There are good answers here already, let me just share one strategy that has not been mentioned yet and that I frequently had success with in the past. It may not apply to every such email one might want to send, but if it does I believe it is the strategy that has the most success.

It is based on the observation that people will usually enjoy talking about there own work more than about some random paper you send them. Your paper may or may not be interesting to them, but their own work certainly is. A good strategy can therefore be to establish a direct connection with the recipients work or expertise. This may not always be possible, in which case I refer you to the other answers. Also in formulating such an email one has to watch out it does not come across as an obvious attempt at flattery.

So essentially my suggestion is to actually show the reader why you think that

[...] my work about A is highly relevant to your work in B, so I think it will be interesting to you too.

Maybe you are challenging their work? Maybe you have some new ideas or suggestions for improvement? Maybe you provide an alternative method? Maybe you have done an interesting experiment that connects to their theory or vice versa? Maybe you build entirely on their work and want to praise how important their work was for you? Or a combination thereof?

My trust in this strategy is based entirely on experience. I personally would at least look at the paper in case of such an email, since missing a directly related work is potentially risky.

There are good answers here already, let me just share one strategy that has not been mentioned yet and that I frequently had success with in the past. It may not apply to every such email one might want to send, but if it does I believe it is the strategy that has the most success probability.

It is based on the observation that people will usually enjoy talking about their own work more than about some random paper you send them. Your paper may or may not be interesting to them, but their own work certainly is. A good strategy can therefore be to establish a direct connection with the recipients work or expertise. This may not always be possible, in which case I refer you to the other answers. Also in formulating such an email one has to watch out it does not come across as an obvious attempt at flattery.

So essentially my suggestion is to actually show the reader why you think that

[...] my work about A is highly relevant to your work in B, so I think it will be interesting to you too.

Maybe you are challenging their work? Maybe you have some new ideas or suggestions for improvement? Maybe you provide an alternative method? Maybe you have done an interesting experiment that connects to their theory or vice versa? Maybe you build entirely on their work and want to praise how important their work was for you? Or a combination thereof?

My trust in this strategy is based entirely on experience. I personally would at least look at the paper in case of such an email, since missing a directly related work is potentially risky.