Do bubble/vacuum diagrams have some physical implication?

While calculating S-matrix elements $$langleOmega|T { phi(x_1)…phi(x_n) }|Omegarangle=frac{langle0|T Big{ phi_0(x_1)…phi_0(x_n) e^{iint d^4xmathcal{L}_{i}[phi_0]}Big}|0rangle}{langle0|T Big{e^{iint d^4xmathcal{L}_{i}[phi_0]} Big}|0rangle}$$

Bubble diagrams come from the numerator when doing wick contraction but they get canceled due to the same contributing terms with opposite sign from the denominator. Since S-matrix element (RHS) is the observables here so does it mean that bubble diagrams don’t carry any physical implication, since in LHS they cancel each other hence no contribution in something observable?

Same idea is conveyed when we’re told to use only connected diagrams while calculating S-matrix elements.