Every program we run works to the same conditions: domains aged 30 days with no sending, the full list verified, plain-text creative, a staged ramp measured against real inbox placement, and pause thresholds agreed in writing before launch. This is the whole thing, published.
The Sending Standard is what we require before we will run a program, and what we decline outright. It covers how domains are built and aged, how data is verified, how creative is structured, how volume ramps, and the thresholds that pause a campaign automatically. It exists so both sides find out early rather than three weeks in.
Why it is public
Because it is the most useful thing we know, and because you should be able to tell before a sales call whether we are going to be difficult about the things you were planning to skip. Every line here came out of programs that worked and programs that did not.
Requirements
Five conditions, before anything sends.
None of these are unusual. What is unusual is refusing to start without them.
Domains and infrastructure
Sending domains aged 30 days, with no sending
Registered, authenticated, and then left alone. A domain that appears and immediately sends volume is treated by receiving providers as exactly what it looks like.
Outbound separated from your corporate domain
Cold volume never runs on the domain your invoices and contracts go out on. If a campaign goes wrong, the damage has to be contained somewhere it does not matter.
Dedicated IPs, isolated per client
Your reputation is yours. One client's bad week cannot become another client's placement problem.
Data
The whole list verified before the first send
Not a sample. Bounces are the fastest way to lose a new domain, and an unverified file is the single most common cause of the emergencies we get called into.
A traceable source
We need to know where the data came from. Scraped files with no origin are declined, because we cannot assess a risk we cannot see.
Suppression applied permanently
Opt-outs, complaints and hard bounces are suppressed across the account, not per campaign.
Creative
Plain text, personalized
No image-heavy templates at cold volume. Tracking pixels and heavy link structures are what filters are built to catch, and at cold volume they cost more than the data they produce.
Links kept minimal
Ideally none in the first touch, or a single calendar link where booking is the actual call to action.
A sequence, not a blast
Initial, follow-up and break-up. A single send to a whole list is the shape that fails most reliably.
Ramp
First sends to 5 to 10% of the list
Small enough that a bad signal is recoverable, large enough to be a real measurement.
Volume climbs over four to six weeks
And only as measured inbox placement allows it. The schedule is a plan, not a commitment: placement decides.
Seed testing before the first live send
A documented placement baseline, per provider, so later movement means something.
The fifth condition is the one people notice: we keep the right to pause your campaign, and we use it.
The aging rule
Thirty days, not fourteen.
The 30-day aging period is the one requirement people push back on hardest, and it is the one we will not move on. It is also the closest thing we have to a proven variable.
Two of our managed campaigns succeeded on domains that sat unused for a month because the client was slow getting approvals. That was not luck, and it was not a nice-to-have that happened to be present. Across the accounts we have run, it is the strongest predictor of whether a program works.
On the number itself
Fourteen days is the absolute minimum if you are doing this yourself and cannot wait, and it is what we will tell you in a consultation when the alternative is sending on a domain registered yesterday. We use 30 because it measurably outperforms. When we control the build, we build to what works rather than to what is tolerable.
The practical consequence is that a build takes four to six weeks before the first send, and that length is set by the aging period rather than by us. It is not dead time — the data, the authentication and the sequences are all prepared while the clock runs — but the first send cannot be pulled forward. If that does not work for your deadline, the honest answer is that outbound is not the right channel for it, and we would rather say so now.
Pause thresholds
Kill criteria, agreed before launch.
Named numbers, set in writing before the first send. Crossing one pauses the campaign automatically rather than starting a conversation.
1
Complaint rate
Google's published bulk-sender threshold is 0.3%, and by the time you are near it the damage is already done. Your program's own ceiling is set below that and agreed before launch.
2
Bounce rate
A rising bounce rate on a verified list means the data has aged or the segmentation is wrong. Either way it is a stop-and-look, not a push-through.
3
Blocklist hit
Any listing on a blocklist that matters stops outbound on that domain immediately. We repair first and resume second, in that order, every time.
Nobody argues with a threshold they signed off on. That is precisely why it is set in advance rather than in the middle of a bad week.
Both sides
What we decline, and what you get for it.
What we won't take on
Send-it-all-next-week timelines
Scraped data, or lists you can't tell us the source of
Single-blast campaigns with no follow-through
Image-heavy, link-heavy creative at cold volume
Lists we've assessed as unsalvageable — we'll say so plainly
Anyone who wants a guaranteed number of leads
What you get in return
Inbox placement verified by seed testing before your first live send
Infrastructure properly built, aged and warmed
Reputation monitoring, blocklist handling and delisting
Disciplined ramp management and campaign execution
A named person who knows your account
Being told early and honestly when something isn't working
If we can't take your project, we'll tell you exactly what would need to change — and you're always welcome to run it yourself on the platform.
Questions
The Sending Standard, answered.
Why do you require 30 days of domain aging?
Because it is the strongest predictor we have. Two of our managed campaigns succeeded on domains that happened to sit unused for a month while the client sorted out approvals, and those accounts outperformed the ones that launched faster. A brand new domain has no history, and receiving providers treat a domain that appears and immediately sends volume as exactly what it looks like. Thirty days of no sending is the cheapest reputation you will ever buy.
Is 14 days of aging ever enough?
Fourteen days is the absolute floor if you are doing this yourself and cannot wait, and it is what we tell people in a consultation when the alternative is sending on a domain registered yesterday. It is not what we do. When we control the build we use 30, because it measurably outperforms and because the cost of getting it wrong is a burnt domain and another three months.
What happens if we do not want to follow the Standard?
Then we will not take the work, and we will tell you why rather than take a deposit and hope. You are welcome to run the campaign yourself on the platform. The Standard exists because our clients share infrastructure with each other and because a campaign that fails costs you far more than the one we declined.
Is the Standard part of the contract, or just a marketing page?
It goes in the agreement for managed programs and partner accounts, not only on this page. The aging period, verification requirement, creative rules, ramp schedule and our right to pause are agreed at signing, so they are never a mid-campaign argument.
Who decides when a campaign gets paused?
The thresholds do. Complaint rate, bounce rate and blocklist status are agreed in writing before launch, and crossing one pauses the send automatically rather than starting a discussion. Nobody argues with a number they signed off on, which is the whole reason it is set in advance.
Does the Standard apply to self-serve platform accounts?
Not as a contractual requirement. If you buy a plan and run your own campaigns, the platform does not stop you sending on a domain you registered this morning. The Standard is what we require when our team is doing the work and our reputation is on the line, and it is published here because it is also the best advice we have for anyone running their own program.
Do you make exceptions?
On sequencing, sometimes. On the aging period, verification and the ramp, no. Those three are what the results are actually built on, and an exception on any of them moves the risk onto every other client sharing the infrastructure.
Get started
Tell us what you want to send.
Volume, data source and timeline. We will tell you whether it clears the Standard, what would need to change if it doesn't, and what the work would cost.
An honest answer on whether your timeline is realistic
What we would require before taking the work on
Free either way, and useful whether or not you become a customer
