This paper presents an overview of how to run the CCP4 programs for data reduction (SCALA, POINTLESS and CTRUNCATE) through the CCP4 graphical interface ccp4i. It covers the determination of the point-group symmetry of the diffraction data (the Laue group), which is required for the subsequent scaling step, examination of systematic absences, which in many cases will allow inference of the space group, putting multiple data sets on a common indexing system when there are alternatives, the scaling step itself, which produces a large set of data-quality indicators, estimation of |F| from intensity and finally examination of intensity statistics to detect crystal pathologies such as twinning. An appendix outlines the scoring schemes used by the program POINTLESS to assign probabilities to possible Laue and space groups. The true space group is only a hypothesis until the structure has been solved. The program POINTLESS examines the symmetry of the diffraction pattern and scores the possible crystallographic symmetry. It uses correlation coefficients rather than R factors to score rotational symmetry operations, as they are less dependent on the unknown scales. A probability is estimated from the correlation coefficient, using equivalent-size samples of unrelated observations to estimate the width of the probability distribution. The protocol for determination of space group in POINTLESS is as follows: (i) From the unit-cell dimensions and lattice centring, find the highest compatible lattice symmetry within some tolerance, ignoring any input symmetry information. (ii) Score each potential rotational symmetry element belonging to the lattice symmetry using all pairs of observations related by that element. (iii) Score combinations of symmetry elements for all possible subgroups of the lattice-symmetry group (Laue or Patterson groups). (iv) Score possible space groups from axial systematic absences. (v) Scores for rotational symmetry operations are based on correlation coefficients rather than R factors. The paper also discusses scaling, which tries to make symmetry-related and duplicate measurements of a reflection equal by modelling the diffraction experiment. After scaling, the remaining differences between observations can be analysed to give an indication of data quality. The program SCALA outputs a large number of statistics, mostly presented as graphs, and a final summary table which contains most of the data required for the traditional 'Table 1' in a structural paper. The paper also discusses intensity statistics and crystal pathologies, including twinning. It describes how to detect anomalous signals and estimate amplitude |F| from intensity I. The paper concludes with a summary of the main questions and decisions in data reduction.This paper presents an overview of how to run the CCP4 programs for data reduction (SCALA, POINTLESS and CTRUNCATE) through the CCP4 graphical interface ccp4i. It covers the determination of the point-group symmetry of the diffraction data (the Laue group), which is required for the subsequent scaling step, examination of systematic absences, which in many cases will allow inference of the space group, putting multiple data sets on a common indexing system when there are alternatives, the scaling step itself, which produces a large set of data-quality indicators, estimation of |F| from intensity and finally examination of intensity statistics to detect crystal pathologies such as twinning. An appendix outlines the scoring schemes used by the program POINTLESS to assign probabilities to possible Laue and space groups. The true space group is only a hypothesis until the structure has been solved. The program POINTLESS examines the symmetry of the diffraction pattern and scores the possible crystallographic symmetry. It uses correlation coefficients rather than R factors to score rotational symmetry operations, as they are less dependent on the unknown scales. A probability is estimated from the correlation coefficient, using equivalent-size samples of unrelated observations to estimate the width of the probability distribution. The protocol for determination of space group in POINTLESS is as follows: (i) From the unit-cell dimensions and lattice centring, find the highest compatible lattice symmetry within some tolerance, ignoring any input symmetry information. (ii) Score each potential rotational symmetry element belonging to the lattice symmetry using all pairs of observations related by that element. (iii) Score combinations of symmetry elements for all possible subgroups of the lattice-symmetry group (Laue or Patterson groups). (iv) Score possible space groups from axial systematic absences. (v) Scores for rotational symmetry operations are based on correlation coefficients rather than R factors. The paper also discusses scaling, which tries to make symmetry-related and duplicate measurements of a reflection equal by modelling the diffraction experiment. After scaling, the remaining differences between observations can be analysed to give an indication of data quality. The program SCALA outputs a large number of statistics, mostly presented as graphs, and a final summary table which contains most of the data required for the traditional 'Table 1' in a structural paper. The paper also discusses intensity statistics and crystal pathologies, including twinning. It describes how to detect anomalous signals and estimate amplitude |F| from intensity I. The paper concludes with a summary of the main questions and decisions in data reduction.