Random points:
1. Your rake: Ouch. TBH, isolating is normally marginal enough that it is possible for that rake to consume all the profits. It'd be pretty close, I think, but I'd keep isolating and tracking my results to determine whether I was making a profit.
My general SWAG for comparing online win rates to live win rates is that win rates in live play in marginal spots or with marginal hands are going to be something like five times higher in live play than online; thus, if an online win rate for isolating limpers is 1bb/hand, I'd expect it to be around 5bb for live, and, if it wasn't in that ball park, I'd be looking for an explanation. But that is just a SWAG, and I actually have very little data supporting it (my own win rate with isolation plays being one data point supporting it, as well as my win rates in other marginal spots).
2. Isolating fundamentals. The basic premise of isolating is to get heads up in position with a person to exploit their tendency to fold too much. So, if they fold pre-flop, that's fine, and if they limp/call then play fit or fold, that's even better.
That's the basic isolation play. The more advanced isolation play is to get heads up with somebody bad and simply exploit whatever tendency they have. In other words, your described villain, someone who limp/calls and then peels wide on the flop, is a poor candidate for basicisolation plays, where you're rooting for him to check/fold to a c-bet. Against this villain, you'll need to change up th basic isolation strategy of more or less c-betting by default. Against this guy, you're going to have to have a hand or some equity.
For that reason, I'm not a fan of your iso with A3o here. Sure, it's ahead of his range, but, most of the time, what you flop is all you're ever going to have to work with. It plays like crap. Much prefer to have T9o or K5s than A3o against a villain who is going to peel wide on the flop.
3. That said, my plan for your hand would be to c-bet the pot maybe 60%, and barrel the turn on any non-heart card that improves our hand with 2 pair, trips or by improving our straight draw; something like A, 2, 3, 5 or a 7. So I'm going to barrel turn about 18 cards, check back all hearts, and usually check back most everything else. Checking back all rivers unimproved, and bluff catching river donk bets.
Isolation plays are where pot control lines really earn their money, because, most of the time, you have crap like you have here. In general, you don't want to be going crazy on these hands. Obviously you want to get max value, but max value for a crappy ace is a completely different thing than max value with AK; it is really easy here to bet your hand strongly enough that you tighten his calling range up to things that beat you. You need to be careful with that.
As a highly relevant aside: when I was coaching online, the vast majority of the people who were having trouble with making a profit on their isolation plays were having trouble that stemmed from being too aggressive. The two most common leaks were: 1. running too many multi-street bluffs just trying to blow the limp/caller off his hand, and 2. Losing bigger pots than they won by failing to pot control.
Both of those leaks were related to the thinking that "this guy limps almost anything, he can't have much here," and just bombing, without giving due consideration to the fact that the villain keeps calling.
I saw it all the time in online databases--aggressive isolators have a tendency to forget that there's no rule against them being outflopped or outdrawn by the fish they isolated. Controlling your aggression in isolation spots is the key to success.
Look at a basic isolation strategy: he limps, you raise, all others fold, he calls, flop comes down, he checks, you bet, he folds, you win the pre-flop money minus rake. Every so often, maybe a quarter of the time, maybe less, he'll call your c-bet. If you have air when he calls, you're going to lose both the preflop money and the c-bet. So a standard win is going to be maybe 6bb, and a standard loss is going to be maybe 15bb. So if we win, say, 3 out of 4, we have a win rate of .75bb/hand. The rest of our win rate is going to have to come from our non-standard wins, those where we flop a hand and he calls the c-bet (of course, we'll have some non-standard losses in that group, too). Winning 3 of 4 is actually probably too optimistic, though. It's probably closer to 2 out of 3, in which case, we have a negative win rate of -1bb/hand on the standard hands, and we need to make ALL of our profit from the non-standard hands where betting continues beyond the flop.
4. Position. The main hazard in isolating a limper is not the limper, it is the players yet to act behind you. An isolation play fails, by definition, when a limper limps, you raise, and you get called by someone yet to act.
For the basic category of isolation plays, where you're expecting to take it down with a c-bet against a limp/call, check/folder, this is an important point to remember: you were bluffing and someone called your bluff. At that point, preflop, your play has failed and you should ordinarily be expecting to give up, unless the flop somehow bails you out (by giving you a hand, equity or a great c-bet texture).
For the more advanced play, where your range is much more playable, it is less of an issue that someone else called.
Needless to say, getting called by someone who will have a positional advantage over you post flop is way worse than getting called by someone else who will be OOP in addition to the limper.
All of which tells us that we should increasingly open up our isolation range as we get closer to the button. TBH, I don't really have an iso range before the hijack, and, even on the hijack, I need the button to be a known quantity who will act predictably in some way before I'm comfortable really opening up my range to iso limpers.
So: if you are not on the button, know the button before you act. He's the most important player to have a read on before you iso a limper; it's more important to have a read on the button than it is to have a read on the limper. Ideally, he's an ABC tight player who will play his cards, period. If he's not, you have to proceed with caution.
An aggro player in the BB capable of squeezing is also a yellow light, probably a red one, for isolating limpers.
Light 3 betting players on an apparent isolation play was a thing online; it was a significant source of revenue for skilled 3 bettors. So jut be on the lookout for players behind you who might be aggro enough to exploit your exploitation of the limpers.
I may think of other stuff later.