If you are just starting out in British family history research, the census records for England and Wales will probably be your first stop. Here you can find your ancestors named, housed, and placed in context. Ten named censuses survive for England and Wales. They were taken every decade from 1841 to 1921 and cover a period when the majority of our ancestors were alive. Earlier censuses took place in 1801, 1811, 1821 and 1831, but they recorded only population totals and contain no individual names. Learning what each census contains, and what it does not, will save you considerable time.

This is not a guide to searching techniques. It is a guide to the records themselves: what each census year recorded, how reliable it is, and what you can reasonably expect to find.

Why the census matters

A census return is a snapshot. On one night, the same night across the whole country, every household was recorded. Knowing the exact date matters: it can explain why an age looks wrong, or why a family member appears to be somewhere unexpected. The dates were as follows: 6 June 1841; 30 March 1851; 7 April 1861; 2 April 1871; 3 April 1881; 5 April 1891; 31 March 1901; 2 April 1911; 19 June 1921.

In censuses before 1911, people usually gave information verbally to an enumerator. The enumerator then copied it into the official book. This explains many spelling errors and transcription inconsistencies you may encounter. Enumerators recorded names, ages and birthplaces as they heard them, not always as intended. The 1911 census changed this, as we will come to shortly.

Find a relative in a census and you often uncover the whole household. This includes parents, children, siblings, lodgers, and servants. You may find a birthplace that points you directly to the next parish to search. You may find an age. It is often imprecise, but it can help you locate a birth or marriage certificate.

The census also tells you things no other record quite captures. It shows who lived with whom, what they did for a living, and whether a family stayed put or scattered across the country. As a source for building a picture of a life, it is hard to beat.

The censuses, one by one

The 1841 census was the first modern census conducted under the Census Act 1840, and it shows. It is the least informative of all ten returns, but it is often the only census record that takes you back far enough to matter.

What it tells you: names, approximate ages, sex, occupation, and whether the person was born in the same county in which they lived. Enumerators rounded adults’ ages down to the nearest five years, so someone listed as 40 may in fact have been anywhere from 40 to 44. If not, the enumerator might record S (Scotland), I (Ireland), or F (Foreign Parts). For those born in England and Wales, the county-of-birth column is a yes or no answer only. It will not give you a specific birthplace.

What it does not tell you: relationships between household members. A woman listed alongside a man and children might be his wife or sister. She could also be his housekeeper or an unrelated lodger. You have to infer relationships from other evidence. Ages are unreliable. The county-of-birth information is of some use as it can help confirm or rule out a theory.

The 1851 census is where British genealogical research really starts to open up. For the first time, enumerators recorded the relationship of each household member to the head of the household (wife, son, daughter, lodger, servant) and the specific parish and county of birth.

That birthplace information is transformative. Instead of knowing only that an ancestor was born somewhere in Yorkshire, you now have a parish name. That is often enough to take you directly into the parish registers and back another generation or more.

Ages were still self-reported and often inaccurate, but the additional detail makes 1851 a far richer source than 1841. If you find an ancestor here, you will usually come away with considerably more to work with.

The 1861 census follows the same format as 1851 and records the same categories of information. Some families moved frequently or are otherwise hard to trace. In these cases, multiple census returns can make the difference between finding an ancestor and losing them.

One thing worth noting: the 1861 census can be particularly useful for tracing children who were born in the late 1850s, when civil registration was still imperfectly enforced and some births went unregistered. A child appearing in the 1861 census gives you a name, an approximate birth year, and a birthplace, which may be all you have.

By 1871 the census format is well established and largely consistent with what came before. Relationships, birthplaces, ages and occupations are all recorded as in 1851 and 1861. The value of this consistency is that you can compare returns across multiple decades and start to build a longitudinal picture of a family: where they lived, how their circumstances changed, who stayed at home and who had left.

The 1871 census predates the full enforcement of compulsory education, which was introduced gradually after the Education Act of 1870 and made mandatory in 1880. This means you may still find children recorded with occupations (errand boy, agricultural labourer, lacemaker) that become less common in later returns as schooling took hold.

The format remains unchanged from 1871, but the quality of the transcription makes this census particularly forgiving when working with unusual surnames or variant spellings. It is often worth trying FamilySearch for difficult 1881 searches even if you normally use Ancestry or Findmypast.

The 1891 census introduced additional employment categories, including whether workers were employers, employed, or working on their own account. This is a small but occasionally useful detail, particularly for tracing artisans, tradespeople, and small business owners.

For Welsh families, the 1891 census is the first to include a column indicating whether a person spoke Welsh only, English only, or both. This helps place a family within the linguistic geography of Wales and can narrow a birthplace search.

In all other respects the 1891 census follows the format established in 1851. Ages remain self-reported and should be treated with appropriate scepticism.

The 1901 census records the same categories as 1891. Its particular value is that it sits at the junction between the Victorian era and living memory: many people researching today will have grandparents or great-grandparents who appear in this return, making it a natural starting point for recent family history research.

The 1901 census was one of the first that organisations digitised and made available online in the early 2000s, and its release helped spark the modern interest in genealogy research. It is well indexed and generally easy to search.

The 1911 census is a significant step forward in detail. Unlike its predecessors, where enumerators copied returns into official books, householders usually completed the 1911 census themselves. This means you are often reading the actual handwriting of your ancestor and, more practically, that the spelling of names and places reflects how the family themselves wrote them.

New information in 1911 includes the number of years a couple had been married, the number of children born to the marriage, and the number of those children still living. That last detail is often quietly devastating: a family with eight children born and five surviving tells a story that no other record quite captures.

The 1911 census also records the rooms each household occupied, which gives a sense of living conditions, and provides more detailed occupational descriptions than earlier censuses.

The 1921 census, released in January 2022 after a 100-year closure, is the most recent and in many ways the richest. It covers England and Wales on the night of 19 June 1921 and contains more information than any previous census.

In addition to the standard categories, the 1921 census records detailed information about each person’s employer and workplace, including the name and address of the business, and whether a person worked full-time or part-time. This employment information is the real innovation of the 1921 return.

It is particularly useful for researchers tracing working-class ancestors: it can place a person within a specific factory, mine, or business, and can open avenues for further research in company or trade union records.

A few things to remember across all census years

Ages are almost always unreliable to some degree. People misremembered, guessed, or deliberately adjusted their ages. A person’s age recorded in 1881 may not align with their age in 1891, and neither may match their birth certificate. Use census ages as a guide to a birth year, not as a fixed fact.

Birthplaces can be similarly inconsistent. The same person may give a different birthplace in different census years, sometimes because they moved young and genuinely did not know exactly where they were born, sometimes because the enumerator recorded what they heard rather than what was said. Treat birthplaces as strong leads rather than certainties until confirmed by a primary source.

What to do when you cannot find someone

Enumerators did not record everyone. People who were travelling, sleeping rough, or simply missed will not appear. If you cannot find someone in a census year, consider whether they might have been in an institution (a workhouse, a hospital, a prison) which would appear under a separate enumeration district.

And finally: the census only covers England and Wales. Scotland had its own census, available through ScotlandsPeople. Ireland had censuses in 1901 and 1911, both the National Archives of Ireland makes freely available. The earlier Irish censuses were almost entirely destroyed by fire in 1922.