Nobody calls it spam when it's obviously about them.
Cold outreach is an interruption, and whether it's a welcome one comes down to how much you knew before you wrote. Narrow the list far enough and you can research one company at a time - what's changing in their market, what that does to a business their size, what the person who owns the problem is actually measured on. Do that and the message is personal because it is personal, not because a merge field made it look that way.
Book Your Impact WeekThe outbound playbook
GTM Engineer allocation
50-100%
Front-loaded. Month one runs at the top of that range, because the list, the research and the tracking all have to exist before anything goes out.
- A list narrow enough to defend name by name, and researched one company at a time.
- Every door that reaches them - email, LinkedIn, phone, post - not just the inbox.
- One variable at a time under test: the profile, the list, or the offer.
Where this playbook comes from.
Outbound is the channel most often done badly, and nearly always for the same reason: the list came first and the thinking came second. Twenty thousand contacts, one message, one channel, and a reply rate that tells you nothing except that twenty thousand people weren't interested. This playbook runs the other way round - a short list, reached everywhere those people actually are.
A Designli go-to-market plan carries several channels at once, and your GTM Engineer decides which ones during the Impact Week that opens the engagement. Outbound earns its place when you can describe your buyer precisely enough to build a list of them by name, and when what you're offering is something a stranger would react to.
The first month is mostly research, not sending. Who is genuinely worth contacting, what's shifting in their market, who inside the business owns the problem and what they're being judged on. Skip that part and you have a blast, whatever you decide to call it.
The whole job is deciding who we bother.
Every name you add to a list makes the message a little more generic, because there's only so much you can know about so many people. So the work is cutting until the list is short enough that you could say something specific and true to every company left on it.
When the messages are landing and nobody replies, the answer is your offer, not more sends. On a list of ten thousand strangers you'd never learn that. On a list of six hundred companies you chose one at a time, a quiet week is a finding.
Which means most of the effort happens before anything goes out. One buyer, one offer, one signal that says this company might care this quarter. Get that part wrong and volume only makes it worse, faster.
How an outbound experiment actually runs.
Six steps, and the first three all happen before anyone hears from you. To keep it concrete, follow one sample customer through them - a company selling job-management software to construction trades, the kind of business where the buyer is an owner in one segment and a project manager in the next.
1. Pick one buyer and one offer.
We narrow to a single buyer - one title, one company shape, one signal that says they might care right now. Everything after this only works if that stays small.
2. Build the list, then score it.
Your engineer builds the universe of companies that fit, from public sources, then scores every one against your criteria. Nobody hears from you until their company has earned a place on the list.
3. Research the company before you write to it.
What's changing in their market, what that means for a business their size, and what the person who owns the problem is measured on. For our sample company that meant knowing the owner decides below a certain size and a project manager decides above it, and that what pushes either of them to look is usually a hiring spurt rather than a software problem.
4. Write to one company at a time.
Each buyer gets one message shape, and the specifics come from the research rather than from a merge field. If a sentence would be equally true of every company on the list, it isn't doing any work.
5. Reach them the way they'd actually answer.
Email, a LinkedIn message, a call, a text, sometimes a letter somebody wrote by hand - the channel follows the person rather than the tooling. Where email carries the load, it leaves from a dedicated domain rather than the one your invoices go out on, and the daily volume stays capped on purpose.
6. Point every reply at one place.
Replies land in a single booking flow, built from day one to hand off to a salesperson when you have one. Then the campaign runs its full window before anyone judges it, and one variable changes at a time.
A typical ramp-up plan.
Every inbox now takes a daily load of AI-written mail sent to ten thousand people at once, which is exactly why the inbox on its own can't carry this. Outbound here means reaching one short list of companies through as many different doors as you can find - email, LinkedIn, a phone call, a text, something that turns up in the post. The first three months tend to run like this.
Month 1: stand it up.
The list gets built and scored, the research behind the first companies gets done, and the tracking goes in behind it. By the end of the month the first messages are going out on the first channel, and the list is an asset you own outright.
Months 2 and 3: the doors multiply.
More channels come online, so one company might get an email, a reply to something their founder posted, and a call inside the same fortnight - each one referring to the same specific thing you found out about them. The picture fills in, and by the end of month three you know which buyer engages, which version of the offer lands, and which half of the list to cut.
Month 4 and on: one audience becomes the next.
A buyer that works becomes the template for the one after it. The scored list, the research pattern, the message shapes and the channel mix all carry over, so the second audience costs a fraction of what the first one did.
What counts, and what doesn't.
Three things are under test, and only three: who your ideal customer actually is, which companies make the list, and what you're putting in front of them. Everything else is execution. When a campaign underperforms, the answer sits in one of those three, and the job is working out which.
So they move one at a time, on a set cadence, and each campaign runs its full window before anyone calls it. Tighten the profile this round, widen the company criteria the next, rewrite the offer after that - and because only one thing changed, the result points at a cause instead of at a guess. Change two together and you've spent a month learning nothing.
What counts is real conversations with people who match the profile you named. Opens and clicks decide who hears from you next, and they never count as a result on their own.
The thing outbound cannot fix is timing. You're arriving on your schedule, not your buyer's, and most of the companies that are right for you aren't thinking about this today. That isn't a reason to send more. It's a reason to keep the list short and the research current, so that when you do arrive, you arrive with something they recognise.
What it does better than any other channel is tell you why. You set the volume yourself - you can decide that six hundred named companies hear from you this month, which is not a thing you can decide anywhere else. And when the answer comes back, you'll know whether it was the profile, the list, or the offer.
Let's find out who's actually worth contacting.
It starts with a free 30-minute call. Tell us who you think your buyer is, and we'll tell you roughly how big that list really is, what it takes to stand the channel up, and whether outbound is the right place for you to start at all.