Expected Sample Size of a Sequential Drift Test

A sequential test accumulates i.i.d. log-likelihood increments $X_1,X_2,\dots$ with $E[X_i]=0.25$. The test stops at $N=\min\{n: |S_n|\ge 3\}$ where $S_n=\sum_{i=1}^n X_i$, and it is given that $E[N]<\infty$ and that the expected stopped statistic is $E[S_N]=2.0$ (reflecting that the upper boundary is hit far more often under this positive drift). Each observation costs $6 to collect. Using a Wald-style identity, find the expected total data-collection cost.