Context. Stellar feedback in high-redshift galaxies plays an important, if not dominant, role in the re-ionisation epoch of the Universe. Because of their extreme star formation (SF), the relatively closer green pea (GP) galaxies are postulated as favorite local laboratories, and analogues to those high-redshift galaxies. However, at their typical redshift of z ∼ 0.2, the most intimate interaction between stars and the surrounding interstellar medium cannot be disentangled. Detailed studies of blue compact dwarf (BCD) galaxies sharing properties with GP galaxies are necessary to anchor our investigations on them. Aims. We want to study in detail UM 462, which is a BCD with emission line ratios and equivalent widths, stellar mass, and metallicity similar to those observed in GP galaxies, and thus it is ideally suited as a corner stone and reference galaxy. Methods. We use high-quality optical integral field spectroscopy data obtained with MUSE on the ESO Very Large Telescope. Results. The electron density (ne) and temperature (Te) were mapped. Median Te decreases according to the sequence [S III] → [N II] → He I. Furthermore, Te([S III]) values are ∼13 000 K, and uniform within the uncertainties over an area of ∼20″ × 8″ (∼1.4 kpc × 0.6 kpc). The total oxygen abundance by means of the direct method is 12 + log(O/H) ∼ 8.02 and homogenous all over the galaxy within the uncertainties, which is in stark contrast with the metallicities derived from several strong line methods. This result calls for a systematic study to identify the best strategy to determine reliable metallicities at any location within a galaxy. The strong line ratios used in the BPT diagrams and other ratios tracing the ionisation structure were mapped. They are compatible with plasma ionised by massive hot stars. However, there is a systematic excess in the [O I]/Hα ratio, suggesting an additional mechanism or a complex relative configuration of gas and stars. The velocity field for the ionised gas presents receding velocities in the east and approaching velocities in the west and south-west with velocity differences of Δv ∼ 40 km s−1, but it is not compatible with simple rotation. The most striking feature is a velocity stratification in the area towards the north with redder velocities in the high ionisation lines and bluer velocities in the low ionisation lines. This is the only area with velocity dispersions clearly above the MUSE instrumental width, and it is surrounded by two ∼1 kpc-long structures nicknamed ‘the horns’. We interpret the observational evidence in that area as a fragmented super-bubble fruit of the stellar feedback and it may constitute a preferred channel through which Lyman continuum photons from the youngest generation of stars can escape. The galaxy luminosity is dominated by a young (i.e. ∼6 Myr) stellar population that contributes only 10% to the stellar mass, as derived from the modelling of the stellar continuum. The most recent SF seems to propagate from the outer to the inner parts of the galaxy, and then from east to west. We identified a supernova remnant and Wolf-Rayet stars – as traced by the red bump – that support this picture. The direction of the propagation implies the presence of younger Wolf-Rayet stars at the maximum in Hα. These may be detected by deep observations of the blue bump (not covered here). Conclusions. The ensemble of results exemplifies the potential of 2D detailed spectroscopic studies of dwarf star-forming galaxies at high spatial resolution as a key reference for similar studies on primeval galaxies.