Klaviyo does not bill you for the emails you send or the people who read them. It bills you for active profiles: everyone in the account who is not suppressed. People who unsubscribed years ago on a different platform and got imported anyway. People who bought once in 2023 and never opened again. People with no email address at all, collected by an SMS form. If they are in the account and not suppressed, you pay for them every month.
Most store owners find this out the day the bill jumps, because the count only moves in one direction on its own. Every pop-up signup, every checkout, every import adds profiles. Nothing removes them unless you do it deliberately.
Why does the bill jump instead of creeping?
Pricing is tiered. You do not pay per profile, you pay for the band your count sits in, and crossing into the next band moves the whole bill at once. Klaviyo also upgrades you automatically when you cross — there is no warning email that asks first. A store hovering near the top of its band can be pushed over by one good month of pop-up signups, and the new price arrives before anyone notices the count moved.
At 50,000 profiles this is real money, and the part that stings is what the count is made of. On accounts we have taken over, the split is consistent: a third of the billable list has not opened, clicked or bought anything in six months. We watched one account send 64,342 emails to an audience like that across two re-engagement flows. Revenue from it: zero. Unsubscribes: 635. Those profiles were being paid for twice — once in the bill, once in the damage to sender reputation.
Which profiles can go without losing sales?
Suppressing a profile removes it from billing without deleting the data. The person can still buy from you, still rejoin the list, still trigger flows if they come back. So the question is not "who do we delete" — it is "who do we stop paying for while they ignore us." The safe cleaning pass, in order:
- No email address. SMS-only or empty profiles that your email plan is counting.
- Hard bounces and repeated soft bounces. They cannot receive anything; keeping them
also hurts deliverability.
- Never-engaged imports. Added more than 90 days ago, zero opens, zero clicks, zero
orders, ever. These are the imported dead weight.
- Long-lapsed with no orders. No engagement in 12 months and no purchase history.
Run them through a sunset sequence first if you want a last attempt, then suppress what stays silent.
One caution before you build these segments: opens are inflated by Apple's Mail Privacy Protection, so "opened recently" keeps profiles looking alive that have not seen an email in a year. Anchor the definition on clicks and orders, and treat opens as a bonus signal, not the spine.
How much does this actually save?
Run the numbers on your own account before touching anything: the count is in Account → Billing, and a segment of "no click, no order, added over 90 days ago" tells you what share of it is dead weight. If that segment is a third of your list and you are near a tier edge, the cleaning pass usually drops you a full band.
The bill is one half of the return. The other half is that every number in your account improves when the dead weight stops receiving sends — which changes how the rest of your email revenue performs, and is the reason we treat the cleaning pass as revenue work, not admin. If nobody in your operation owns this pass, that is a gap worth closing before the next tier does it for you.
