After preliminary schooling Hollerith attended the School of Mines of Columbia University and was graduated in 1879.
Hollerith became an assistant to his teacher, William Petit Trowbridge, in the Census of 1880. He worked on the statistics of manufacturers and prepared an article, "Report on the Statistics of Steam- and Water-Power Used in the Manufacture of Iron and Steel, " for the Report on Power and Machinery Employed in Manufactures. His work on the census brought him into contact with Dr. John Shaw Billings, from whom came the suggestion of Hollerith's main invention. In a letter to a friend written nearly forty years later he described the origin of the idea: "One evening at Dr. B's tea table he said to me, 'There ought to be a machine for doing the purely mechanical work of tabulating population and similar statistics. ' " Hollerith thought the problem could be solved and later offered Billings a share in the project.
In 1882 he went to the Massachusetts Institute of Technology, as instructor in mechanical engineering. He disliked teaching, however, and after a year moved to St. Louis, Missouri, where he experimented on electromagnetically operated air-brakes and other types of brakes for railroads.
From 1884 to 1890 he was attached to the Patent Office in Washington. During these years he worked on the problem of perfecting mechanical aids in tabulating statistical information. By the time the Census of 1890 was to be taken he had invented machines that would record statistical items, by a system of punched holes in a non-conducting material, and would also count those items by means of an electric current passed through the holes identically placed. The system was given trial in tabulating mortality statistics in Baltimore, and in compiling similar data in New Jersey and New York City.
In competition with two alternative methods of tabulation, it was chosen for use in compiling the Census of 1890. It did a sample piece of work in less than half the time required by the other systems, and the commission estimated that in dealing with the returns expected at the approaching census the new machine would reduce the labor days by more than two-thirds. Subsequently the machines were improved by the addition of a mechanical feeding device.
In 1890 the Franklin Institute of Philadelphia, reporting that Hollerith had made the outstanding invention of the year, gave him its highest award, the Elliott Cresson medal. The Hollerith machines were used in 1891 in recording the census returns in Canada, Norway, and Austria. Although they revolutionized statistical technique, American scholars gave little attention to them at the outset, probably because statistical interpretation had not been carried as far in the United States as elsewhere. But in Europe technical articles about their value appeared in England, France, Germany, Austria, and Italy.
Hollerith attended the Berne session of the International Statistical Institute in 1895 and commented upon a paper by an Austrian member. Between 1890 and 1900 the machines were successfully adapted to handle types of mass enquiries in which addition was an element, and thus they could be used in tabulating railroad freight statistics and the data assembled in the agricultural census.
In 1896 Hollerith organized the Tabulating Machine Company, incorporated in New York, to manufacture the machines and to sell the cards used with them. In 1911 that company was consolidated with the Computing Scale Company of America and the International Time Recording Company of New York to become the Computing-Tabulating-Recording Company, later known as the International Business Machines Corporation, of which Hollerith was retained as consulting engineer until 1921.
Hollerith died at his home in Washington, of heart disease, at the age of sixty-nine.
On September 15, 1890, Hollerith married Lucia Beverly Talcott. She, with their six children, Lucia, Nannie, Virginia, Herman, Richard, and Charles, survived him.