Outbound sales: the multichannel playbook

Outbound sales is the work of choosing the companies you want as customers and starting the conversation yourself, by email, LinkedIn or phone, before they have shown any interest in you. Most teams treat it as a channel. They buy a tool, write a sequence, load a list and judge the whole idea by the reply rate of that one campaign. When the numbers disappoint, they blame cold email, the market or the SDR.

The reply rate is the last thing that happens, though. Before it come the choice of accounts, the offer, the data, the sending setup and the sequence itself. After it come the replies someone has to sort and the meetings someone has to hold. Outbound is a system of six parts, and it performs at the level of its weakest part. A strong sequence sent from a domain that lands in spam books nothing. A clean list paired with a vague offer books nothing either.

This playbook walks through the six parts in order, explains what good looks like in each, and points you to the deeper guide where one exists. It is written for founders and heads of sales who are deciding how to run outbound, whether they build it in-house or hand it to a team that runs outbound as a service.

In this guide:

• What outbound sales is (and how it differs from inbound)

• The six parts of an outbound system

• 1. ICP and offer

• 2. Data and list building

• 3. Infrastructure and deliverability

• 4. Multichannel sequences

• 5. Reply handling and booking

• 6. Reporting and iteration

• Outbound in Europe

• Build in-house, buy a service, or mix?

• Putting the six parts together

What outbound sales is (and how it differs from inbound)

In outbound, you choose the buyer. You decide which companies should hear from you, which people inside them, and when. In inbound, the buyer chooses you. They find an article, an ad or a referral, and they fill in a form or book a call.

That one difference shapes everything that follows. Inbound leads arrive with some intent, but in the mix and the volume the market decides. Outbound leads arrive with no intent at all, but you control exactly who they are and how many of them you contact. That control is why outbound suits companies with a narrow, high-value market they can name account by account, and companies entering a market where nobody is searching for them yet.

The price of that control is that every reply has to be earned. The prospect did not ask to hear from you, so the message has to explain quickly why it is for them and why now.

Two roles usually share the work. The SDR (sales development representative) researches accounts, runs the sequences and books meetings. The AE (account executive) holds those meetings and closes the deals.

Most B2B companies end up running both motions, because each one covers a gap the other leaves. If you are deciding which to build first, compare the two on cost, speed and fit with your market.

The six parts of an outbound system

Each part feeds the next, and the last one sends you back to the first.

1. ICP and offer. Who you target (your ICP and the size of that market) and what you offer them.

2. Data and list building. Turning the ICP into named accounts and verified contacts, with tools such as Sales Navigator, Apollo and Clay.

3. Infrastructure and deliverability. The secondary domains, inboxes, authentication and warmup that get email into the inbox.

4. Multichannel sequences. Email, LinkedIn and phone, each with its own role.

5. Reply handling and booking. Sorting replies quickly and turning the good ones into held meetings, then handing them over cleanly.

6. Reporting and iteration. Measuring positive replies, meetings held and pipeline created, then changing one thing at a time.

The order matters most when something goes wrong. A weak result usually shows up in part 6 while the cause sits further up. A low reply rate can mean the copy is weak, but it can just as easily mean the emails are landing in spam (part 3), the list is full of people who do not fit (part 2), or the offer gives the reader no reason to answer (part 1).

1. ICP and offer

Your ICP (ideal customer profile) describes the company that is most likely to buy and most likely to get value from what you sell. A useful ICP goes past industry and headcount. It names the geography, the business model, the tools the company already uses, and above all the trigger that makes now a good moment to talk: a new funding round, a leadership change, a hiring push in a specific team, an expansion into a new country.

Inside each account, the ICP also names the people. Who feels the problem every week? Who owns the budget? Who signs? In most B2B deals these are different people, and the sequence each of them receives should differ too.

Your TAM (total addressable market) is the number of accounts that match the ICP. Counting it early tells you two things. It tells you how long the program can run before you start contacting the same companies again, and whether outbound alone can plausibly reach your pipeline target. A small TAM calls for more research per account and a slower, more personal approach. A large one leaves room to test segments against each other.

