When you search for a Greek or Hebrew word in Accordance, Accordance ignores any breathing marks, accents, or vowel points. We do this because we don't want a search to fail just because you entered a rough breathing mark instead of a smooth one, or a sere instead of a seghol.
To see how this works, open the tagged Greek New Testament (GNT-T) and do a search for εις. When the search is complete, choose Analysis from the Details pop-up menu.
An Analysis of your search will open in a new zone and list every lexical form that was found by your search. Note how it finds εἷς (with a rough breathing mark) and εἰς (with a smooth breathing mark). Accordance doesn't require those kinds of characters to be specified for a search to be successful.
Now, suppose you only want to find the lexical form εἷς and do not want to find εἰς? As I mentioned earlier, Accordance ignores vowel points and breathing marks, so even if you enter εἷς and re-run the search, you'll still find both εἷς and εἰς. Try it yourself to verify this behavior. Copy εἷς from the Analysis window and paste it into the argument entry box of your Search window. When you click the Find button, you'll notice that the number of hits remains the same and the Analysis window still lists both εἷς and εἰς.
So what do you do when you want Accordance to consider accents, breathing marks, and vowel points? How do you get it to stop ignoring them? You do it simply by placing an equals sign (=) in front of the word you're searching for. The equals sign tells Accordance to do an exact search for the word as you've entered it. This means of course that you have to get all those little marks right for the search to be successful. Try putting an equals sign in front of εἷς and running the search again. Now you'll see that all occurrences of εἰς are excluded.
Most of the time, you won't need to worry about accents, breathing marks, and vowel points; but on those occasions when you do want them to be considered, just remember the equals sign.