# Why 100% double opt in beats a bigger list

> Consent is filed under compliance and it belongs under deliverability. A smaller confirmed list reaches more inboxes than a larger unconfirmed one, every time.

Source: https://hiipe.io/blog/why-100-percent-opt-in-beats-a-bigger-list
Published: 2026-11-17

Consent gets filed under compliance, discussed with a lawyer, and then ignored by the people who actually send. That is the wrong filing. It belongs under deliverability, because a smaller confirmed list reliably puts more messages in front of more humans than a larger unconfirmed one.

**Key takeaways**

- The bigger list often delivers fewer messages, because reputation is scored per sender.
- Confirmation removes exactly the addresses that would have damaged the send.
- Watch one number: your reported spam complaint rate.

## The arithmetic, on your own numbers

Take your list and split it in two: addresses that confirmed, and addresses that
were added some other way. Then run the same send to both and compare not opens
but **messages that reached an inbox rather than a spam folder**.

The pattern is consistent enough to plan around. Here is the shape, as an
illustration rather than a measurement, because the real figures are yours:

| | 10,000 unconfirmed | 4,000 confirmed |
|---|---|---|
| Invalid or abandoned addresses | A meaningful share | Near zero, they clicked to confirm |
| Complaint rate | Higher, from people who forgot signing up | Very low |
| Effect on sender reputation | Downward, cumulatively | Neutral to positive |
| Messages reaching an inbox | A fraction of the list | Most of the list |
| Messages reaching a human | Fewer than the smaller list | More |

The last row is the whole article. It is not an argument about ethics, though the
ethics also point the same way. It is that reputation is scored per sender, and a
bad send does not just underperform, it makes the next send worse.

## Why mailbox providers behave like this

Because they have one job, which is to keep their users' inboxes usable, and they
do it by scoring senders. Volume, complaint rate, bounce rate, engagement,
authentication. [Google's published sender
guidelines](https://support.google.com/a/answer/81126) ask bulk senders to keep
the reported spam rate below 0.10 percent, never to reach 0.30 percent, and to authenticate properly, and other
providers apply comparable thresholds.

Notice what that threshold means at scale. Three complaints in a thousand is not
many angry people. It is the ordinary rate at which humans forget having signed
up for something eight months ago, which is precisely what confirmation, and
regular sending, prevent.

## What full opt-in actually requires

Four things, and only the first is the checkbox everyone thinks of.

**An affirmative act.** They do something, deliberately. Not a pre-ticked box, not
a condition of entry, not a side effect of buying a ticket.

**A confirmation step.** A message to the address, and an action on it. This is
what makes the address real rather than typed.

**A stored record.** The wording shown, the timestamp, the surface. Without the
record you have a belief, and a belief is not a defence.

**A visible way out.** One click, honoured immediately. Making unsubscribe hard
converts an unsubscribe into a complaint, which is the single most expensive
substitution in email.

## The three real costs, stated plainly

**The list gets smaller.**  This is uncomfortable to report
internally and it is the correct outcome, because the removed addresses were not
producing anything.

**Growth looks slower.** Any metric that counts additions will worsen. Change the
metric to confirmed additions, once, and stop reporting the other one.

**There is more friction at capture.** A confirmation step loses some people who
would have engaged. That is a real loss and it is smaller than the reputation
damage the same people cause when they arrive unconfirmed and forget.

## What it buys back

Sends that land. Segments that behave predictably. The ability to go quiet for
three months and come back without being filtered, which for a seasonal business
is not a nicety, it is the entire year, as set out in [festival marketing happens
in the eleven months without a festival](/blog/festival-marketing-between-editions).

And a list you can defend. Someone will eventually ask why they are receiving
this, and the answer "on this date, on this page, you clicked this button that
said this sentence" ends the conversation, which is the same argument as [a guest
list is not a contact list](/blog/a-guest-list-is-not-a-contact-list).

## The re-permission send, and when it is worth doing

Every team eventually inherits a list nobody can vouch for. There is one honest
way to deal with it and several tempting ones that are worse.

| The approach | What happens | Verdict |
|---|---|---|
| Mail it all and see | High complaint rate on the first send, reputation damage that outlasts the list | Never |
| Mail the recent slice, watch, expand | Slow, safe, and you learn the shape of the list | The one to use |
| Ask everyone to re-confirm | Small response, and the responders are worth more than the whole file was | Good, if you can accept the number |
| Quietly keep it and never send | No risk, no value, and it still counts on a per-contact bill | The most common, and the most expensive |

The third row surprises people. A re-permission send typically returns a fraction
of the file, and that fraction outperforms the original list on every metric that
matters, because it is composed entirely of people who chose twice.

## The one number

Reported spam complaint rate, checked at every send. Under 0.1 percent is
healthy. Approaching 0.3 percent is a list quality problem that no subject line
will solve, and the fix is always the same: confirm, segment, and send to the
people who asked.

## Where HIIPE fits

The reason this is hard in practice is that consent usually lives in a different
system from the sending, so the record and the send drift apart. In HIIPE the
consent is a field on the person in the fanbase, written at capture with its
wording and timestamp, and it stays with the contact, in the fanbase and in every export, so the proof travels with the person you write to. [How the pieces
connect](/solutions) is the short version.

**Read next** : [A guest list app fills the room. It does not build a contact list](/blog/a-guest-list-is-not-a-contact-list) · [The receipt is the part nobody owns](/blog/first-party-fan-data-and-the-receipt)
## Citable facts

- Google's published sender guidelines ask bulk senders to keep the spam rate reported in Postmaster Tools below 0.10 percent and to avoid ever reaching 0.30 percent.
  Source: Google, Email sender guidelines
- A bigger list lowers deliverability when part of it never asked to be there. Complaints are counted against the sending domain, not the campaign, so one bad import damages every message that follows.
  Source: HIIPE analysis

## FAQ

### What is double opt in?
The person gives you an address, and then confirms it by acting on a message sent to that address. The second step proves the address exists, that they control it, and that the interest was real, all of which the first step only assumes.

### Does double opt in shrink your list?
Yes, usually by a noticeable margin, and that is not the cost people think it is. The addresses it removes are the ones that were never going to open anything, plus the ones that would have complained.

### Is this a legal requirement?
In some jurisdictions and for some cases, confirmation is expected practice rather than strict law. But the argument here is not legal at all. It is that mailbox providers reward it and the arithmetic is on its side.

### What single number should I watch?
Your reported spam complaint rate. Google's sender guidelines ask bulk senders to keep it below 0.10 percent and never to reach 0.30 percent. If yours is anywhere near that, list quality is the cause and no subject line will fix it.

### Is single opt-in ever acceptable?
It is common and in many jurisdictions lawful. The argument here is not legal, it is that confirmation removes exactly the addresses that damage a send, so single opt-in trades a larger list for a worse one. If you use it, watch the complaint rate weekly rather than monthly.

### What is a good open rate?
Whatever yours was last month, moving up. Industry averages mix newsletters, receipts and promotions across every sector, so they tell you nothing about a venue mailing four hundred people about a Tuesday. Compare yourself to yourself.

### How do I clean an old list I inherited?
Do not mail it to find out who is alive: that is exactly the send that damages your reputation. Send once to the most recent slice only, watch the complaint rate, and expand only if it stays low. Anything older than two years should be treated as history rather than audience.