The offer is what you ask the prospect to say yes to. A cold prospect has no reason to give a meeting to a stranger, so a strong offer lowers the cost of saying yes. It might be a specific finding about their company, a short audit, a benchmark from their segment, or simply a clear and narrow reason to talk. A weak offer is a feature list followed by a request for a call. A quick test is whether the prospect could explain, in one sentence, what they get from replying and why the email was sent to them.

The ICP and the offer are worth getting right before anything else, because every later part inherits their quality. For a worked example of how buyers, triggers and offers look in one vertical, see our lead generation report for IT services companies.

2. Data and list building

List building turns the ICP into three layers of data: the accounts that match, the people inside them, and contact details you can trust. Each layer has its own tools.

Finding accounts and people. LinkedIn Sales Navigator is the standard starting point because it searches LinkedIn's own network, which LinkedIn puts at more than 1 billion members as of November 2026 (Sales Navigator). As of the same date, LinkedIn's own prospecting guide lists 15 account filters and 36 lead filters (more than 50 in total). You can save a search and get notified when new accounts match it, and an "account has buyer intent" signal is available on the Advanced and Advanced Plus plans only.

Contact data. Sales Navigator tells you who to contact but not how to email them. Data platforms such as Apollo fill that gap. As of November 2026, Apollo presents itself as "one connected GTM system", with B2B contact data, prospecting, outreach sequences and a dialer on the same platform (Apollo). Enrichment tools such as Clay work differently. Clay calls its approach waterfall enrichment, which means it searches several data providers one after another until it finds a valid match. The logic behind a waterfall is that no single provider covers every country, industry and seniority equally well, so asking several in sequence fills more gaps than relying on one.

Verification. Every email address should be verified before it enters a sequence. Bounced emails hurt the reputation of the sending inbox, which hurts every later campaign sent from it (more on this in part 3).

Two habits separate good lists from big ones. The first is building a separate, hand-researched list for your top accounts instead of running them through the same filters as everyone else. The second is storing the reason each account is on the list, such as the trigger, the tool they use or the role they are hiring for, so the sequence can use it.

If you are choosing between the main data tools, the core question is whether a database is enough or whether you need an enrichment workflow on top. Whichever you pick, estimate what a workflow will cost before you run it on a large list.

3. Infrastructure and deliverability

Deliverability is whether your emails land in the inbox, the spam folder or nowhere. It is the part teams skip most often, because a broken setup still sends emails and still shows them as delivered. The emails simply never reach anyone.

The usual setup has four pieces:

• Secondary domains. Cold email goes out from domains that are separate from your main company domain, typically close variations of it. If a cold campaign damages a sending domain's reputation, your invoices, customer email and team inboxes stay untouched.

• Several inboxes. Volume is spread across multiple inboxes on those domains, so no single inbox sends enough to look like a bulk mailer.

• Authentication. Every sending domain needs SPF, DKIM and DMARC records, which prove to the receiving server that you are allowed to send from that domain.

• Warmup. New inboxes start slowly and build volume over time, so mailbox providers see a sending history before the first campaign goes out.

Mailbox providers publish their rules, and Google's are the ones most B2B senders meet first. As of November 2026, Google's email sender guidelines say that everyone sending to Gmail accounts must:

• set up SPF or DKIM for their sending domains

• have valid forward and reverse DNS records for their sending domains or IPs

• send over a TLS connection

• keep the spam rate reported in Google Postmaster Tools below 0.3%

• format messages according to RFC 5322 and never impersonate Gmail in the From header

Google treats anyone who sends close to 5,000 messages or more to personal Gmail accounts within a 24-hour period as a bulk sender. It counts every message sent from the same primary domain together (its example counts mail from a subdomain toward the main domain), and a sender that meets the threshold once is permanently considered a bulk sender (bulk sender FAQ). On top of the list above, bulk senders must set up SPF, DKIM and DMARC (the DMARC policy can be set to "none"), align the From domain with either the SPF or the DKIM domain, and support one-click unsubscribe in marketing and subscribed messages, with a clearly visible unsubscribe link in the body. Both sets of requirements have applied since February 1, 2024. Google also recommends keeping the spam rate below 0.10% and never reaching 0.30%. The page also warns that Gmail will begin using a DMARC quarantine enforcement policy, so mail that impersonates a Gmail From: header might not be delivered. No date is given.

