You match the number of electrons with the expression for $K$ itself, which in turn depends on the coefficients you use to balance the equation. If you do everything right you get the same algebraic relation that reflects the correct thermodynamics.
Thus for hypobromite disproportionation we may have
$\ce{(3/2)BrO^- <=> (1/2)BrO3^- + Br^-}$
with two electrons transferred due to one bromine atom being reduced from $+1$ to $-1$ oxidation state. Therefore,
$K=\color{blue}{\dfrac{[\ce{BrO3^-}]^{1/2}[\ce{Br^-}]}{[\ce{BrO^-}]^{3/2}}=10^{2E/(0.059\text{ V})}}$
Or, we can render it as
$\ce{3BrO^- <=> BrO3^- + 2Br^-}$
with four electrons transferred due to two bromine atoms being reduced from $+1$ to $-1$ oxidation state. Therefore,
$K=\color{blue}{\dfrac{[\ce{BrO3^-}][\ce{Br^-}]^2}{[\ce{BrO^-}]^3}=10^{4E/(0.059\text{ V})}}$
The two formulations will give different $K$ values. But the blue parts of the equations are equivalent; the second equation is just the first one with both sides squared. So if you properly formulate $K$ based on the coefficients along with the electron count, you get the same empirically accessible relationships between the reactant and product concentrations/activities.