The Agencies That Build the List Win the Renewal
There are two kinds of appointment-setting agencies: the ones that send to whatever list the client hands over, and the ones that own the data layer — building each client's list themselves and standing behind it. The second kind charges more, books better meetings, and keeps clients longer, because the quality of every meeting traces back to who was on the list in the first place. A perfectly written sequence sent to the wrong 2,000 people books meetings that no-show, stall, or never close — and the client churns blaming you.
Apollo is the tool that makes owning the data layer practical: a B2B contact database with title, industry, and signal filters, saved personas you can maintain per client, and sequences to send from. ColdRelay is the layer underneath that — the secondary domains, mailboxes, and dedicated IPs Apollo actually sends through. This guide covers how list-owning agencies wire the two together: encoding each client's ICP as a saved persona in Apollo, sourcing from the database instead of buying third-party lists, and sending it all from infrastructure that never touches the client's primary domain.
Why List-Owning Agencies Run Apollo on ColdRelay Infrastructure
When list-building is part of your service, your outbound stack has three layers, and Apollo covers the middle two: data (the contact database, filters, and saved personas) and sending logic (sequences with email, call, and LinkedIn steps). What Apollo doesn't provide is the bottom layer — the domains and mailboxes those sequences send from. Apollo sends through whatever mailboxes you link under Settings → Mailboxes; the deliverability of those mailboxes is the infrastructure's job, not Apollo's.
That's where ColdRelay fits: dedicated mailboxes on isolated Azure tenants with dedicated IPs, DNS (SPF, DKIM, DMARC) pre-configured, provisioned in about an hour, with 95%+ inbox placement. The pairing matters more for list-owning agencies than for anyone else, because data quality and deliverability multiply. A precisely built Apollo list hitting spam folders wastes the research; a clean inbox placement rate on a junk list wastes the infrastructure. Tight personas plus 95%+ placement is the combination where each layer's quality actually shows up in booked meetings.
The pairing is additive, not competitive: ColdRelay is the infrastructure layer, Apollo is the data and sequencing layer on top. You keep Apollo's database, personas, and multi-channel sequences — you just send them from mailboxes built to land.
Visit Apollo →From Client ICP to Sending: The Apollo + ColdRelay Setup
Run the closed-won analysis before touching Apollo
Start every new client with their CRM, not the database. Pull their last 20-40 closed-won deals and extract the pattern: titles that signed, company sizes, industries, and any common trigger (new funding, hiring for a role, a tech they run). This is the raw material for the client's ICP — and it's the difference between a list built from evidence and a list built from the client's guess about who their buyer is.
Encode the ICP as a saved persona in Apollo
Translate the closed-won pattern into Apollo's filters — job titles, seniority, industry, employee count, technologies, and intent signals — and save it as a named persona for that client. One persona (or two or three, if the analysis splits cleanly) per client, named so any setter on the team can find it. Saved personas turn list-building from a one-off project into a maintained asset: rerun the saved search monthly and fresh contacts matching the proven profile flow in.
Provision the client's mailbox pool on ColdRelay
While the persona work happens, order the client's dedicated sending pool — fresh secondary domains themed to the client, never their primary domain. ColdRelay supports 100-150 mailboxes per domain, everything lands on isolated Azure tenants with dedicated IPs, DNS pre-configured, in about an hour. No warmup period before sending; warmup runs continuously as part of each mailbox's daily budget, so the pool is ready as soon as the list is.
Link the pool in Apollo under Settings → Mailboxes
Connect the ColdRelay mailboxes in Apollo via Settings → Mailboxes so sequences can rotate across the pool. Set each mailbox's daily send limit in Apollo to 2 outbound emails per day, matching ColdRelay's per-mailbox budget of 4 sends/day total — 2 outbound + 2 warmup. Don't be tempted to raise the per-mailbox limit when the list is large; add mailboxes instead, and let the budget protect the placement rate your list quality depends on.
Build a multi-step sequence from the persona, then launch
Create the client's sequence in Apollo with the saved persona as the source, mixing email steps with call tasks and LinkedIn touches — multi-channel sequences are where Apollo earns its keep for setting work, since a prospect who saw a connection request answers the email at a higher rate. Because every contact came through the persona filters, the copy can reference the ICP traits directly instead of hedging with generic openers.
The List-Owning Setter's Apollo Playbook
Sell list ownership as the service, not a line item
Most setting agencies position on sends and meetings; position on the data instead. 'We don't send to bought lists — we build yours from your closed-won pattern and maintain it monthly' is a pitch the agency blasting a purchased CSV can't make. It justifies a higher per-meeting price, because the client is buying qualified meetings, and the qualification happened before the first email was written.
Refresh personas from meeting outcomes, not just monthly reruns
The saved persona is a hypothesis; every meeting is a data point. Each month, compare which Apollo filter segments produced meetings that progressed in the client's pipeline versus ones that stalled, and tighten the persona accordingly — drop the title that books-but-never-buys, widen the industry that quietly overperforms. Agencies that close this loop send to a better list every month on the same infrastructure; agencies that don't are just rerunning their first guess.
