The tool seems to be just calculating the Euclidean distance between the two points (the square root of the sum of the squared differences between the coordinates). This doesn't make any sense for latitudes and longitudes, which are not coordinates in a Cartesian coordinate system. Not only is this number not a meaningful distance, but it no longer contains the information required to reconstruct a distance from it, so you won't be able to calculate anything meaningful from it; you need to go back to the latitudes and longitudes themselves.
To calculate distances between points given by latitudes and longitudes precisely, you need to know which geoid was used as a reference in specifying them. But since you only want to get within 95% of the answer, you can safely assume that the Earth is a sphere.
There are two possible meanings for "the distance between two points" on a sphere. You can take the Euclidean distance between the two points (the actual points, not their latitude/longitude coordinates like your tool does), or you can take distance along the shortest curve along the surface of the Earth. Again, if you only want to get to within 95% of the answer and the distances are as small as in your example, the difference is negligble, so you can take the Euclidean distance, which is easier to calculate.
To get the Euclidean distance, you can first calculate the Cartesian coordinates of the points from their latitudes and longitudes. Denoting the latitude by $\theta$, the longitude by $\phi$ and the Earth's radius by $R$ (with $R\approx 6371 \mathrm{km}$), these are given by
$$\vec{r}=\left(\begin{array}{c}x\\y\\z\end{array}\right)
=
\left(\begin{array}{c}
R\cos\theta\cos\phi
\\
R\cos\theta\sin\phi
\\
R\sin\theta
\end{array}\right)\;.
$$
Then you get the distance between them using
$$d(\vec{r_1},\vec{r_2})=\sqrt{(x_2-x_1)^2+(y_2-y_1)^2+(z_2-z_1)^2}\;.$$
Since you seem to have small distances and aren't interested in precision, you can simplify this by expanding the trigonometric functions around one of the points, or, for greater precision, around the midpoint $\theta=(\theta_1+\theta_2)/2$, $\phi=(\phi_1+\phi_2)/2$:
$$\vec{r_2}-\vec{r_1}\approx
R\left(\begin{array}{c}
\sin\theta\cos\phi(\theta_2-\theta_1)-\cos\theta\sin\phi(\phi_2-\phi_1)
\\
\sin\theta\sin\phi(\theta_2-\theta_1)+\cos\theta\cos\phi(\phi_2-\phi_1)
\\
\cos\theta(\theta_2-\theta_1)
\end{array}\right)\;,
$$
$${}$$
$$\lvert\vec{r}_2-\vec{r}_1\rvert\approx R\sqrt{(\theta_2-\theta_1)^2 + \cos^2\theta(\phi_2-\phi_1)^2}\;.$$