I'm looking at an example on how to obtain the $\mathcal Z$-transform from a finite length truncated exponential sequence, namely: $$x[n] = \begin{cases} a^N &\text{for} & 0 \leq n \leq N-1\\[2ex] 0 & \text{otherwise} \end{cases} $$
The $\mathcal Z$-transform from that sequence is: $$ \frac {1}{z^{N-1}} \cdot \frac {z^{N} - a^N}{z - a} $$
It can be seen that there are $N-1$ poles at the origin and $N-1$ zeros at $z = a$ (because the pole in $z=a$ is cancelled by a zero). My doubt is on the location of the zeros over the $z$-plane.
I know that the $N-1$ zeros at $z = a$ can be expressed as:
$z = a\cdot1\ \text{and}\ 1 = e^{j2\pi k}$ so: $$ z_k = ae^{j2\pi k}\quad\text{with}\quad k = 1,2,\ldots,N-1 $$ Yet in the example I'm looking at, they say that the zeros are located at $$z_k = ae^{j\frac{2\pi k}{N}}\quad\text{with}\quad k = 1,2, \ldots, N-1$$ So they are spaced at $\frac{2\pi}{N}$ instead of being all at the same place and I don't understand where that $N$ came from.