Market statistics · Methodology
How Laudvi Measures Local Market Reputation
The short answer
This page explains, in plain language, exactly what our numbers mean and how they’re produced, so a business owner, a journalist, or a search engine can understand and cite them accurately. It applies to every Laudvi market-statistics page. This is methodology version 1.0.0.
What we measure
For a given trade in a given local market, we report a small set of aggregate figures: how many businesses we analyzed, the median Google star rating, the median Google review count, and the range that the middle 50% of businesses fall into. These describe the group as a whole. We deliberately do not publish any single business’s numbers, and we don’t publish the highest or lowest values, because a single extreme business tells you nothing useful and could identify a specific company.
How we define a market
A market is defined by geography, not by a mailing address or a directory listing. A business belongs to a market when its actual location falls within a defined radius of the market’s center. Our first combined market, Sarasota–Bradenton, is the deduplicated union of two areas — roughly a 22-kilometer radius around Sarasota and an 18-kilometer radius around Bradenton. A business that would fall in both is counted once. Because it’s built from location, the market is not simply “every business that shows up for a city search.”
How we classify a trade
A business belongs to a trade (for example, HVAC) when its primary business category matches that trade’s defined set of categories. We keep a controlled list of the categories that count as core to each trade and a list of categories we explicitly exclude — for instance, general contractors, hardware stores, and auto shops that can otherwise contaminate a home-services trade. Classification is decided by these rules, consistently, not case by case.
Which businesses are counted
A business enters the analyzed group only when all of the following are true:
- its location is confirmed inside the market’s geographic boundary;
- its category is confirmed as part of the trade;
- it is not marked permanently closed;
- it is not flagged as contamination (an adjacent business type that isn’t really in the trade).
A business that is missing either confirmation is left out rather than assumed in — so the group never quietly includes something we couldn’t confirm.
How we avoid double-counting
The same business can appear more than once across the underlying data. We merge records only when a strong, unique identifier matches — never on a similar-sounding name alone, which would risk merging two genuinely different companies. Each real business is counted once.
Sample size, and why it’s a sample
Every statistic is reported alongside the number of businesses behind it. That number is a sample — the businesses we were able to analyze in the market — not a guarantee that we captured every business that exists. We only publish a market’s statistics when the analyzed group is large enough (at least 100 businesses) that a median is meaningful and no single business can be inferred from it. We do not claim government- or census-grade coverage.
Median, middle 50%, and percentiles
The median is the midpoint: half the businesses are above it and half below. It’s a better “typical” figure than an average because a few businesses with enormous review counts would drag an average upward and mislead. The middle 50% is the range from the 25th to the 75th percentile — the band that the middle half of businesses fall into. A percentile is calculated as the share of businesses at or below a value, using standard linear interpolation. Review-count figures are rounded to whole numbers and ratings to two decimal places (a median across many businesses can carry more precision than any single one-decimal Google rating).
What’s excluded
We exclude permanently closed businesses, businesses we couldn’t confirm as being in the market or the trade, and contamination categories. When a business has no Google review count on record, it still counts toward the market’s size but not toward the median review count — so we also report how many businesses the review figures are based on.
Snapshots and dates
Each statistic comes from a fixed snapshot — a point-in-time reading of the data, not a live query against Google. That’s why every page shows a “Data collected” month. The snapshot is stable: the number you see and cite won’t silently change out from under you. When we refresh the data, we produce a new snapshot with a new date rather than editing the old figures.
Freshness and limitations
“Data collected” tells you when the underlying readings were taken; it is not a claim that the numbers are current to today. If a snapshot ages past our freshness window, we remove the numbers rather than present stale figures as current. These are honest limitations worth stating plainly: the figures are a sample of one point in time for one market and trade, drawn from public business data; they are not a live feed, not a complete census, and not a prediction of any individual business’s results.
Why only anonymous aggregates
We publish market-level aggregates only. No page identifies a specific business, and no individual business’s rating or review count is shown. The published figures are computed by Laudvi from the analyzed group — they are our own aggregate market intelligence, and they are designed so that no individual business can be reconstructed from them.
See where your business stands
These market figures answer “what’s typical?” To see how your own numbers compare, use Compare My Business — you enter your own Google rating and review count, and it places them against the same anonymous local benchmark. Your figures stay yours; we never pre-fill a business, rating, or review count.