The term “desertification” is burdened by its colonial origin and later development based more on ad hoc effects descriptions than on proper understanding of the socio-ecological processes involved. It is ambiguous and cases are difficult to detect for two reasons. First, desertification is often tackled from a symptomatic, rather than systemic, viewpoint. Second, desertification is characterized using an array of attributes, a variable subset of which must be present, but none of them being decisive. In this essay, we propose a set of principles a case must fulfil to qualify as desertification. These are: (1) Desertification causes land degradation; (2) Prioritizing economic efficiency exclusively results in negative environmental and social externalities; (3) Desertification takes place in coupled but spatially separated land subsystems; (4) Desertification affects large geographical areas; (5) Resources origins and sinks in a desertification setup may be spatially telecoupled; (6) Early detecting of desertification requires to combine “slow” and “fast” variables of socio-ecosystems; (7) Abrupt land use changes can foster land degradation; (8) The perception of some "fast" socioeconomic variables can lead to a misperception of desertification; (9) Aquifers and their associated basins make coupled exploited subsystems of desertification; (10) Desertification risk is different from desertification condition; and (11) A main problem facing desertification monitoring and remediation is the desertification baseline. This proposal is timely in view of some recent frameworks that demand a precise and consistent characterization of desertification: Sustainable Development Goals in general, and particularly Land Degradation Neutrality, and Convergence of Evidence.