Size the pool to the persona, not the other way around
List-owning agencies have a discipline blast shops don't: the persona caps the list. If a client's saved search yields 4,000 matching contacts, a 30-mailbox pool at 2 outbound sends/day per mailbox (4/day total with 2 warmup) works through it in about two months with multi-step sequences — right-sized. Provisioning 100 mailboxes against that persona just pressures someone to loosen the filters to feed the capacity, which is how good lists die. Scale the pool when the persona genuinely supports it.
Make data quality a reported metric, alongside meetings
Report bounce rate and meeting-to-opportunity conversion to the client every month, next to meetings booked. Low bounces prove the list is real; meetings that turn into pipeline prove it's right. When renewals come up, the agency that can show 'under 2% bounce, 40% of our meetings reached your proposal stage' is defending the data layer it owns — not just a meetings count the client can get quoted cheaper elsewhere.
Typical Appointment-Setting Benchmarks (Apollo + ColdRelay)
| Metric | Benchmark | Notes |
|---|---|---|
| Inbox placement rate | 95%+ | Dedicated IPs and isolated tenants — the multiplier on every hour of persona work |
| Bounce rate on persona-built lists | <2-3% | Apollo-sourced contacts via saved personas, versus 5-10% typical on purchased third-party lists |
| Meeting-to-opportunity conversion | 30-50% | The list-quality metric — meetings from closed-won-modeled personas progress; blast-list meetings stall |
| Outbound capacity per mailbox | 2/day | 4 sends/day total per mailbox — 2 outbound + 2 warmup |
| Time from ICP analysis to first send | 2-4 days | Persona build is the long pole; the ColdRelay pool provisions in about an hour |
What It Costs: Apollo + ColdRelay
You pay per mailbox per month for the infrastructure, with volume tiers that drop as you scale (see the table below). Because list-owning agencies size pools to personas, capacity tends to grow deliberately with the client roster — and total mailbox count across clients is what moves you down the tiers. DNS, dedicated IPs, and isolated Azure tenants are included.
Apollo is billed separately on its own per-seat plans, which cover the contact database, export credits, saved personas, and sequences. For a list-owning agency it's effectively your data subscription and sending software in one bill.
The useful framing is cost per qualified meeting: Apollo's seats and credits buy the list quality, ColdRelay's mailboxes buy the placement, and both are fixed and predictable per client. Since persona-built lists convert meetings to pipeline at a higher rate, the same infrastructure spend produces meetings worth defending at renewal — which is where the margin in this model actually lives.
| Mailboxes | ColdRelay price / mailbox / month |
|---|---|
| 1–199 | $1.00 |
| 200–999 | $0.85 |
| 1,000–4,999 | $0.70 |
| 5,000+ | $0.55 |
Each mailbox sends 4 emails per day — 2 outbound to prospects + 2 warmup. ColdRelay provisions mailboxes on isolated Azure tenants with dedicated IPs; Apollo handles the sending, sequencing, and inbox rotation on top.
Frequently Asked Questions
Is ColdRelay a competitor to Apollo?
No — they're complementary layers of the same stack. Apollo is the data and sequencing layer: the B2B contact database, saved personas, and multi-channel sequences. ColdRelay is the infrastructure layer underneath: the secondary domains, mailboxes, and dedicated IPs that Apollo sends from. A list-owning agency uses both — Apollo to build and sequence each client's list, ColdRelay to provide the mailboxes it sends through.
Why not just send from the mailboxes a client already has?
Because the client's primary domain carries their billing, support, and sales email — and cold volume on it puts all of that at risk. ColdRelay provisions fresh secondary domains and mailboxes on isolated Azure tenants with dedicated IPs, so the outbound you run for a client is completely walled off from the domain their business depends on. You link those mailboxes in Apollo under Settings → Mailboxes and sequence as normal.
Does building lists in Apollo instead of buying them actually change deliverability?
Meaningfully, yes. Purchased third-party lists routinely bounce at 5-10%, and bounces damage sender reputation no matter how good the infrastructure is. Contacts sourced through Apollo's database via a saved persona typically bounce under 2-3%, which protects the 95%+ inbox placement your ColdRelay pool delivers. Infrastructure and data quality multiply — owning both layers is the point.
How many mailboxes should I provision per client when the persona defines the list?
Size the pool to the persona's contact count, not a generic template. Each mailbox sends 2 outbound emails/day (4/day total including 2 warmup), so a 30-mailbox pool delivers ~1,800 outbound sends a month — enough to work a 3,000-5,000 contact persona through a multi-step sequence over a quarter. If a client's saved search supports more, scale up: ColdRelay fits 100-150 mailboxes per domain and new capacity provisions in about an hour.