Abstract: Our knowledge of $B$-meson decays to hadrons is limited, and about 40% of the total $B$ width is not known in terms of exclusive branching fractions. Therefore, the unmeasured decays are usually simulated with relevant assumptions and coarse approximations for the description of the dynamics, as in the PYTHIA fragmentation model. This limits the capability of understanding and controlling the backgrounds of many $B$-decay analyses. A large part of the Belle II experiment physics program relies on the so-called $B$-tagging, i.e. identifying the partner $B$ meson produced in association with the signal $B$ meson to infer the properties of the signal. The impact of our limited knowledge of hadronic $B$ decays on $B$-tagging and Belle II measurements in general are discussed in this seminar. The Belle II collaboration is doing a great effort to mitigate the problem, studying new high-purity hadronic $B$ decay channels. The unknown fraction of the total $B$ width is spread across multiple exclusive channels, therefore improvements are not expected from single results, but require the systematic exploration of a significant fraction of them. This effort is presented, with a particular attention to the recent $\overline B\to D^{(*)} K^- K^{(*)0}_{(S)}$ and $B^-\to D^0\rho(770)^-$ Belle II measurements.