There is a $+2\sigma$ deviation in the average Higgs-signal strength for all the $7+8+13$ TeV data up to Summer 2018. We find that a slight reduction of the bottom-Yukawa coupling can fit the data better than the standard model. We suggest an extension with a vector-like quark doublet, of which the right-handed component of $b'$ mixes nonnegligibly with the standard model $b$ quark. We show that the mixing would induce a reduction of the bottom Yukawa coupling. Simultaneously, the coupling of the $Z$ boson to the right-handed $b$ quark increases, which could reduce the forward-backward asymmetry of bottom production at LEP and bring it closer to the experimental value.