'Pollination syndromes' is the name given to the various ways in which flowering plants have evolved 'strategies' to disperse their pollen - e.g. by insects, wind, etc.

With regard to insect pollination, it was noted as far back as the time of Aristotle that some plants are associated with 'specialist' insects which perform their pollination.

From the plant's 'point of view' it's clearly beneficial if a 'specialist' insect visits its flowers, because the pollen - which is species specific - will be carried to another plant in the same species.

It's unclear however, what advantages this 'syndrome' may have for the insect - which is interested in feeding rather than assisting pollination.

Why should animals specialize on a plant species, rather than move to the next flower of any species? Although pollinator constancy was recognized by Aristotle, the benefits to animals are not yet fully understood.The most common hypothesis is that pollinators must learn to handle particular types of flowers, and they have limited capacity to learn different types. They can only efficiently gather rewards from one type of flower.
[…]
Although a recent review concluded that there is "overwhelming evidence that functional groups exert different selection pressures on floral traits" the sheer complexity and subtlety of plant-pollinator interactions (and the growing recognition that non-pollinating organisms such as seed predators can affect the evolution of flower traits) means that this debate is likely to continue for some time.
Source : Wikipedia

Due to its puzzling nature, the possible advantages and disadvantages of these types of pollination syndromes are still being debated.

There are several hypotheses regarding why insect specialisation exists - including 'Memory Limitation' and 'Learned Satiation' etc - see Wikipedia

Note: Recent studies are suggesting that plant specialisation is less prevalent than was once thought. Currently estimated at around 30%.

Further reading : "Generalization in pollination systems, and why it matters" Ecology, 77 (4): 1043–1060 (archived).