Brian has suggested a great approach to begin with. Do let us know in case you have any further queries.
I've attached the two experiences we're looking to test - the control and an alternate experience. I changed the HTML of the page for the alternate experience to show "Tower Family Suite" as the top offer.
There will be times when the Tower Family Suite is not available, while the other room types are available. We do not want to show Tower Family Suite at all, when the suite is not available. When it's not available, it will not show for the control cell. But for the alternate experience, based on the HTML I create, it will continue to show.
Hopefully that helps!
Can you please elaborate your query as this seems to be a little bit tricky w.r.t. to what you're trying to achieve hence I want to be sure of the end result you're looking to achieve.
There are a few ways to approach this but this is what I would do:
1. Create an HTML offer with jQuery that looks for your "Suite" content and if it finds it, the offer replaces or adds what content you want.
2. Your denominator or visitor count will be wrong since this code will fire on pages without "Suite" content so include within the jQuery an mboxDefine/mboxUpdate with a mbox parameter to identify Exp. A or Exp B. This mboxDefine/mboxUpdate will also need to be included for Default Content.
3. Set up a monitoring campaign to run alongside your A/B test that is using that nested mbox. Apply Experience level targeting to Exp A and Exp B that maps to the mbox parameters of the nested mbox.