Let $G$ a finite group and $H$ and $K$ two sub-groups of $G$.
Why is the map
\begin{array}{rcl}
\Psi: G/({H\cap K}) & \longrightarrow &G/H \times G/K \\
g(H\cap K) & \longmapsto & (gH, \ gK) \\
\end{array} well-defined ?
I do not understand why the fact of having $$g(H\cap K)=g'(H\cap K) \implies (gH, \ gK)=(g'H, \ g'K)$$ answers the question.
Concretely what does that kind of question mean?
NB : It's an important condition to prove Poincaré's Lemma
The elements of a quotient group $G/N$ are cosets $$ gN = \{ gn : n\in N \}. $$ Different $g$ might yield the same cosets of $N$. In fact $gN=g'N$ if and only if $g^{-1}g'\in N$: Let $g^{-1}g'\in N$, take $g'n\in g'N$, then $$g'n = g\underbrace{g^{-1}g'n}_{\in N} \in g'N,$$ so $g'N\subseteq gN$. Using the same trick we show that $gN\subseteq g'N$, so $gN=g'N$. Conversely, having $gN=g'N$, we have $g'=g'e\in g'N=gN$, so $g'=gn$ for some $n\in N$. Then $g^{-1}g'=n\in N$.
Now whenever you define a map $f:G/N\to X$ using a definition of the form $gN\mapsto \ldots$, you have to check that taking different representatives of $gN$ yields the same image, i.e. $f(gN)=f(g'N)$ whenever $gN=g'N$. In your case, \begin{align*} \Psi\colon G/({H\cap K}) & \longrightarrow G/H \times G/K \\ g(H\cap K) & \longmapsto (gH, gK), \\ \end{align*} take $g(H\cap K)=g'(H\cap K)$, so $g^{-1}g' \in H\cap K$. This means $g^{-1}g'\in H$ and $g^{-1}g' \in K$, so we have $gH=g'H$ and $gK=g'K$, so $(gH, gK)=(g'H, g'K)$ as desired.