms when it bites. There is no third option, and the second is how most guarantees in this market actually work.
One thing we deliberately don't publish: a hit rate. We could quote an impressive percentage, you'd have no way to check it, and it would be exactly the kind of unverifiable number this page exists to argue against. Instead we show you the baseline document and the contract clause before you sign.
Why we'll never guarantee a ranking position
This is the guarantee people ask for, and refusing it costs us deals every month. Five reasons, in order of how much they should bother you.
- Nobody controls Google's index. Google's own guidance on choosing an SEO tells you to be wary of anyone guaranteeing a number one ranking. That isn't legal caution, it's a description of how ranking works.
- Position guarantees are trivially gameable. Pick ten keywords with no search volume and you'll hit position one by Thursday. Guarantee met, report superb, pipeline unmoved. This is the most common way ranking guarantees get honoured.
- Positions aren't stable objects any more. The same query returns different results by device, location, personalisation and whichever AI summary sits on top that week. Position three is an average of a distribution, not a place.
- A ranking isn't a lead. You can win a competitive position and get fewer clicks than last year because an AI Overview absorbed the answer. We'd have kept our promise while your business got worse.
- The reliable way to hit an aggressive position guarantee is to buy links — a liability parked on your domain to satisfy a clause in ours. Why ranking guarantees are a lie covers what usually happens next.
How to read anyone else's guarantee
Apply this to us as much as to anybody. Six questions, answerable before you sign rather than on request afterwards. Most guarantees in this market survive question one and die on question three.
- What exactly is measured? Leads, revenue, traffic or positions? If it's positions, ask which keywords and what their monthly search volume is.
- Where is it measured, and who owns that system? Your CRM is good. Their rank tracker is not.
- What is the starting value, today, in writing? If nobody can state it before work begins, there's nothing to beat.
- What's the remedy? Free work, cash back, or credit against future invoices? Credits are the common answer and the weakest — a discount on a service you've just decided you don't want.
- What voids it? Real guarantees have client obligations. Reasonable ones name dev hours, approval turnaround and CRM access. Unreasonable ones void everything if you miss a single deadline.
- Who signs it? A clause in the contract is a commitment. A sentence on a landing page is copywriting.
Where we land, and what it costs you
SEO runs from ₹75,000 a month, and from ₹40,000 for smaller sites, ex-GST. Month-to-month after the first quarter, 30 days' notice, and you keep every asset if you leave — accounts, documents, drafts, the keyword map, the lot.
The guarantee sits in the contract, not on a banner: beat your frozen trailing-90-day organic lead baseline inside 90 days, or we keep working free until we do. Three clients a month, because that's what the promise can carry.
If we tell you on the first call that we won't take the work, that's the guarantee doing its job — and the reason we're declining is usually the thing capping your results regardless of who you hire. Our pricing is published in full, bundles included, so you can do the arithmetic before spending an hour on a call.