When outbound sales slow down, most revenue teams try changing the messaging approach or increasing their outreach volume. Rewriting templates is cheap, and raising daily quotas is relatively fast.
But this approach assumes the problem lies at the tactical level, when in fact an outbound lead generation process is a complex system of numerous touchpoints built upon each other. A failure in an early layer doesn’t become apparent until much later, when it turns out that the outreach channel was picked incorrectly or a lead database contains wrong contacts because it hadn’t been refreshed for 18 months.
Since 2017, our outbound lead generation agency, Belkins, has run client acquisition for 1,000+ businesses across 50+ industries, and that gap between where a team believes outbound fails and where it is actually broken is a consistent pattern in the data. Below are proven steps to avoid this gap, audit your current situation, and build a lead generation process for a predictable, manageable, revenue-focused pipeline.
Step 0: How to diagnose what’s broken in your outbound lead gen process
Before you change anything, work through the five checks below in the order they appear. Each one assumes the check above it passed, so the first failure you hit is the layer worth your immediate attention.
Is the list right? Do these companies match our compliance policy, use a relevant tech stack, have the problem we solve, the budget to fix it, and the authority to act?
Are the records accurate? Are these people still in these roles, and do the email addresses and direct numbers still work?
Is the message relevant? Does it connect to something true about this specific account rather than about the segment it belongs to?
Are the channels working as one sequence? Does each touch build on the last rather than restart the conversation?
Is the exit criterion honest? Would you have taken that meeting yourself?
In our practice, the two most frequent findings are wrong targeting and a generic message with surface-level personalization, and they present differently. Targeting faults show up as prospects saying the offer is not relevant, redirecting you to other departments, or raising the same objection over and over again. Messaging faults show up as low reply and positive engagement rates while the audience itself still looks correct.
Once you understand the core of your process, it’s time to audit what is considered an effective outreach campaign. Sends, dials, connection requests, and open rates can all hold steady while pipeline goes to zero, because activity is the last thing to degrade when a motion breaks. That creates a specific trap for anyone inheriting an existing outbound function, since the dashboard looks healthy while it measures effort.
Open rate deserves particular suspicion. For example, enterprise mail security gateways automatically open and crawl inbound messages and can inflate the number toward 100%. And vice versa, some outreach automation platforms can trigger spam filters when the open-rate tracking pixel is on. So our team treats this performance indicator as a supporting signal on infrastructure health rather than as evidence of interest.
What actually moves first when a campaign degrades is narrower: our delivery team watches response rate against the campaign’s benchmark, bounce rate, and engagement rate, then escalates any unexpected decline to the deliverability team before meeting volume is affected.
Eventually, your priority KPIs should be pipeline growth, deal velocity, cost per meeting and cost per deal, show rates, and return on investment. Each can indicate the weak spot in the process. For instance, a high lead-to-appointment rate that doesn’t transform into real meetings for your sales team might signal a broken scheduling or follow-up approach.
Step 1: Check your market size and unit economics before designing an outbound campaign
Two questions decide whether outbound is worth running at your organization at all: whether your market is big enough, and whether your deal size covers what it costs to win one. The math works the same way whether you run the campaign yourself or hire someone to do it.
Size your target market against the volume of outbound needs in order to iterate
Outbound needs volume before it produces meetings, because you have to test audiences, messages, and channels to find out what works. Every test uses up contacts. If your market is small, you spend the whole list learning how to sell to it and have nobody left to sell to.
Belkins’ documented minimum to launch a campaign is 9,000 contacts. After the pilot, a typical campaign uses around 1,500 new contacts a month, though that number changes with the campaign goals, the strategy, budget calibrations, seasonal events, and how much manual validation the research needs.
A small market does not mean you should send more email. Where a company sells at a high average contract value (ACV) and has 100 to 500 named target accounts, we point them to account-based marketing (ABM) instead, which works through individually tailored messaging and long-term engagement rather than volume. Where that company also has no customer results to point to yet, we usually suggest referrals or assist with targeted events or content marketing first.
Model cost per qualified meeting against your average contract value and gross margin
Work out what the campaign costs, divide it by the number of meetings you expect, then multiply by the share of meetings your team actually closes. Compare the result to your ACV and your gross margin. Sometimes, revenue leaders get this wrong in two ways: they measure against top-line revenue instead of gross margin, and they use the close rate they hope for instead of the one their CRM shows.
