Temptation Bundling: Glue the 'Should' to the 'Want'

The short answer: glue the boring thing to the thing you actually want

Behavioral scientist Katy Milkman ran the foundational experiment on this in 2014 at Wharton (source). Her team gave subjects an option to listen to addictive audiobooks (Hunger Games, Twilight) only at the gym. Compared with controls, the bundled group attended the gym 51% more often. Milkman named the technique temptation bundling — gluing a should-do (gym, taxes, cleaning) to a want-to-do (podcast, audiobook, favourite coffee, comedy show). The effect is large and robust because it converts an internal motivational struggle into an external choice architecture problem the brain finds much easier.

Why "just do it because it's good for you" doesn't work

Most should-do tasks are routinely outperformed by want-to-do alternatives because the want-to-do has immediate reward and the should-do has delayed reward. The brain doesn't argue about this; it just goes where the dopamine is. Bundling moves the immediate reward inside the should-do — the audiobook is the dopamine, the gym is the carrier. The should-do still gets done, but for the right brain reasons. This is not a discipline replacement; it's a discipline substitute that doesn't burn willpower because none was required.

Conditions for a bundle to actually work

• The want has to be genuinely wanted. Lukewarm wants don't pull the should-do behind them. A great podcast you've been waiting to start; the comedy show you'd actually pay for; the audiobook you've been recommended five times — these work. "Educational podcasts" you think you should listen to don't, because they're another should-do you're stacking. The want has to be hedonic, not virtuous.

• Restrict the want to the should context. If you can listen to the audiobook anywhere, the bundle dissolves — you'll listen on the sofa instead. Milkman's strongest condition was the restricted-bundling group who could only listen at the gym. Pick a want that you'd be willing to wall off and use only in the should-do context. The walling-off is the technique, not optional decoration.

• Match the want to the cognitive load of the should. Bundling a podcast with treadmill running works (low-load physical). Bundling a podcast with deep-focus coding doesn't (high-load mental). For high-load shoulds, pair with low-load wants like a great coffee, a candle, a specific playlist — sensory bundles rather than attention-competing ones.

• Stack with a stable cue, not a wish. "Audiobook only at gym" works because the gym is a discrete event. "Audiobook only when I'm being productive" doesn't, because productive is internal and slippery. The should has to have a clean external boundary so the bundle has somewhere to fire.

Six concrete bundles

• Premium podcast or audiobook only during exercise.

• Favourite coffee only at the desk while doing the dreaded admin task.

• Watching the show you love only while doing the laundry / folding.

• Comfort meal only on the night you process invoices or do taxes.

• Walk in the favourite park only while making the hard phone call.

• Specific music playlist only during the weekly review or planning session.

Why this pays double for ADHD

ADHD-brain motivation is interest-based rather than importance-based. The standard productivity model — "do the important thing first" — runs into this hard. Temptation bundling sidesteps the problem by making the important thing also interesting through the bundle. The brain doesn't have to choose between dopamine and obligation; the bundle provides both. This is one of the few productivity techniques that gets stronger with ADHD attention rather than weaker, because it works on the brain's natural reward system instead of fighting it.

Where it fails (and the repair)

• Leaking the want outside the bundle. The most common failure: you listen to the audiobook on the sofa one evening because "just this once". The bundle weakens; within a fortnight it's gone. The fix is to actually wall off the want; if you can't, pick a different want that you can.

• Bundling with a marginal should. If the should is something you'd just skip altogether, the bundle won't fix it — the want will just go unused too. Bundling works on shoulds you'd reluctantly do; it doesn't manufacture motivation for genuinely-not-going-to-happen tasks.

• Habituation. If you bundle the same audiobook with the same gym session for three months, the want becomes ordinary and the bundle weakens. Rotate the want — new podcast, new audiobook, new playlist — every few months to keep the pull fresh. Bundling is a regenerable rather than permanent technique.

FAQ

Isn't this just bribing myself?

Yes — and the data says it works much better than not bribing yourself. The cultural moral overlay against self-rewards mostly hurts the person carrying it; behavioural research is consistent that engineered rewards in the right place reliably move behaviour. There's no virtue badge for finishing the task through pure suffering when a bundle would have done the same work cheaper.

What if I don't have any "want" strong enough to bundle?

Look at where your discretionary time actually goes when you let yourself drift. The activities that pull you in for an hour without effort are your candidate wants. If you genuinely can't find one, the issue might be upstream — burnout, depression, or a missing dopamine source the bundle can't manufacture. In that case the technique alone isn't the right tool.

Can I bundle the want with multiple shoulds?

Carefully. The want needs to stay distinctive — if you allow it during three different shoulds, the bundle effect dilutes proportionally. Better to have one strong bundle than three weak ones. If you want multiple bundles, pair multiple wants with multiple shoulds, one-to-one.

What if I miss the should and want to use the want anyway?

Don't. The whole technique depends on the want being tied to the should. The first time you let yourself break the bundle is the day the bundle starts to lose force. Treat it like a contract with yourself; the cost of breaking it is the technique's effectiveness, not just the moment.

How long does a bundle stay effective?

Variable — usually two to six months before habituation softens it. The fix is rotation: new audiobook, new show, new bundle. The underlying technique can run indefinitely; the specific bundles refresh periodically. Don't expect any single bundle to be a forever fix; expect the rolling system to be.