Free · no login required
Estimate an era from sequential or letter-prefix serials; randomized serials cannot be dated.
Rolex used sequential numbers, then letter-prefixed series, and later randomized serials. Only the first two systems support a public date estimate.
Depending on age, inspect between the lugs or on the inner rehaut. Documentation can help confirm the transcription.
Serial tables are approximate, and randomized serials from roughly 2011 onward cannot be publicly dated.
These rules explain documented marking patterns. They do not prove that an item is authentic.
Rolex serials pass through three distinct systems: sequential numbers, letter-prefixed sequences, and randomized identifiers. Public tables can only estimate the first two, and their year boundaries vary between specialist datasets.
Classify the marking before consulting a date table. A randomized identifier is not a disguised year code.
| Approximate period | System | What a public lookup can show |
|---|---|---|
| Before the late 1980s | Sequential numeric serial | Approximate production period; some early sequences repeat |
| 1987–2011 transition | One letter followed by digits | Approximate prefix window |
| About 2010/2011 onward | Eight-character randomized serial | No public production year |
These windows combine widely used collector references. Adjacent prefixes overlap because cases and movements were not consumed in a perfectly linear calendar sequence.
| Prefix | Indicative window |
|---|---|
| R | 1987–1988 |
| L | 1988–1990 |
| E | 1990–1991 |
| X | 1991–1993 |
| N | 1991–1992 |
| C | 1992–1993 |
| S | 1993–1995 |
| W | 1994–1996 |
| T | 1996–1998 |
| U | 1997–1998 |
| A | 1998–1999 |
| P | 2000–2001 |
| K | 2001–2002 |
| Y | 2002–2003 |
| F | 2004–2005 |
| D | 2005–July 2006 |
| Z | July 2006–December 2007 |
| M | January–November 2008 |
| V | December 2008–2010 |
| G | 2010–2011 transition |
G-series and randomized identifiers overlap during the transition. Do not turn either system into a precise month or sale date.
A numeric table is best used as a broad consistency check. Rolex restarted its sequence, and published tables disagree at many boundaries.
| Indicative year or window | Serial band |
|---|---|
| 1926 | 1–20,189 |
| 1927–1930 | 20,190–29,131 |
| 1931–1932 | 29,132–29,561 |
| 1933 | 29,562–30,822 |
| 1934 | 30,823–34,335 |
| 1935 | 34,336–36,855 |
| 1936 | 36,856–40,919 |
| 1937 | 40,920–43,738 |
| 1938 | 43,739–71,223 |
| 1939 | 71,224–99,774 |
| 1940 | 99,775–106,046 |
| 1941 | 106,047–143,508 |
| 1942 | 143,509–230,877 |
| 1943 | 230,878–269,560 |
| 1944 | 269,561–302,458 |
| 1945 | 302,459–367,945 |
| 1946 | 367,946–529,162 |
| 1947 | 529,163–628,839 |
| 1948 | 628,840–709,248 |
| 1949–1951 | 709,249–726,638 |
| 1952 | 726,639–855,725 |
| 1953 | 855,726–999,999 |
| 1954 | 23,000–96,999 (restarted sequence) |
| 1955 | 97,000–133,060 |
| 1956 | 133,061–223,999 |
| 1957 | 224,000–327,999 |
| 1958 | 328,000–399,452 |
| 1959 | 399,453–515,999 |
| 1960 | 516,000–643,152 |
| 1961 | 643,153–743,999 |
| 1962 | 744,000–823,999 |
| 1963 | 824,000–1,008,888 |
| 1964 | 1,008,889–1,099,999 |
| 1965 | 1,100,000–1,199,999 |
| 1966 | 1,200,000–1,538,434 |
| 1967 | 1,538,435–1,751,999 |
| 1968 | 1,752,000–1,899,999 |
| 1969 | 1,900,000–2,241,881 |
| 1970 | 2,241,882–2,589,294 |
| 1971 | 2,589,295–2,890,458 |
| 1972 | 2,890,459–3,200,267 |
| 1973 | 3,200,268–3,567,926 |
| 1974 | 3,567,927–3,862,195 |
| 1975 | 3,862,196–4,115,298 |
| 1976 | 4,115,299–4,999,999 |
| 1977–1978 | 5,000,000–5,737,029 |
| 1979 | 5,737,030–6,433,999 |
| 1980 | 6,434,000–6,520,869 |
| 1981 | 6,520,870–7,099,999 |
| 1982 | 7,100,000–7,399,999 |
| 1983 | 7,400,000–8,070,021 |
| 1984 | 8,070,022–8,613,999 |
| 1985 | 8,614,000–8,899,999 |
| 1986 | 8,900,000–9,399,999 |
| 1987 | 9,400,000–9,999,999 |
Brand and archive material is preferred where it exists. Public decoding tables are identified as compiled references or observed patterns so their uncertainty stays visible.