The starting point is identifying what you are trying to achieve with the bridge. Specifically, if you don't care about weight/mass, then any very hard wood will do, including rosewoods and ebonies. If you are trying to minimize weight/mass, then less dense woods could be worth exploring along with differing designs that facilitate lighter/weight and softer woods.
The combination of material selected and design chosen must resist string wear/cutting, must prevent the saddle from leaning or distorting over the life of the instrument and must not split over the life of the instrument. Some woods just aren't sufficiently hard and strong to resist wear, prevent splitting and prevent the saddle from leaning, regardless of the design used. Balsa wood comes to mind, though I'm certain that a successful design could be created that uses balsa. Some makers/manufacturers use design features that facilitate the use of softer, lighter and less strong woods. For example, instead of putting the pin holes along a straight line, which can lead to splits between pin holes, some makers use a design where the pin holes are not in a line. Another example is leaving more wood between the saddle and the sound hole-edge of the bridge, to strengthen the saddle slot as well as using an enclosed/blind saddle slot rather than a through saddle slot. Gore details using alternate layers of carbon fibre and wood in his lighter-weight bridge design.
In short, both material and design need to be considered: one design that works well in rosewood or ebony might not work well in much softer, weaker woods, though some design changes could facilitate the successful use of those woods.
I haven't tried a bridge made of holly - I wouldn't like the aesthetic of a white bridge with a dark fingerboard - but have used padauk, walnut, rosewood, ebony, pao ferro, zebra wood and a few others. Many woods can be made to work if one chooses a complimentary design and chooses the materials carefully.
Charles