Context. A key subject in extragalactic astronomy concerns the chronology and driving mechanisms of bulge formation in late-type galaxies (LTGs). The standard scenario distinguishes between classical bulges and pseudo-bulges (CBs and PBs, respectively), the first thought to form monolithically prior to disks and the second gradually out of disks. These two bulge formation routes obviously yield antipodal predictions on the bulge age and bulge-to-disk age contrast, both expected to be high (low) in CBs (PBs). Aims. Our main goal is to explore whether bulges in present-day LTGs segregate into two evolutionary distinct classes, as expected from the standard scenario. Other questions motivating this study center on evolutionary relations between LTG bulges and their hosting disks, and the occurrence of accretion-powered nuclear activity as a function of bulge stellar mass ℳ⋆ and stellar surface density Σ⋆. Methods. In this study, we have combined three techniques – surface photometry, spectral modeling of integral field spectroscopy data and suppression of stellar populations younger than an adjustable age cutoff with the code REMOVEYOUNG (ℛ𝒴) – toward a systematic analysis of the physical and evolutionary properties (e.g., ℳ⋆, Σ⋆ and mass-weighted stellar age 〈t⋆〉ℳ and metallicity 〈Z⋆〉ℳ, respectively) of a representative sample of 135 nearby (≤ 130 Mpc) LTGs from the CALIFA survey that cover a range between 108.9 M⊙ and 1011.5 M⊙ in total stellar mass ℳ⋆,T. In particular, the analysis here revolves around ⟨δμ9G⟩, a new distance- and formally extinction-independent measure of the contribution by stellar populations of age ≥ 9 Gyr to the mean r-band surface brightness of the bulge. We argue that ⟨δμ9G⟩ offers a handy semi-empirical tracer of the physical and evolutionary properties of LTG bulges and a promising means for their characterization. Results. The essential insight from this study is that LTG bulges form over 3 dex in ℳ⋆ and more than 1 dex in Σ⋆ a tight continuous sequence of increasing ⟨δμ9G⟩ with increasing ℳ⋆, Σ⋆, 〈t⋆〉ℳ and 〈Z⋆〉ℳ. Along this continuum of physical and evolutionary properties, our sample spans a range of ~ 4 mag in ⟨δμ9G⟩: high-⟨δμ9G⟩ bulges are the oldest, densest and most massive ones (〈t⋆〉ℳ ~ 11.7 Gyr, Σ⋆ > 109 M⊙ kpc−2, ℳ⋆ ≥ 1010 M⊙), whereas the opposite is the case for low-⟨δμ9G⟩ bulges (〈t⋆〉ℳ ~ 7 Gyr) that generally reside in low-mass LTGs. Furthermore, we find that the bulge-to-disk age and metallicity contrast, as well as the bulge-to-disk mass ratio, show a positive trend with ℳ⋆,T, raising from, respectively, ~ 0 Gyr, ~ 0 dex and 0.25 to ~ 3 Gyr, ~ 0.3 dex and 0.67 across the mass range covered by our sample. Whereas gas excitation in lower-mass (≲ 109.7 M⊙) bulges is invariably dominated by star formation (SF), LINER- and Seyfert-specific emission-line ratios were exclusively documented in high-mass (≳ 1010 M⊙), high-Σ⋆ (≳ 109 M⊙ kpc−2) bulges. This is in agreement with previous work and consistent with the notion that the Eddington ratio or the black hole-to-bulge mass ratio scale with ℳ⋆. The coexistence of Seyfert and SF activity in ~20% of higher-ℳ⋆, high-Σ⋆ bulges being spectroscopically classified as Composites suggests that the onset of AGN-driven feedback does not necessarily lead to an abrupt termination of SF in LTG nuclei. Conclusions. The continuity both in the properties of LTG bulges themselves and in their age and metallicity contrast to their parent diskssuggests that these components evolve alongside in a concurrent process that leads to a continuum of physical and evolutionary characteristics. Our results are consistent with a picture where bulge growth in LTGs is driven by a superposition of quick-early and slow-secular processes, the relative importance of which increases with ℳ⋆,T. These processes, which presumably combine in situ SF in the bulge and inward migration of material from the disk, are expected to lead to a non-homologous radial growth of Σ⋆ and a trend for an increasing Sérsic index with increasing galaxy mass.