A long sales cycle changes the question. Instead of asking whether you can afford the campaign, ask whether you can keep paying for it through the months between launch and the first deal closing. That is a cash-flow problem, and it is what sinks business cases that otherwise add up. For outsourced programs, our commercial team treats an ACV of roughly $15,000 to $25,000 as the point where the numbers start to work, because closing 2 to 4 new contracts at that level covers the annual cost.
📚 Relevant reading: B2B lead generation cost: how much to pay for leads?
Confirm your internal team can work the meetings before you launch
Outbound works for a fairly narrow set of companies: a market large enough to test against, a sales cycle long enough to justify paying for the campaign over months, deal sizes that cover the cost of winning them, and someone in-house ready to take the meetings.
A company can pass the first three and still fail on the fourth. The most common reason a qualified client’s campaign stalls is that nobody owns the meetings, so no one is assigned to run the discovery call, follow up afterward, or keep CRM records clean.
Step 2: Build the target account list and define the signals that time your outreach
Nearly all the decisions that form an outbound campaign and determine its success are made at this stage.
Define an ICP in a way a researcher can build a list from without asking extra questions
An ideal customer profile (ICP) is usable and correctly defined when it contains three things:
The firmographics you can filter on
The events that make a company need you now
The job titles that control the budget and have decision-making authority
Test it by handing the definition to whoever builds your leads database. If they come back with questions, it is not finished.
The most correct way to create an ICP is to crystallize it from your own closed deals, looking at these important factors:
The segments benefit the most from your offering
The segments that take less time and effort to close
The clients that deliver you the most revenue
The types of companies with the highest retention rates
When a client brings us an ICP they already trust, our account strategists check it against their case studies, their strongest customer segments, and where deals have actually closed. The list we end up working from is often different from the one we were handed, because we are looking for the audiences that return the most revenue rather than the segment that books the most meetings.
💡
Case in point: That first example is close to the definition
we built for Virayo, a B2B SaaS SEO agency in a saturated market, where narrowing the list rather than rewriting the pitch produced 3 closed-won deals with key accounts in 5 months.
Separate timing signals from fit signals when you prioritize accounts
Fit tells you which companies belong on the list. Timing tells you which ones to contact this week. Most teams collect the first kind and treat it as the second, so they work through a list of good-fit companies in whatever order it was exported and reach most of them at the wrong moment.
The signals our SDR teams find most predictive of a booked meeting are things the prospect actually does:
Accepting a LinkedIn connection request
Opening or clicking a shared document or proposal link more than once
Visiting pricing or case study pages repeatedly
Replying with anything at all, including an objection, and then going quiet
Raw email opens and generic website clicks are the two to distrust. Both get sold as intent data, and both are unreliable enough that we would not change our contact order because of them.
Knowing which software a company runs is often worth more than its firmographics, because it gives you something specific to write about. If you know an account runs a particular backend system, you can open with the integration problem that system creates rather than with the company name dropped into a merge field.
Source, verify, and refresh contact data on a recurring schedule
Ready-to-buy lead databases fail mostly because some of the people moved to other companies, and some were never the right people to contact. Both show up later as poor engagement rates, which is why they’re usually misread as a copy problem. Records go out of date continuously, so checking them has to be a routine rather than a one-off cleanup.
Let’s look at how quality lead sourcing can be realized in practice. Belkins’ research system pulls firmographic, technographic, and intent data from 100+ sources in real time, and a human researcher checks every record before an SDR uses it. That step catches what software cannot. For example, a tool will confirm that a Marketing Director’s address is valid and the headcount matches your filter, while a researcher will notice that the company is a B2C franchise where head office controls all purchasing, or that a “Director of Sales” at a 12-person company is a full-cycle rep with no budget.
💡
Case in point: Autodesk Fusion, the cloud-based design and manufacturing platform, came to Belkins with roughly 8,000 contacts that were largely unchecked and did not match their ICP. We rebuilt the database and validated around 1,000 genuinely ICP-fit contacts from that batch. The engagement produced 35 qualified appointments in under 4 months, 110% of the agreed target, with 14.3% of engaged prospects converting to an appointment.
