In vector notation we will take the $j$-direction to be parallel with the axis of rotation and the $i$-direction to be perpendicular with the axis of rotation.
In the absence of the Earth's rotation,
$$F_G = N = (-\frac{Gm_1m_2}{r^2}\cos\theta)\tilde{i} + (-\frac{Gm_1m_2}{r^2}\sin\theta)\tilde{j}$$
It is easily seen that the above equation represents the perceived force of gravity in the absence of rotation. Now the centripetal force acts only in the i-direction, since it acts perpendicular to the axis of rotation.
If we let $R_{rot}$ be the radius of rotation, then the centripetal force is $m_1\omega^2R_{rot}$, which for an angle of latitude of $\theta$ corresponds to $m_1\omega^2r\cos{\theta}$
$$N = (-\frac{Gm_1m_2}{r^2} + m_1\omega^2r)\cos{\theta}\tilde{i} + (-\frac{Gm_1m_2}{r^2})\sin{\theta}\tilde{j}$$
By comparing this equation to the case shown earlier in the absence of rotation, it is apparent that as $\theta$ is increased (angle of latitude), the effect of rotation on perceived gravity becomes negligible, since the only difference lies in the $x$-component and $\cos\theta$ approaches 0 as $\theta$ approaches 90 degrees latitude. However it can also be seen that as theta approaches 0, near the equator, the $x$-component of gravity is reduced as a result of the Earth's rotation. Therefore, we can see that the magnitude of $N$ is slightly less at the equator than at the poles. The reduced apparent gravitational pull here is what gives rise to the slight bulging of the Earth at the equator, given that the Earth was not originally as rigid as it is today (see other answer).
In the case of the Earth, that minimum energy configuration is a surface over which the sum of the gravitational and centrifugal potential energies are constant. Something that makes the Earth deviate from this equipotential surface will result in an increase in this potential energy. The Earth will eventually adjust itself back into that minimum energy configuration. This equipotential surface would be an oblate spheroid were it not for density variations such as thick and light continental crust in one place, thin and dense oceanic crust in another.
In terms of force, the quantity we call g is the gradient of the gravitational and centrifugal potential energies (specifically, $\vec g = -\nabla \Phi$). Since the Earth's surface is very close to being an equipotential surface and since that surface in turn is very close to being an oblate spheroid, gravitation at the poles is necessarily slightly more than it is at the equator.
This gravitational force will not be normal to the surface at places where the surface deviates from the equipotential surface. The tangential component of the gravitational force results in places where water flows downhill and in stresses and strains in the Earth's surface. The eventual responses to these tangential forces are erosion, floods, and sometimes even earthquakes that eventually bring the Earth back to its equilibrium shape.
Update: Why is this the right picture?
Based on comments elsewhere, a number of people don't understand why energy rather than force is the right way to look at this problem, or how the second law of thermodynamics comes into play.
There are a number of different ways to state the second law of thermodynamics. One is that a system tends to a state that maximizes its entropy. For example, put two blocks at two different temperatures in contact with one another. The cooler block will get warmer and the warmer block will get cooler until both blocks are at the same temperature, thanks to the second law of thermodynamics. That uniform temperature is the state that maximizes the entropy of this two block system.
Those two blocks only have thermal energy. What about a system with non-zero mechanical energy? Friction is almost inevitably going to sap kinetic energy from the system. That friction means the system's mechanical energy will decrease until it reaches a global minimum, if any. For a rotating, dissipative, self-gravitating body, that global minimum does exist and it is a (more or less) oblate spheroid shape.