We recommend treating the bulk-sender list as the baseline even if your domains never come close to 5,000 messages a day. It describes what a mailbox provider considers a well-run sender, and a cold program, where the recipient never asked to hear from you, has the least room for spam complaints. Make opting out easy in every email and honor every request straight away. An unsubscribe costs you one contact. A spam complaint costs you reputation on the whole domain.

Deliverability needs watching, not just setting up. Check spam rates in Postmaster Tools, watch bounce rates per inbox, and rest any inbox whose numbers start to slip before it drags its domain down.

4. Multichannel sequences

A sequence is the ordered set of touches a prospect receives, and the cadence is how those touches are spaced. A multichannel sequence spreads the touches across email, LinkedIn and phone. That works best when each channel has its own job instead of repeating the same message in three places.

Email carries the message. It reaches a large list at low cost, the prospect can read it whenever suits them, and it can hold enough detail to explain the offer. The first email should say why this person, why now, and what they get from replying. Each follow-up should add something new, such as a different angle, a relevant example or a simpler ask, rather than "just bumping this up." Our guide to cold email follow-up covers how many to send, how far apart, and what each one should say.

LinkedIn builds familiarity. A profile view, a connection request without a pitch, or a thoughtful comment on a post means your name is not new when the email arrives. Stay inside LinkedIn's own limits on invitations and messages, because a restricted account takes a channel out of the sequence for everyone.

Phone gives the strongest signal per touch. A short conversation can tell you more than weeks of email. Calls work best for top-tier accounts and for prospects who have already engaged in some way, and they go better when the caller can refer to an email the prospect has seen. Rules on calling businesses differ by country (see the Europe section below).

Three rules hold across channels. Every touch should make sense on its own, because the prospect may see only one of them. Personalization should come from the segment and the trigger, which tell the reader why the email is relevant, rather than from a forced compliment about their latest post. And the sequence should stop as soon as the prospect replies on any channel, asks to be removed, or a colleague answers on their behalf.

5. Reply handling and booking

Replies are where pipeline is won or lost, and they need a process as clear as the sending side. Every reply should land in one of a few categories, each with an agreed next step.

Speed matters most in the first row. A prospect who replies with interest is thinking about the problem right then, and that attention does not last. A sensible standard is to answer positive replies the same working day, offer two or three specific times as well as a booking link, and send a calendar invite with a one-line agenda as soon as a time is agreed. A short reminder before the meeting, with something relevant to discuss, reduces no-shows.

The handover from SDR to AE is the step that most often breaks. The AE should receive a short note with who the prospect is, what they replied, which problem they mentioned and what was promised to them. A prospect who has to repeat everything in the first meeting starts the deal with a worse opinion of you.

If you would rather buy booked meetings than run this process yourself, the build or buy section below compares the options.

6. Reporting and iteration

Outbound reporting should follow the money backwards. In order of how much they matter:

• Pipeline created. The value of qualified opportunities that came from outbound meetings.

• Meetings held. Meetings that actually happened, which is different from meetings booked.

• Meetings booked. The output of reply handling.

• Positive replies. Replies that show interest or a referral, as opposed to all replies.

Beneath these sit health metrics that explain the result rather than define it: bounce rate, spam rate, total reply rate and opt-out rate. Opens no longer belong on the list. As of November 2026, Apple's Mail Privacy Protection downloads remote content in the background by default, so for Apple Mail readers with the protection on, tracking pixels fire whether or not anyone reads the email, and Smartlead's 2026 benchmark leaves open rates out for that reason. Hunter even found that campaigns without open tracking had a higher reply rate, 7.4% against 4.4%, across 31 million emails sent by its users in 2025 (Hunter, vendor-reported).