Step 3: Orchestrate messaging and channels as one sequence
A well-crafted email still performs badly when it arrives on a schedule that ignores what the prospect has been doing. That’s why we treat sequencing and messaging as a whole.
Decide which accounts are worth researching by hand
What limits personalization is research hours, not ideation, so you have to decide which accounts justify this.
Belkins splits a client’s list into two tiers:
The top 10% to 20% get manual research. For them, copywriters work out which problem that specific company is likely dealing with and the people on its buying committee. Then they pick the client result that speaks to it most directly, ideally an outcome from a close competitor or from a company at the same stage of growth.
The remaining 80% to 90% get a proven copy framework filled with variables our tools extract automatically, such as a cost saving calculated against industry benchmarks or a supplier relationship we have verified.
We never let AI write a whole message. The framework stays human-first because fully generated copy reads inconsistently. Furthermore, our proprietary research on using AI for lead gen showed that almost 38% of revenue leaders consider AI robotic output quality as their biggest challenge in using the technology. Senior buyers tell us that outreach sounding canned costs them more than a lost reply, since it worsens the reputation of the company that sent it.
💡
Case in point: Mapping the buying committee is what makes the top tier worth the hours.
For Terrascope, a carbon management platform selling to large enterprises, we wrote separately for chief sustainability officers, sustainability managers, procurement leads, and CFOs, because a procurement lead and a CFO want different things from the same product.
Set your touch count and channel order from your sales cycle and buyer seniority
Copying someone else’s cadence rarely works, because the right number of touches depends on your deal velocity, your buyers’ seniority, the channels those decision-makers actually use, and how much research went into each account. Our standard sequence for a client selling to senior buyers runs 14 to 21 days across three channels:
Email: three waves. Wave one carries the strongest hook and the best proof point; wave two is a short bump in the same thread; and wave three opens a separate thread with a new angle. We rarely send more than 3 waves — our data shows replies drop sharply with each new follow-up.
LinkedIn: four to five waves, starting only after the connection request is accepted. We send the invite with no message, which has consistently produced the best acceptance rates. The first three waves stay conversational, and waves four and five turn direct.
Calling: 2 to 3 attempts running in parallel with the digital touches.
Which of those three channels carries the most weight depends on the persona more than on the industry. For instance, plant and operations managers need more calling; engineering and analytical leaders reply best to email; marketing and HR leaders are easiest to reach on LinkedIn.
Cold outreach also works better when the person has already come across your company somewhere, whether through a LinkedIn post, an article, or a conference. That recognition changes how they treat the first message, which makes brand and content inputs to outbound performance rather than a separate funnel. This is why we describe our model as allbound rather than purely outbound.
What we recommend to avoid is sending the same message across three channels on parallel schedules. It looks like omnichannel, but every touch restarts the conversation instead of continuing it.
📚 Relevant reading: Omnichannel lead generation for B2B: how to do it right
Trigger calls from engagement signals instead of a fixed dial schedule
Calling someone who has just engaged with your company works better than calling down a cold list, both in how often you connect and in how the conversation goes. Our SDRs review engaged prospects every week, meaning the ones who accepted a connection request, opened a shared document more than once, or replied and then went quiet, and they call while it is still recent and refer to it directly.
Our own benchmarks show the size of the difference. Cold dialing connects roughly 3% to 15% of the time, and we treat 7% as the working target. Calling into an engagement signal starts at 15% and reaches 25% or higher.
This also settles two opposite worries we hear from clients: that cold calling will damage the brand, and that email-only outreach is too lazy, since calling someone who just engaged with you is neither. It does need infrastructure, skilled SDRs, and a tracking system in place first.
Protect your sending domains and check where your emails land
Sender reputation, mailbox warm-up, authentication, and sending limits need weekly attention, because a damaged sending domain undoes everything else you have built. The mistake we see most often is a company running high-volume campaigns from its main corporate domain without SPF, DKIM, and DMARC configured, which puts the reputation of its everyday business email at risk.
Belkins checks deliverability for each mailbox and each project through Folderly, our proprietary deliverability platform, which shows where emails land in real time and identifies DNS or server faults before a campaign scales. We warm spare domains and mailboxes in advance for every project, so when monitoring shows 2 in 10 emails from one mailbox going to spam, we take that mailbox out of rotation and move its volume the same day.
