Account-based marketing agency
We build the account list, the research behind every name on it, the sequence and the emails, and we run them from your own domain.
Fifty to two hundred named accounts you can defend one by one.
What changed at the account, and why the timing is now.
Written for one company at a time, sent from your own domain.
The work
Every one of these is a template we publish free, in the editor, with the HTML yours to take.
Four things changed underneath B2B outbound, and all four are a matter of public record. Every claim below links to where we read it, so you can disagree with the source rather than with us.
5,000
messages a day is where the gates close
Google, Yahoo and Microsoft now reject mail that fails authentication rather than filtering it into a spam folder where somebody might still find it. Any domain sending five thousand or more messages a day has to pass SPF, DKIM and DMARC, carry a one-click unsubscribe, and hold spam complaints under 0.3%, with 0.1% the number to actually work to.
Sources: powerdmarc.com and redsift.com.
89%
inbox placement, for senders who comply
Senders who meet those requirements average about 89% inbox placement. Senders who do not see roughly a fifth to a third of their mail go to spam. That gap is not a copywriting problem, and no amount of subject line testing closes it.
Sources: powerdmarc.com and redsift.com.
2 weeks
is how long LinkedIn banned an AI SDR for
The category that promised to automate all of this ran into trouble in public. 11x has been reported to have heavy churn and disputed customer claims, LinkedIn banned Artisan for about two weeks, and Artisan’s own chief executive said he would “cringe in pain” at the emails his early product wrote. The tools sold volume into the one channel that had just started punishing volume.
Sources: salesmotion.io and refontelearning.com.
The list
is the biggest lever, ahead of the agency
The biggest lever in an account-based program is the account list itself, ahead of the agency running it and ahead of whatever software it is run on. The research, the sequence and the spend are all downstream of who is on the list, which is why the first weeks of any engagement go there and nowhere else.
Source: salesmotion.io.
Four things. Most engagements start with one of them and add a second once the first is working. What each one leaves out is written down here too, because that is the part people find out later.
Named-account lists, research on each account, the sequence, and the emails themselves.
Copy, design, sending setup and the follow-up cadence, run from your own domain.
Designed once, built to render everywhere, handed over as HTML you own.
ESP setup, deliverability basics, and reporting that says what actually happened.
They are not three sizes of one idea. The work, the cost per account and the honest ceiling are different in each. In all three drawings an orange mark is an account and a white mark is something we make, so the first and the last are the same shape pointing opposite ways.
Custom material for a single account, and marketing and sales contacting it in step. Everything on the right of the drawing is made for the one mark on the left.
Expensive per account, and worth it only where one deal pays for the quarter.
Five to fifteen accounts that share a situation. One research spine and one argument, with the evidence swapped per account. This is where most of what we do sits.
The research carries across the cluster, so the cost per account falls without the message going generic.
A segment treated as a segment. One thing made, then sent to everybody in the tier. It is the mirror of the first drawing, with the arrows pointing the other way.
Useful for warming the outer ring of a list, and the tier most likely to be demand generation wearing an account-based label.
This is the whole of it for a single name on the list. Six things get made, in this order, and the last one is the only part the account ever sees.
Five to eight tests one person can check from outside the company in about five minutes. Written before the list, because a rule invented to justify an account already on it is not a rule.
Named accounts, each with a one-sentence reason and a date beside it, in a form somebody else can argue with. It is what makes the monthly review possible and what survives whoever leaves.
Every claim that will appear in the email, with the link beside it. Anything that cannot be opened does not get written.
Not the company. The people inside it who can say yes, and the ones who can say no, which is usually the longer list.
Written for one company, sent from your domain, with a real person's name on it. It opens with something true about them rather than about us.
Sales reads it before it goes. Accounts come off the list at the monthly review as well as go on.
Real, rendered here, and plain on purpose: an account-based first touch is not a designed template, it is a short note about that company’s own problem.
The merge tags mark where the researched detail goes, so the first paragraph cannot be skipped. Open it.
There is no discovery phase with a deck at the end of it. The call is a call, the plan is a document, and the first month builds something.
Day one
A short call about who you sell to, what has been tried, and what a good quarter would look like.
Week one
Within a week you get a written plan: the accounts, the sequence, the emails, the timeline and the cost. No retainer until you have read it.
Month one
The first month builds the list and the emails. Nothing is sent until you have approved every message.
An agency that fits everybody fits nobody. There are four situations where hiring us is the wrong call, and they are common enough that we would rather put them on the home page than discover them together on the second call.
Below about twenty million a year, recurring or not, one person in house who owns the list and does the writing usually beats a retainer. They keep learning your market between campaigns, which is the part an agency has to relearn every quarter. Hire them first and talk to us when the program has outgrown one person.
Sources: salesmotion.io and npws.net.
Pilots run about five to fifteen thousand dollars a month and full programs fifteen to forty thousand. That fee usually sits on top of the ad spend and the platform licences rather than including them. If the maths only balances by assuming a result, it is not a budget, and we would rather say so now than three invoices in.
Sources: salesmotion.io and npws.net.
A fuzzy list loses money whoever runs it, because everything after it, the research, the writing and the spend, gets applied to the wrong companies at full price. Cut the list to the accounts you can argue for one by one. Some teams do that and find they no longer need an agency.
Source: salesmotion.io.
If the plan is to reach as many people as possible for as little as possible, we are the wrong shop, and we would do that job worse than the tools built for it. Our whole method is expensive per account and only pays back when the account is worth it.
The names are agreed and sales is bought in, and there is still no one whose actual job this is on a Tuesday afternoon.
The research has to start from nothing, and your team has no closed-won pattern in that market to copy yet.
Mail is going out, the domain is shaky, the emails look assembled rather than written, and replies have gone quiet.
Below is the real editor with a real template loaded, not a screenshot of one. Change anything in it, export the HTML and take it with you. There is no account, no export limit and no paid tier, and there is not going to be one: it exists because our own client work needed it and it is cheaper to maintain in the open.
Click anything, change it, export the HTML. Nothing to sign up for.
Tell us what you are trying to fix. If we are the wrong people for it, we will say so in the reply and point you at what we would do instead.
If a form is not your thing, hello@draftship.io reaches the same inbox.
Pilots in this market run about five to fifteen thousand dollars a month and full programs fifteen to forty thousand, and we work inside that range. What you pay depends on how many accounts are covered and whether a defensible list already exists. The figure is in the written plan before you commit to anything, and it does not include your ad spend or your platform licences.
Sources: npws.net and salesmotion.io.
The first call is usually the same week you write. The written plan follows within a week of it. The first month goes on the list and the emails, and nothing is sent until you have approved every message, so a first real send lands around week four or five. Anyone promising replies in week one is sending from a domain they are willing to burn, or from somebody else's list.
An honest account of what has already been tried, your closed-won deals, the list if one exists, access to your ESP, and one person who can approve copy without convening a committee. Two hours from you in the first week and about an hour a week after that.
Yes, from your domain and your ESP rather than ours. SPF, DKIM and DMARC are set up and verified before anything goes out, volume stays inside what your domain can carry, and we watch the complaint rate weekly. Sending from a lookalike domain we owned would protect our reputation at the cost of yours.
You do, as it is made. The account list, the research notes, the copy and the HTML live in files you already have access to, not inside a builder you have to keep renting. If we stop working together you lose us, not the work.
Then we tell you, and we tell you what we think went wrong. Some quarters produce no meetings, and when that happens the answer is usually the list rather than the writing. We will not report opens at you to cover it, there is no commitment past the month you are in, and you can stop after the pilot without an argument.