Fill rate: what to check before you buy a business list
Auspector ·
Fill rate is the share of records in a list in which a given field is actually filled. It turns a record count into a usable count: of our 24,256,924 US records, 83.2% carry a phone number — 20,182,528 records with one — while the share “with a website” in the same market runs from 57.2% to 63.3% depending on what you count as a website.
What is fill rate?
Fill rate is the number of records in which a field is filled, divided by all records — calculated one field at a time. In the United Kingdom, 5,109,461 records with a phone fill rate of 72.9% make 3,724,315 records with a phone number. For your purposes, a database is only as large as the field you need.
Two things fill rate does not tell you. It does not say the value is right: a number on a record is not a number someone answers. That is accuracy, and only a sample shows it. Nor does it say how many distinct businesses stand behind the records — more on duplicates below.
One rate per field, never an average
Phone, address, website and category are filled at very different rates, and the field you need is the only one that matters. A single “average fill rate” can hide a phone rate far below it. For a call team the number is phone; for a field team, address and district; for a web agency, the records without a website.
Combined rates help only when they are defined. Ours is “complete”: address, phone, category and district all present on the same record — website is deliberately left out. In the United States 68.5% of records are complete, in the United Kingdom 62.1%.
Why “website” needs a definition
No field moves as much with its definition. A web address on a record can be the business’s own domain, a page on a booking, appointment or ordering platform, or a social media profile. Count all three and you get one number; count own domains only and you get another.
| Market | Own website | Platform page | Social profile only | No web address |
|---|---|---|---|---|
| United States | 57.2% | 4.5% | 1.6% | 36.7% |
| United Kingdom | 53.1% | 9.4% | 2.2% | 35.3% |
| Canada | 59.1% | 2.6% | 2.1% | 36.2% |
| Australia | 59.7% | 6.2% | 2.4% | 31.7% |
Own websites alone: 57.2% of US records. Add platform pages and you get what our coverage report calls “website information”: 61.7%. Add social profiles as well: 63.3%. In the United Kingdom the same three readings give 53.1%, 62.5% and 64.7%. None of them is wrong, but a website rate quoted without its definition cannot be compared with anything. How the share varies across 246 territories is in what share of businesses have a website.
Why fill rate differs by country and by segment
Across our 15 priced markets the phone fill rate runs from 66.2% (Saudi Arabia) to 83.2% (United States), and website information from 18.1% (Saudi Arabia) to 73.4% (Netherlands). A supplier’s global average says little about your market.
| Market | Records | Phone | Website information | Complete |
|---|---|---|---|---|
| United States | 24,256,924 | 83.2% | 61.7% | 68.5% |
| Germany | 6,827,622 | 74.0% | 68.2% | 60.8% |
| France | 6,255,198 | 67.6% | 56.9% | 59.7% |
| Italy | 5,284,956 | 67.8% | 50.6% | 64.1% |
| United Kingdom | 5,109,461 | 72.9% | 62.4% | 62.1% |
| Türkiye | 4,240,250 | 67.6% | 27.9% | 64.1% |
| Canada | 3,928,438 | 80.7% | 61.6% | 46.9% |
| Spain | 3,820,491 | 67.2% | 52.0% | 62.4% |
| Australia | 2,453,567 | 76.3% | 65.9% | 58.3% |
| Netherlands | 1,814,661 | 78.5% | 73.4% | 69.2% |
| Saudi Arabia | 1,116,655 | 66.2% | 18.1% | 61.2% |
| Switzerland | 1,037,581 | 77.1% | 70.6% | 72.2% |
| United Arab Emirates | 676,797 | 79.2% | 40.2% | 72.2% |
| Sweden | 601,539 | 72.8% | 67.3% | 64.7% |
| Singapore | 331,792 | 76.0% | 39.4% | 35.5% |
Within one country, segments differ as much. In the United States, website information is lowest for auto repair shops (43.9%) and highest for restaurants (73.1%); phone runs from 80.9% for cafés to 97.0% for real estate agencies.
| Segment | Records | Phone | Website information |
|---|---|---|---|
| Restaurants | 693,619 | 91.6% | 73.1% |
| Cafés | 111,848 | 80.9% | 69.3% |
| Hair and beauty salons | 609,882 | 89.6% | 48.3% |
| Real estate agencies | 731,953 | 97.0% | 72.2% |
| Auto repair shops | 384,915 | 91.4% | 43.9% |
| Construction companies | 534,718 | 94.5% | 56.1% |
New records are a segment too: 71.6% of the US records that arrived in the 30 days to October 23, 2026 carry a phone number, against 83.2% across all US records. A rate measured on the whole database tells you little about the slice you are actually buying.
