A hard bounce is a permanent delivery failure — the address doesn’t exist, the domain is dead, or the server is explicitly rejecting mail. A soft bounce is a temporary failure — the mailbox is full, the server is down, or the message was rate-limited. For Instantly users, the distinction matters because hard bounces destroy sender reputation immediately, while soft bounces that repeat consistently become just as dangerous if left unaddressed.
What a Hard Bounce Actually Means for Your Sending Domain
A hard bounce is a permanent, unrecoverable delivery failure that signals to receiving mail servers that you’re sending to unverified lists. Every hard bounce you generate chips away at your domain’s sender reputation score. Instantly’s internal deliverability thresholds treat hard bounce rates above 3% as a warning event — and Google Postmaster Tools will flag your sending domain for review once bounce signals accumulate at volume. In practice, a single campaign sent to a dirty 10,000-address list can generate enough hard bounces to get a freshly warmed inbox flagged within 48 hours. That’s not recoverable with a warmup sequence. That requires domain reputation work and sometimes a new sending infrastructure entirely.
What a Soft Bounce Actually Means — and Why Most Agencies Dismiss It Too Quickly
A soft bounce is a temporary delivery failure, typically caused by a full mailbox, a server timeout, or a greylisting event. The dangerous mistake is treating all soft bounces as benign. What we see consistently across high-volume cold email agencies is that a significant subset of soft bounces — particularly from catch-all domains — are not actually temporary. They’re catch-all servers that accept every inbound connection at the SMTP layer but silently discard or bounce the message after the session closes. These show up as soft bounces in Instantly’s reporting, but the underlying address is effectively dead. You’ll never get a read, a reply, or a conversion from it. And if you keep hitting it, you keep generating negative deliverability signals.
The Catch-All Problem No One Talks About
Catch-all domains are configured to accept mail sent to any address at that domain, regardless of whether the individual mailbox exists. Standard verification tools — including binary valid/invalid classifiers — mark these addresses as valid because the server responds positively during the verification handshake. They’re not invalid. But they’re not safe either. Industry data suggests catch-all addresses account for 20-35% of most purchased B2B lead lists, and a substantial portion of those are dead mailboxes sitting behind an accepting server. Running Instantly campaigns against unclassified catch-all addresses is one of the most common causes of creeping soft bounce accumulation that eventually tips a domain into deliverability trouble. The address never hard bounces cleanly. It just quietly drains your sender reputation over weeks.
Why Your Current Verification Tool May Be Missing This
Most first-pass verification services operate on cached database lookups. They check whether an address existed and was valid the last time they crawled it. If a mailbox was deactivated after that crawl, it still returns as valid. You load it into Instantly, hit send, and collect hard bounces on addresses that passed verification. This is the core failure mode of database-reliant verification — it has no mechanism for detecting addresses that became invalid between the last crawl and your send date. Real-time SMTP probing solves this by initiating a live connection to the receiving mail server at verification time, not six weeks ago. VerifyFlow’s second-pass verification does exactly this — classifying every address as SAFE, PROTECTED, RISKY, or DEAD based on a live probe, not a historical record. RISKY addresses include catch-all domains with low delivery confidence. DEAD addresses failed the live handshake. You decide which segments to suppress before Instantly ever touches them.
The Practical Threshold That Matters
Instantly flags campaigns when hard bounce rates exceed 3%. Google’s spam rate threshold for Gmail delivery sits below 0.10%. These two numbers are not in sync — you can stay under Instantly’s warning threshold and still be quietly destroying your Gmail deliverability. The only way to stay clean across both systems is to suppress addresses that carry bounce risk before they generate any signal, not after. Waiting for bounce data to tell you your list is dirty means the damage is already done.
Frequently Asked Questions
Q: Does Instantly automatically suppress soft bounces after they occur?
A: Instantly will flag and suppress hard bounces from future sends automatically. Soft bounces are handled differently — Instantly may retry soft-bounced addresses, which is appropriate for genuinely temporary failures but problematic if the soft bounce is coming from a catch-all address that will never deliver. Manual suppression of persistent soft bouncers is recommended after any campaign with a soft bounce rate above 2%.
Q: Can a clean ZeroBounce or NeverBounce result still produce hard bounces in Instantly?
A: Yes. Both tools rely on verification data that can be weeks or months old. If a mailbox was deactivated after their last crawl, it will still pass as valid. Live SMTP probing at send time catches these recently deactivated addresses before they enter your Instantly sequence and generate hard bounces.
Q: What bounce rate should Instantly users target to protect domain reputation?
A: Keep hard bounce rates below 1% per campaign as a working threshold — not the 3% Instantly warning level, which is a reactive floor, not a safe operating ceiling. For Gmail-heavy lists, monitor Google Postmaster Tools directly and treat any spam rate above 0.08% as an active incident requiring list review.
Before your next Instantly campaign, run your list through a second-pass SMTP verification pass and suppress every address classified as RISKY or DEAD — especially any catch-all domains you haven’t individually validated. That single step is what separates agencies that maintain clean sending infrastructure from the ones rebuilding after a deliverability incident.