When referring to a sealer for stone, the word SEALER is wrong. Well, technically it is not, but the reason why I said it's wrong is because a sealer for stone is totally different from any other sealer most people are familiar with.

A sealer is perceived as a topical coating of sorts that's meant to protect the surface of the sealed object from traffic and spills, to produce a finish (polished, matte, or satin), and to fill all of the little nicks, fissures and other surface imperfections.

A sealer for stone is none of that ? None!

And that is why I said that the word sealer is wrong when referring to stone. The right word is impregnator.

An impregnator is a below the surface of the stone sort of sealer.

It's a product made of two major components:

A resin of sorts that could be silicone, siloxane, silane, ester epoxy, alphatic fluorochemicals, acrylics, etc.
A carrier that could be a petroleum-based solvent or simply water.

The resin is dissolved by and within the carrier.

What Does An Impregnator Do?
The only thing that an impregnator does is dramatically reduce the natural absorbency rate of the stone by filling in the spaces between the single minerals composing the stone. These spaces are known as pores - End of the list of performances.

This reduction in absorbency rate (or porosity) of the stone will make it possible that staining agents which get spilled on the stone will be kept at bay, on the surface of the stone, for a period of time much longer than if the stone was not sealed.

How Does It Work?
 
MB-4 impregnator/Sealer
The solution goes inside the stone
The carrier (solvent or water) evaporates.
The resin stays in and cures, thus partially clogging the pores of the stone.
The most important phase of the application of an impregnator sealer is the total and thorough removal of the entire residue of the product from the surface of the stone.

At the end of the sealing job, the surface of the stone is as bare as it was before the sealing procedure was started.

Now the question is: How does an impregnator (sealer for stone) go inside the stone?
Quite simply, the stone absorbs it.

So far we've learned a couple of important things:
A sealer for stone only helps prevent deeply embedded stains by delivering a reaction time. The reaction time is how much time you'll have to blot the staining agent off of the stones surface before it begins to sink in. (The better the quality of the impregnator in relation to the particular stone to be sealed, the longer the reaction time will be.)
Because of the way impregnators are designed and work, they cannot - and in fact do not - offer any protection whatsoever to the surface of the stone.
The Natural Absorbency Of Stone.
This side of mono-mineral rocks (i.e.: gemstones), every multi-mineral stone is somehow porous. While there are stones that absorb liquids like sponges, there are stones that are naturally so dense that no liquid is thin enough to be absorbed by them.<

The latter types of stone - there are quite a few - can't be technically sealed because no impregnator will ever stand a chance of being absorbed by the stone. Since they won't absorb any liquid, it is pretty intuitive that they will never get stained.

What is interesting to note is that while certain stones have an absorbency rate that indicates their ability to absorb liquids (above 0.2%), they in fact do not absorb anything due to their dramatically increased surface tension once polished.

For Example, travertine is rated at 0.4% to 1.0%. In its rough form it does absorb liquids, though slowly. If you polish it, it effectively will not absorb a single drop of anything. In fact, nobody ever reported any stain on a polished piece of travertine.

In conclusion, only a certain number of stones can be sealed and, more importantly, the performance of an impregnator is only limited to the reduction of the stones natural absorbency rate if it is - even when polished - above the 0.2% cut off point.

How Do I Know If I Need To Seal?
How does the average consumer know if their stone should be sealed without that kind of information? It is quite simple and down to earth:

Spill some water in a couple of spots on the stone to be tested.
Let it sit for 10 minutes or so.
Wipe it dry and observe if the areas under which the water has been sitting have become (temporarily) any darker than the rest.
If so, and if the stone is installed in an environment where staining spills are likely (i.e.: a kitchen).
If not, or if the stone is to be installed where the likelihood of spillage is minimal or nil altogether, it would be a totally useless exercise that will only help put the kids of the impregnator's maker through college.