Your code reads the first row of the table outside the loop then starts the loop with the second data line. This is probably the best solution. My only reluctance is that means the code to read the data is in the macro in two places, which always opens up the possibility of them not being identical. I always try to never duplicate code if I can help it.



The other solution that I have used is to check inside the loop if it's the first iteration, but that puts code inside the loop that is unnecessary after the first iteration, which I also don't like.

What if I change the datatype of the variable to Variant? Is there a value, like Null, that I can initialize it to that can never be a match for anything that could ever be in a Word table?

But you still need to initialise it somehow. And, of course, the whole point of the comparison is to change the highlight when they cease to be identical. Rather more complicated code could be used to do that within the loop, but the code would then run more slowly overall, which brings us to:
Nor do I.
I can't see the point of that - one could just as easily initialise it as:
StrTitle = ""
(or not even bother initialising it), but then the shading would necessarily start at the second data row (undesirable), though the code could be modified to make it start at the first data row.

__________________
Cheers,
Paul Edstein
[Fmr MS MVP - Word]

The point is, as I said, initializing it to something that cannot be in the table. That way, the compare in the first pass through the loop will always fail, which is what I need.

But even if you cannot see a point, do you know if there is a Variant value that will fail a compare with any possible text value?

That fails if the data in the first row of the table is "". I admit that that is highly unlikely, but as Murphy (or someone) said, anything that can go wrong will go wrong.