Benchmarks are useful as a sanity check, not as a target. One recent large dataset is Smartlead's State of Cold Email 2026 (reply rate benchmark, vendor-reported), which covers more than 850 million emails sent through its platform in the first half of 2026. It defines reply rate as replies divided by contacts and only includes senders with at least 500 emails in the period. On that basis, the median sender reached 0.74% (about one reply per 135 contacts), the top quarter 1.43% and the top 10% 2.63%.

Other platforms measure differently and report higher averages. Instantly's 2026 benchmark report puts the 2025 average at 3.43%, counting all replies divided by total emails sent, and Hunter puts the average sequence reply rate for 2025 at 4.5%. Both are vendor-reported.

Two cautions apply. The gap between these platforms shows how much the definition and the sender mix move the number, and each figure comes from one platform's users. Your own baseline, per segment and per offer, is the number worth beating.

Iteration means changing one variable at a time and giving each change enough volume to read. When results drop, walk the system from the bottom up: check deliverability first, then list quality, then the sequence, and only then the offer and the ICP.

Outbound in Europe

Outbound in Europe follows the same six parts, with two extra layers: data protection and country rules.

GDPR applies to business contacts. A name and a work email address are personal data. Recital 47 of the GDPR says that processing personal data for direct marketing "may be regarded as carried out for a legitimate interest" (source). The word "may" matters, because legitimate interest requires a balancing test you should be able to document. Article 21 gives every recipient the right to object to direct marketing at any time (source). In practice, that means recording why each contact fits your ICP, being able to say where the data came from, making opt-out easy, and keeping a suppression list that every tool respects.

Country rules differ. Electronic marketing is also governed by national law, and the strictness varies. Germany is the usual example. Section 7(2) no. 2 of the German Unfair Competition Act (UWG) treats advertising by email without the recipient's prior express consent as unreasonable harassment, and the wording covers any recipient, not only consumers. For phone calls, section 7(2) no. 1 requires at least presumed consent from business recipients (source). For Germany, a sensible plan therefore leans more on LinkedIn and on carefully targeted calls than on cold email. Check the rules for each target country with qualified counsel before launch. This section is an overview, not legal advice.

Plan each country separately. In our view, there is no single European reply rate worth planning around. Language, local holidays, how much cold email a market already receives and where the sender is based all change how a campaign lands, so it is worth testing each country as its own segment instead of averaging them together.

Build in-house, buy a service, or mix?

Once the six parts are clear, the real question is who runs them. There are three common answers.

Build in-house. You hire SDRs, buy the tools, and own every learning and every data point. You also own the ramp-up: hiring, onboarding, domain warmup, list building and the months before the system reaches a steady state. This model fits companies that have someone with outbound experience to manage the team.

Buy a service. An outsourced SDR team runs the whole system under your brand, from list to booked meeting, with infrastructure and tools already in place. It is usually faster to start and easier to stop, and you keep the AEs who close. Our guide to outsourced SDR explains what the model includes and when it fails. If you only want meetings and not the full program, appointment setting is a narrower option with different trade-offs.

Mix. Many teams split the work. An external partner builds the infrastructure, the data and the sequences, while in-house SDRs or AEs handle replies and calls.

Four questions usually settle it. How soon do you need pipeline? Do you have someone who can manage SDRs? How much of your market knowledge lives in people's heads rather than in documents? And would you rather own the learning curve or rent someone else's? If you do go external, ask any partner to show how they run each of the six parts above before you sign. To see how we structure a program ourselves, here is how we run outbound.

Putting the six parts together

Outbound sales rarely fails because cold email stopped working. It fails because one of the six parts is weak and the team keeps fixing a different one. Pick an ICP you can name account by account, give the prospect an offer worth replying to, build a verified list, send from infrastructure that reaches the inbox, run sequences where each channel has a job, answer replies fast, and measure pipeline rather than opens. Then loop back to part 1 and tighten it.

For more depth on two parts of this playbook, our guide to outsourced SDR covers what an SDR-as-a-service team includes and how it is priced, and our lead generation report for IT services companies shows the system applied to one market.

Want this system running without building it yourself? Book a call and we will walk you through how we run outbound and whether it fits your market.