If you run outbound in-house, ask these questions all the time: which domains are sending, where those emails are landing, and who is responsible for monitoring.
📚 Relevant reading: How to improve email deliverability: Belkins' framework
Step 4: Qualify meetings, hand them off, and re-engage the uninterested accounts
A booked meeting is not the finish line, and the failures that can come after it might cost a fortune. What you do between the booking and the sales call decides whether your team trusts outbound or writes it off.
Decide on and document qualification criteria
Write down what counts as a qualified meeting before the campaign starts, then hold to it. Without that, “qualified” gradually comes to mean whatever the SDR managed to book. We settle this during onboarding, alongside the targeting and exclusion criteria. There is no universal standard, since the criteria depend on what you sell and to whom.
A meeting with the wrong person costs more than no meeting at all, because it uses an AE’s time and makes your sales team doubt the whole strategy. When a client’s team rejects a booking, our SDR and account strategist read back through the conversation log and CRM notes, record why it was rejected in HubSpot, and change the research criteria for the next batch of contacts within 24 to 48 hours.
📚 Relevant reading: B2B lead qualification process we use for clients
Brief the AE and run a confirmation sequence to protect show rates
The account executive taking the call needs the context the SDR already has. The meeting brief we use at Belkins covers contact details and company firmographics, the full conversation history across email, LinkedIn, and phone, what the prospect said they were interested in, and any objection or expectation raised.
Then, it’s crucial to secure solid show rates, and it doesn’t come with luck. Our SDRs verify the calendar invitation was delivered and accepted, then send a reminder one day and a few hours before the call. We treat 70% or higher as the benchmark for meetings actually held, and a project that books plenty of meetings while missing that number is usually on its way to churn.
Prevent contacting the same accounts
If you already have AEs or BDRs prospecting, they risk contacting the same account. This is a process problem with ordinary solutions: do-not-contact lists that exclude accounts your team is already working on, clear territory or segment splits, and one shared CRM where all activity is recorded.
Belkins works inside the client's own HubSpot or Salesforce wherever possible, and in their communication tools, so both teams see activity as it happens instead of reconciling it afterward.
💡 Case in point: Fire Safety & Protection ran this with 30+ salespeople already in place, and the split held because Belkins handled research and first contact while their own team handled everything from the first meeting onward.
Plan how you will contact the accounts that were not ready
Most companies you contact are not looking to buy right now, and writing them off means throwing away a list you paid to research and verify. As we do for our projects, we recommend cooling contacts down for 30 to 60 days after a completed sequence, then re-engage with a genuinely fresh pass — a different subject line, angle, new case study, or webinar invitation.
That matters because a second attempt reading like a fifth follow-up produces irritation rather than replies. A meaningful share of the meetings our campaigns book comes from these following attempts rather than from first contact. This is why outbound rewards running it for longer than a quarter. Where sales cycles are long, it may be the only reason the numbers work at all.
📚 Relevant reading: How we nurture B2B leads: Belkins’ strategy
Step 5: Track pipeline conversion and set ramp expectations you can defend
For roughly the first quarter — which is about as long as the ramp-up takes — both working and failing campaigns produce similar-looking numbers. Unless you focus on pipeline-related metrics, you cannot tell which one you have until the budget is spent.
Set expectations for month four rather than month one
How quickly outbound produces results varies widely between companies. The figures below are what we typically see across client campaigns, and your own timeline will depend on cycle length, deal size, number of approvers, and infrastructure you have.
We get campaigns live in 14 days and see tangible results in 30 to 90 days, with steady performance from month four onward. Months 1–3 are calibration, covering infrastructure warm-up, list building, message testing, and the first real signal on what converts.
What you should commit to internally is a curve with a checkpoint on it. By the second monthly review, about a third of the pilot's projected appointments should be booked, and the pipeline created in month one should be visibly moving. First closed deals usually appear around month five or six, but this still depends on your industry benchmarks.
💡
Case in point: a fintech investment platform came to Belkins after fragmented work with several vendors that had converted at barely 3%. Results arrived in the first month, and the campaign expanded from email into LinkedIn and intent-based calling after month four. Over 15 months, it produced 346 appointments at a 9% conversion rate, three times what the previous vendors managed, and the monthly quota rose from 15 to 30 once performance settled.
