State of Local Lead Generation 2026: What We Measured

This is everything we can state with a measurement behind it, gathered in one place. 285,733 businesses checked across 98 cities in 22 countries, of which 88,803 — 31.1% — have no website at all.

Everything here is measured rather than surveyed, and unlike most annual reports in this industry the figures are queried when this page is built rather than typed into it, so what you are reading is the index as it stands rather than as it stood when somebody wrote a paragraph.

Read live at build time. Countries under 500 checked businesses are not shown.

The spread is the finding rather than any single row. A fourfold difference between the top and bottom of that table means no global figure is usable for a decision, and the variation inside each country is wider still.

Indian cities by gap rate, at least 400 checked. Also read live.

The pattern holds across every market we measure: the gap runs inversely to city size. Second cities carry it, capitals do not. In Britain that means Liverpool and Birmingham against London; in India, Morena and Kota against Bengaluru and Mumbai.

Inside a single city the spread is nearly as wide — neighbourhood-level rates range from the mid twenties to above fifty percent — which makes the area a better planning unit than the city.

Category variation is the widest axis in the entire dataset, over tenfold from top to bottom, and it beats both country and city as a predictor.

The last row against the first is the whole argument about niche lists.

The single most consistent finding across all of it: the gap is a property of the business model rather than of the country. Every high-gap category anywhere has the same shape — a local customer who arrives in person and chooses on proximity rather than research. Country moves the level; the business model sets the order.

The finding we did not expect, and the one most useful in a sales conversation.

Businesses without websites carry roughly a third of the Google reviews of businesses with them — a median of 36 against 120 in India — and the effect holds inside every large category we tested, so it is not a category artefact.

But their ratings are identical. 4.77 against 4.74 for service businesses, 4.79 against 4.79 for educational institutions, and in two categories the businesses without websites rate slightly higher.

Which rules out the convenient explanation. They are not worse businesses. They are equally good businesses that fewer people know about — and the causation is genuinely unproven in either direction, so nobody should be promising an owner that a website produces reviews.

Google's Places data carries no email field, which is why the third row is not a coverage gap.

That table decides the channel question on its own. Email tooling — finders, verifiers, sequencers — returns blanks against this market by construction, and the entire outbound stack built around a named person at a domain has nothing to attach to.

What works instead, in order: walking in for anything counter-based, WhatsApp at low volume for appointment businesses, and the phone for verticals where the owner sits at a desk. Cold calling here does not fail at the pickup — a business line is published on purpose and gets answered — it fails at the handoff, because whoever is nearest the counter answers it.

The numbers that decide whether any of this is a business.

Data. A licensed Places call costs about ₹3.08 and returns several businesses, so the underlying data is roughly ₹1–10 a prospect. Scraped Maps data is cheaper still at around ₹0.26 a record, with terms-of-service exposure as the trade.

Bought leads. Shared leads run around ₹800 and exclusive around ₹3,000, at conversion rates that put cost per customer above ₹40,000 either way — which does not work against a ₹25,000 project. That entire model is priced for industries where a customer is worth ten to a hundred times more.

Conversations. Roughly one in twenty qualified conversations closes, half of contact attempts reach a decision-maker, and ten clients takes about 250 qualified prospects.

Lifetime value. A ₹25,000 first website is worth roughly ₹1.5 lakh over the relationship on conservative assumptions, of which the build is about 17%. The care plan and referrals are two-thirds.

It is our index, not a census. We cover 98 cities. A business we have never scanned is not in these figures, and a country low in the table may simply be one we have not worked.

Every rate is conservative. A business counts as having a website if there is anything in its listing's website field — including a Facebook page, a directory profile, or a domain that no longer loads. The real first-website opportunity is larger than every number here.

Correction runs one way. When a later check finds a business has built a site, we correct the record, which makes our own headline figure smaller over time. That is the right behaviour for a measurement and the wrong behaviour for a marketing statistic, and it is why these numbers move when the widely-quoted ones do not.