Something which might be painfully obvious to some, but which never seems to be discussed explicitly is that there are two general types of sound change.
The first one is 'oral' sound change, the other is 'aural' sound change. This terminology is pretty horrendous due to the very similar sounding words oral and aural, but bear with me, this is text, surely you can figure out which one of the two is meant as long as I don't mix them up ;-)
Let's start off with the oral sound change. Oral sound changes are the type of sound changes that we would often call 'lenition', they ease the pronunciation for the speaker. A good example of this would be the intervocalic voicing (and subsequent fricativisation) of voiceless stops in Romance languages.
Lat. lupus 'wolf' Spanish lobo 'id.'
This has an increased ease of speaking as the vocal chords of the speaker do not have to stop vibrating to pronounce the consonant in between the vowels.
This doesn't mean all oral sound changes are lenitions. for example the voicing of Latin s to z which then became r in intervocalic position is hard to see as a lenition. the s > z shift is, but z > r is rather a fortification.
Aural sound change is different. This has nothing to do with ease of speaking. It rather replaces one sound with another because it is very similar in sound. A great example of this is London English where we find that [θ] has shifted to [f].
English think [ˈθɪŋk] London English [ˈfɪŋk]
Although these two sounds are absolutely phonetically distinct, they are very similar. While the articulation is nothing alike, the only thing these two words have in common is the fact that they're voiceless fricatives. But going from an interdental fricative, which involves putting your tongue between your teeth, to a labio-dental fricative which doesn't involve your tongue at all, but rather your lower lip and upper teeth is not obvious.
So now we have made this distinction between oral and aural soundchange, how does this help us?
Well, for one, it may help to get insight on the development. It seems obvious that an aural sound change is much more likely to occur due to intensive contact with non-native speakers unfamiliar with a certain sound and thus replacing it with another sound, similar phonetically but not necessarily articulatorily.
This does not mean only aural sound changes can happen due to intensive language contact. If a language still has an active soundlaw that says 'voice all intervocalic voiceless stops', such a oral sound change might rub off onto the contact language after all.
On the other hand, maybe we could hypothesise that oral sound change is likely to take place more often in fairly isolated communities with little contact. If a language is not spoken with non-native speakers there is less reason to be absolutely meticulous about the pronunciation of the language to be understood. This is comparable to the fact that casual speech often has far more oral sond changes than does formal speech. For example in Dutch you can reduce every single unaccented vowel in a word to [ə].
Formal Dutch fonologie [ˌfo.no.lo.ˈχi] 'phonology' Informal Dutch [ˌfo.nə.lə.ˈχi]
For more familiar and more native communication, more reduction is possible while retaining mutual understanding.
Of course things are never so clear cut, and surely you can come up with examples where it's not very clear to which of the two forms it belongs. But I thought I'd just throw this out there.