Track meeting-to-pipeline conversion as your primary quality signal
Of everything you can measure, the share of meetings that turn into real pipeline tells you the most, because it separates a volume problem from a quality challenge. Plenty of meetings with poor conversion usually means the targeting is wrong, while good conversion with too few meetings usually means the market is too small or your team lacks capacity. Alongside it, we recommend reporting qualified appointments against target, the share of meetings actually held, engagement rate as a proportion of all replies, and cost per qualified meeting.
Emails sent, dials made, connection requests, and impressions should come off the dashboard rather than move further down it. As long as they stay visible, whoever is defending the budget upward will reach for whichever number looks best, and effort always looks better than results. We stopped reporting raw open rates for the same reason, since security software inflates them toward 100% even while a campaign is being filtered into spam.
📚 Relevant reading: B2B sales pipeline conversion rates: Belkins’ data
Should you build outbound in-house, buy software, or outsource it?
Step 1 answered whether outbound is worth running for your business. Now, we’ll answer who should run it in your particular case.
Work out what an in-house SDR team really costs
Building internally costs considerably more than the salaries suggest:
Minimum of two junior SDRs: roughly $100,000 to $120,000 a year combined, base pay plus commission
A manager who has done the job before: $80,000 to $100,000 or more
Tools, covering CRM, contact databases, deliverability warming, dialers, and Sales Navigator: roughly $15,000 to $20,000 a year
Allow 3 months before new hires are productive, and the first-year total passes $200,000. Turnover here is built into the role rather than a sign of bad management, because SDRs who do well get promoted into AE positions and the ramp starts again. Sales leaders tell us they have lost several SDRs to internal promotion in a single quarter, then spent two to three months hiring replacements before training could begin. Building in-house still makes sense where you have steady volume, a manager who has built this before, and a product that needs real technical depth in the first conversation.
📚 Relevant reading: Outsourced vs. in-house sales development
Identify the processes software won’t do for you
Tools make an existing process faster. They don’t supply the judgment behind the research, the writing, the qualification, or the willingness to keep changing your targeting, and those four are what decide results. But buying tools without the people to run them usually produces an expensive unused stack and sending volumes larger than the domain setup can safely handle. Software is the right answer for a team that already has a working process and wants more output from it, though it rarely helps a team that needs the process itself.
Check whether you are ready to outsource
Outsourcing works where the ACV clears roughly $15,000 to $25,000, the company can sustain the monthly investment for at least 12 months without expecting immediate returns, there are verifiable customer outcomes to point at in the messaging, and dedicated internal reps are ready to take every booked meeting. It fails where the market sits below the contact floor, the ACV cannot absorb the acquisition cost, positioning is unresolved, or nobody has been assigned to receive the meetings.
One criterion sits above the rest: what transfers to you regardless of how the engagement ends. Every validated database, custom subdomain, mailbox, messaging sequence, and CRM record is your intellectual property, and our offboarding hands over full ownership along with the templates, battle cards, and strategy documents built during the partnership. If none of that stays with you, you rented an outcome rather than building infrastructure — a fair question to put to any outbound lead generation agency you are evaluating, including Belkins.
Run both at once while you build the team
This is a third option between selling founder-led and funding a full SDR team. Here’s how it looks in real life with our agency: Belkins handles the technical setup and first contact while the client hires and trains their own reps. It works only if ownership is split clearly and both teams share one CRM, so no account gets approached twice. When the client takes over, they inherit everything we built: warmed mailboxes and domains, verified lists, tools, and sequences that were already performing.
Where to start
If you already have an outbound lead gen process running, start with the diagnostic at the top of this article. Work down the five checks, stop at the first one that fails, and fix that layer before you touch anything else. The fault usually sits earlier in the process, and finding it costs far less than another quarter of rewrites.
Building from scratch changes the order. Answer the two viability questions first, because no amount of good execution will rescue a market too small to test against or a deal size that cannot cover what it takes to win a customer.
And if the outsourcing criteria described your business fairly well, book a call. We will look at what you are running now and tell you whether outbound is worth the investment at this point, including when the answer is no.