Which country’s business data is the most complete?
Completeness — address, phone, category and district all on the same record — is also a property of territories, and it ranges enormously. The ranking below uses the same measurement as the market table above, restricted to the 229 territories with at least 1,000 records; tinier territories swing too much to rank.
| Territory | Records | Complete records |
|---|---|---|
| Belarus | 182,864 | 83.0% |
| Taiwan | 1,276,484 | 82.6% |
| Russia | 3,993,926 | 80.3% |
| Liechtenstein | 5,344 | 79.2% |
| Japan | 7,222,870 | 77.7% |
| Luxembourg | 60,091 | 76.9% |
| Bahrain | 60,208 | 76.0% |
| South Korea | 3,810,686 | 75.5% |
| San Marino | 3,914 | 74.9% |
| Poland | 2,456,144 | 73.6% |
One caution at the top: Liechtenstein and San Marino sit under 10,000 records — the rate is real, the base is small.
| Territory | Records | Complete records |
|---|---|---|
| Northern Mariana Islands | 6,423 | 23.4% |
| Sudan | 183,902 | 23.4% |
| Syria | 275,091 | 23.1% |
| Turks & Caicos Islands | 1,492 | 20.8% |
| St. Pierre & Miquelon | 1,835 | 15.3% |
| Eritrea | 3,690 | 10.2% |
| Micronesia | 5,517 | 10.1% |
| Marshall Islands | 2,554 | 8.0% |
| Kiribati | 4,807 | 6.9% |
| Bermuda | 223,629 | 1.2% |
At the bottom, Turks & Caicos Islands, St. Pierre & Miquelon, Eritrea, Marshall Islands, and Kiribati carry under 5,000 records; treat those rates as descriptions of thin coverage rather than verdicts.
| Territory | Records | Complete records | Rank |
|---|---|---|---|
| United States | 24,256,924 | 68.5% | 26 |
| United Kingdom | 5,109,461 | 62.1% | 52 |
| Germany | 6,827,622 | 60.8% | 59 |
| France | 6,255,198 | 59.7% | 67 |
| Spain | 3,820,491 | 62.4% | 48 |
| Türkiye | 4,240,250 | 64.1% | 43 |
Duplicates make a fill rate look better than it is
A record is a place, so a chain with forty branches can put the same head-office number on forty records: forty filled phone fields, one conversation. The same happens with businesses that work at the customer’s address rather than their own — one operator, many listed areas. In our new US records, keeping one record per phone number would remove 31.6% of the dryer vent cleaning records that have a phone and 10.2% of the dumpster rental records, against 1.7% of restaurant records. Ask for unique phone numbers, not filled phone fields.
Six questions to ask any data supplier
- When was it measured? A fill rate without a date may describe a database that no longer exists.
- For which field? One rate each for phone, address, website and category — not one average.
- For which segment? Your country, your categories, new or all records.
- By which definition? What counts as a website, a phone number, a complete record?
- How many unique values? Unique phone numbers, not filled phone fields.
- Can I see a random sample of my segment? Not a hand-picked showcase.
How to check a sample yourself
The quickest test takes an hour. Draw 100 records at random from the segment you would buy — not the first 100 rows of the file. Note how many of them have each field you need: that is your fill rate. Check how many of the phone numbers are distinct. Then try ten numbers: how many reach the business is accuracy, the one thing a fill rate cannot measure.
Our own fill rates for all 246 territories are in the coverage report, counted over the whole database rather than estimated from a sample. More questions worth asking before you buy are in business databases: three kinds, and how fill rate feeds into the price of a lead in new business leads: where they come from, how they’re priced.