They would like to keep them on the site longer and have them explore their products more thoroughly in case they miss what they were looking for initially.
The missing information is so much, it begs a “reading between the lines” answer.
My nose tells me the missing part of the paragraph was “…And is there a quick fix slap-dash gimmick we can just throw up there without all this ‘user testing stuff.’ Our site is just the way we want it, the only problem is that’s not how our users want it.”
There are certain things which reduce bouce rate. There are certain things which reduce exit rate. With some overlap, which to use differ for each.
Barring any real information, my guess is the whole site was built around buying – not shopping behavior. People get to the site, get exactly what they are looking for (or not), then leave.
Congratulations, your clients have a standard, general, that’s-only-way-we-know-how-to-build-'em web site. A site which soothes the programmer’s zero or one binary sensibilities. And that’s the problem.
You may not have built it. But you’re charged with fixing it.
Your client wants people to shop. The site only supports buying. In other words, it’s set up like a vending machine. And you’re all looking for what to write on a Post-It to slap up on the vending machine to cure it from being treated as a vending machine by the users your client doesn’t want to interact with. (Hence a zero information forum post rather than a conclusive user test)
In other words, you’re looking for cough syrup to cure emphysema, my nose tells me.
One, understand why this is a user test problem. Not an “answer from the back of the book” problem. Suggestions on which widget to drop into a site may not help. Certainly not based on what little there is to go on here.
Next, its a problem of what actually gets executed. Were a “nag” screen helpful, the same mindset that forged the current bounce/exit rate will result in a fail. Even if you’re not the dev for the site, my guess is you’ll have to pry control out of the client’s cold, dead, hands.
You’re at the next stage from tacking on a cart – visual merchandising design versus PHP programing. Understanding users rather than nagging them.
I mean, seriously, “should I put up a NAG screen” doesn’t send up foreshadowning red flags? I don’t care if this is the best thing you could ever do, what meager, pitiful scraps I have to go on bodes ill for the way it gets accomplished.
This is like guessing how many jelly beans in a jar …when you aren’t shown the jar, and not told what’s in it – just to guess. My guess will be 12,354 jelly beans or a REALIGN (not a redesign).
When (if) it is a mind set problem, a redesign won’t help. You’d get the same site with more modern look and gimmickry. You have to do a realign, from the client-as-customer to the client’s actual real life customers.