The formula, and the two inputs everyone leaves out
Most keyword research stops at volume and difficulty. Two columns, sorted descending, and the plan writes itself. Fast, tidy, and in B2B wrong almost every time.
The revenue a keyword can produce is a chain of five multipliers, and volume is only the first:
Monthly searches → click-through at your realistic position → conversion to a genuine next step → close rate → deal value.
Volume gets the attention because it's the only one a tool hands you. The last two decide the outcome, and they live in your CRM. Multiply properly and the hierarchy flips: a 40-search query with a 5% demo rate and an ₹8 lakh contract value beats a 40,000-search head term with a 0.8% order rate and a ₹1,200 basket — not slightly, by a multiple.
The worked example: 40 searches versus 40,000
A model, not a case study. Every assumption is stated so you can swap in your own — the point is the shape, not these figures. Left column: a B2B software company with an ₹8 lakh average contract value, targeting a narrow bottom-funnel query. Right column: a D2C brand with a ₹1,200 average order value, targeting a category head term.
| Step | B2B keyword — 40 searches/mo | B2C keyword — 40,000 searches/mo |
|---|---|---|
| Monthly searches | 40 | 40,000 |
| Realistic CTR at position 1–3 | ~30%. Narrow query, few ads, little competition. | ~12%. AI Overviews, shopping units, ads and People Also Ask take the screen before your listing. |
| Clicks per month | 12 | 4,800 |
| Conversion to a real next step | 5% to a demo request — only a buyer types this query. | 0.8% to an order — a head term is mostly research intent, not purchase intent. |
| Per month | 0.6 demos | 38 orders |
| Per year | ≈7 demos | ≈460 orders |
| Close rate | 25% of demos become customers | n/a — the order is the sale |
| Annual revenue | ≈1.8 deals × ₹8,00,000 = ₹14.4 lakh | 460 × ₹1,200 = ₹5.5 lakh |
| Gross profit at typical margins | ~80% software margin ≈ ₹11.5 lakh | ~35% D2C margin ≈ ₹1.9 lakh |
| Difficulty to rank | Low. Almost nobody writes for a 40-search query. | High. The whole category is fighting for it, some of them for a decade. |
Why the B2B number is fragile, and what to do about it
Honesty check, because that table is more precarious than it looks.
1.8 deals a year means the real outcome is one deal or two. One either way moves the ₹14.4 lakh figure by ₹8 lakh — the entire variance of the model sitting inside a single sales conversation. Three consequences follow, and they're the difference between a plan that survives a board meeting and one that doesn't.
- Never build a B2B case on one keyword. Build it on 30–50 buying queries. Individually each is noise; collectively they're a forecast. The maths only stabilises at portfolio level.
- Model the pessimistic case out loud. Present ₹14.4 lakh and deliver one deal and you've missed by half. Present a range — one to three deals — and you were right either way.
- Treat close rate as a shared number. If sales closes 10% instead of 25%, SEO didn't underperform; the model did. Agree it with the sales lead before anyone publishes a projection.
Content formats that actually convert on each side
The inverted maths changes what you build, not just what you target. B2B and B2C reward almost opposite formats, and the commonest failure is a B2B company running a B2C content plan because that's what the blog templates assume.
B2B: bottom-funnel first, awareness later
Build the pages a buyer reads when they already have a problem and a budget. Low volume, hard to write well, almost no competition.
- "X vs Y" comparison pages, including against competitors. Your buyer runs that search whether or not you show up. Someone owns the page; it may as well be you, honestly written.
- "Alternatives to X" pages for whatever your buyer is currently stuck with.
- Pricing pages that contain prices. Hiding the number costs more qualified traffic than it protects.
- Integration and compatibility pages — "does it work with Tally", "SAP integration". Tiny volume, near-perfect intent.
- Security, compliance and procurement content. The person who kills the deal sits in legal or IT.
- Calculators and templates that make the reader input their own numbers. Nothing qualifies a lead like arithmetic about their own business.
B2C: coverage, freshness and proof
Breadth and conversion rate. Volume genuinely matters here, because the deal size can't carry a thin funnel.
- Category and product pages that are the destination, not a thin gateway to a PDP.
- "Best X under ₹Y" and buying guides — high commercial intent, and the price filter qualifies for you.
- Specification, size, ingredient and material explainers. They win the long tail and quietly reduce returns.
- Review-rich pages. Real user content is a conversion lever and a source of long-tail coverage you'd never think to write.
- Freshness. A B2C category page from 2023 loses to one updated last month, in a way B2B pages simply don't.
Attribution across a six-month B2B cycle
This is where B2B SEO gets killed in board meetings — not because it didn't work, but because nobody could prove it did.
A six-month cycle means somebody reads your comparison page in January on a work laptop, forgets you, gets budget in April, searches your brand directly, and converts in June from another device. Default GA4 attribution calls that direct, and the page that did the work gets nothing. Fix it in the CRM, not the analytics — no configuration makes a tool built for single-device paths see a six-month committee decision.
- Stamp landing page and referrer onto every form submission as hidden fields, pushed into the CRM record. Survives cookie expiry, device switching and incognito.
- Add a self-reported "how did you first hear about us" field. Imperfect, and still the most useful attribution data most B2B companies have.
- Measure organic pipeline, not revenue, for the first two quarters. Revenue lags too far to steer with; pipeline says the same thing four months earlier.
- Set the baseline before anything starts. Freeze trailing-90-day qualified organic leads. Without it, month seven becomes an argument about whose dashboard is right — and it's exactly what our guarantee measures against: beat your own frozen baseline in 90 days or we keep working free until we do. See how to set an SEO baseline.
- Track first-touch separately from last. In B2B the first touch is the informational page and the last is your brand name. Last-touch-only reporting concludes that SEO's job is ranking for your own company name.
Why B2B sites need fewer pages, built deeper
The last inversion, and the one most agencies get backwards because page-count contracts are easier to sell.
A B2C site can justify thousands of URLs — every product, category and filter with real demand behind it. A B2B site usually has a few hundred addressable queries, perhaps 40 of them genuinely commercial. Publishing 400 posts against 40 real queries doesn't cover more ground; it creates keyword cannibalisation, splits internal links across near-duplicates, and starves the pages meant to win.
B2B pages also serve more than one reader. The champion wants to know it fixes their problem. Their manager wants the price. IT wants the security page. Procurement wants contract terms. A 700-word post serves one of those people; a proper page serves all four, which is why it takes three days rather than three hours.
The rule: if you can't name the buying question a page answers and the person who asks it, don't build it. In B2C that leaves you with an empty site. In B2B it leaves you with the only pages that were going to earn anything.
We run SEO from ₹75,000/mo, and on B2B accounts that buys fewer pages than clients expect and considerably more work per page. SEO for SaaS and SEO for B2B go deeper on what that looks like month to month.