What is this paper about?This text does not contain any new results, it is just an attempt to present, in a systematic way, one construction which makes it possible to use some ideas and notions well-known in the theory of integrable systems on Lie algebras to a rather different area of mathematics related to the study of projectively equivalent Riemannian and pseudo-Riemannian metrics. The main observation can be formulated, yet without going into details, as follows:The curvature tensors of projectively equivalent metrics coincide with the Hamiltonians of multi-dimensional rigid bodies.Such a relationship seems to be quite interesting and may apparently have further applications in differential geometry. The wish to talk about this relation itself (rather than some new results) was one of motivations for this paper.The other motivation was to draw reader's attention to the argument shift method developed by A. S. Mischenko ana A. T. Fomenko [49] as a generalisation of S. V. Manakov's construction [44]. This method is, in my opinion, a very simple, natural and universal construction which, due to its simplicity, naturality and universality, occurs in different ares of modern mathematics. There are only a few constructions in mathematics of this kind. In this paper, the argument shift method is only briefly mentioned but the main subject, the so-called sectional operators, is directly related to it.We discuss some new results obtained in our three papers [9,10,14]. In this sense, the present work can be considered as a review, but I would like to shift the accent from the results to the way how using algebraic properties of sectional operators helps in solving geometric problems. That is why the exposition is essentially different from the above mentioned papers and details of the proofs, which are not directly related to our main subjects, are omitted. Two first sections are devoted to the definition and properties of sectional operators, in the following four, we discuss their applications in geometry. Also I would like to especially mention the note [8], in which we discussed the properties of sectional operators in a very general setting and which was conceptually very helpful for this paper. I am very grateful to all of my coauthors, Vladimir Matveev, Volodymyr Kiosak, Dragomir Tsonev, Stefan Roseman and Andrey Konyaev.