A practical cold outbound operating guide

This is an operating guide, not an Asphia client case study or a performance benchmark. It outlines choices a team can document and test: narrowing a target before writing, matching channel use to the audience, and defining what counts as a qualified handoff before launch.

The setup most people get wrong

Before the playbook, the trap: buying a large list, sending before the audience and controls are documented, and measuring only send volume. That creates little evidence about what the team should change next.

Volume and relevance are variables to test in the specific market and offer. A documented ICP, exclusions and source context make it easier to review a list before a send; they do not by themselves predict meetings or protect a domain.

So we start with who, not what. Before copy is drafted, the team should be able to explain why an account is in scope and what source supports the proposed context.

Illustration: list volume is not a substitute for documented fit.

Step 1: build a narrow, verified list

We start with the ideal customer profile, not the data tool. Title, company size, industry, region, and a buying trigger. Then we source.

• Apollo for raw data. Company and contact pull against the ICP filters. This is the wide net, deliberately oversized so we can cut it down.
• Clay for enrichment and verification. Every contact gets role-matched, email-validated, and enriched with the signals we will use in copy. Funding, hiring, tech stack, recent moves.
• Hard cut to fit. We drop everyone who does not clear the bar. A list that starts at 4,000 often ships at 900. The cut is the work.

The output should document why an account was selected and what source supports the proposed context. Relevance may inform a legitimate-interests assessment where that basis is available, but it is not a lawful basis by itself. Teams must assess purpose, necessity and the balance of interests, provide required transparency and honor objections, while following the electronic-marketing rules that apply in each jurisdiction. Obtain legal review for the intended campaign.

Compare list-quality signals, unsubscribes, complaints and qualified replies against your own documented baseline. Do not assume a smaller list will outperform a larger one without evidence from the specific market and offer.

Step 2: write copy that earns a reply

With a tight list, the copy job is to be relevant and short, not clever. We lead with the result, name the specific reason this person is getting the email, and ask for one thing.

What goes into every sequence:

• A first line that proves we did the homework. Tied to the trigger from enrichment, not a fake compliment.
• One claim, one number. A specific outcome we can defend, not a wall of benefits.
• One ask. A short call, framed as a question, not a calendar dump.
• Audience-appropriate copy per language. A native-language reviewer can check whether a draft fits the audience and local business context. Translation quality alone does not determine deliverability or reply outcomes.

We do not send a 200-word manifesto. The buyer scans the first line and decides in two seconds. If line one does not earn line two, nothing else matters.

Step 3: run LinkedIn and email together

Channels can serve different roles. Use only the channels, permissions and platform rules your team can operate and measure; do not assume adding a channel improves outcomes.

Channel	Job in the sequence	Tools we run	What it does best
LinkedIn	Warm the name and the face	GetSales, Heyreach	Recognition before the ask, lower-pressure touch
Email	Carry the ask and the proof	Sendkit / Manyreach, Smartlead / Instantly	Detail, links, scheduling, a real opt-out
Combined	Coordinate approved touches	All of the above	A workflow the team can measure and review

One possible sequence is a LinkedIn touch, an email with the actual pitch, then an approved follow-up that does not simply repeat the message. Test cadence, channel and frequency against the audience, permissions and platform rules rather than treating any sequence as universal.

We do not automate to the point of looking like a bot. The connection note is human, the email is human, and a real person handles every reply.

Step 4: protect deliverability or nothing ships

The best copy in the world books nothing from the spam folder. Sending discipline is not optional, it is the whole foundation under the campaign.

• Separate domains from the main brand. Dedicated secondary domains can reduce exposure to the company’s primary domain, but they do not eliminate deliverability or reputation risk. Monitor authentication, volume, engagement, and domain configuration throughout the campaign.
• Mailbox readiness before volume. Assess authentication, mailbox and provider state, and the chosen sending controls before scaling. The required preparation and timing depend on the environment and provider guidance.
• Low caps per inbox. We keep sends low per mailbox and scale by adding inboxes, not by pushing one harder. Filters punish spikes.
• Local-time sending. Each email fires in the recipient’s time zone and work week. A perfect email at 3am their time is a wasted email.
• Compliance review from the start. Assess the applicable privacy and electronic-marketing rules, use a working opt-out and a clear sender identity, and document the chosen basis with appropriate legal review.

This work is easy to overlook. Smartlead and Instantly can support rotation and warmup, while Sendkit and Manyreach can support sequencing. The principle holds whatever you use: manage sending risk and avoid raising volume faster than the evidence supports. Google’s Email sender guidelines describe authentication and sender requirements that may apply.

Step 5: qualify hard and count only real meetings

Here is where we differ from most agencies. We do not count activity, and we do not count every call.

A meeting is qualified when four things are true:

• Right title. The person can actually buy or strongly influence the buy.
• Real budget signal. Not “interesting, maybe next year.” A live reason to spend now.
• A problem we solve. Their pain maps to the offer, not a generic curiosity call.
• They show up. No-shows do not count. A booked meeting the buyer skips is not a meeting.

Positive replies route to a responder in the buyer’s language, live, within the hour where we can. Speed to reply beats any subject-line trick. Slow replies lose warm buyers.

We screen out the noise before it hits the client calendar. Out-of-office, “send me info,” wrong department, and tire-kickers never become “meetings.” Clients pay for meetings that can become deals, so we only hand over meetings that can become deals.

Metrics to define and track

Agree metric definitions before launch. The metrics a team may monitor, per campaign and market, include:

• Positive reply rate. Define the response categories and denominator consistently.
• Qualified handoffs. Define the acceptance criteria, owner and denominator before reporting.
• Show rate. What share of booked meetings the buyer actually attends. This is where qualification pays off.
• Cost per qualified meeting. The number a client can compare against any other channel.

When a campaign underperforms, the metrics tell us where. Low replies means the list or the copy. Good replies but few meetings means the qualification or the ask. We fix one variable at a time, in isolation, without breaking the parts that work.

And when a client wants to own the machine instead of renting the meetings, we hand over the same system, configured and documented, so their team runs it in-house via our done-with-you outbound model. The playbook does not change. Only who operates it does.

No operating choice guarantees a meeting. A documented ICP, source-aware drafting, human review, channel controls and clear qualification criteria give a team a more auditable way to learn from its own results.