"Data" can mean two rather different things, and this poses a style problem.

1. In computing, communications and the like, data is universally thought of as stuff that has meaning only in bulk: the digital 1s and 0s that are stored on a hard disk or sent down a phone line and mean nothing individually. Used in this context, "data" is a mass noun: it takes a singular verb and can be referred to as "it". This is standard usage and no one could possibly object.
2. Many scientists still think of “data”, meaning experimental results, as a set of results whose individual members are identifiable and meaningful. This gives the word a genuinely plural character in these people's minds, and some get understandably upset when we treat "data" in this context as a mass noun, and give it a singular verb or refer to data as "it".

Making "data" singular in some contexts and plural in others is not the answer: it is likely to lead to confusion and inconsistency.

There are, however, two ways to deal with "data" in the second sense that should keep everyone happy:

• frame the copy in a way that makes it impossible to tell whether "data" is being treated as singular or plural.
• where this is not possible, replace the word "data" with an alternative, such as "readings", "measurements", "results", "findings", etc.

In direct quotes, using a plural verb is permissible if that is what the speaker or writer did, as long as it does not clash with a singular “data” nearby.