Agency Cold Email, Built on an Apollo Data Supply Chain
Most agency cold email advice starts at the sending step and works backward. This guide starts where the campaigns actually start: the list. Agencies that run Apollo don't buy data — they manufacture it. A client's ICP becomes a saved persona, the persona becomes a filtered search, the search becomes a fresh, signal-qualified list, and that list flows into sequences sent from infrastructure the agency controls. Apollo is the data and sequencing layer; ColdRelay is the layer underneath it — the domains, mailboxes, and dedicated IPs that Apollo's sequences send from.
This guide covers the full supply chain: encoding each client's ICP as a saved persona in Apollo, provisioning a ColdRelay mailbox pool to carry the volume, linking those mailboxes under Apollo's Settings → Mailboxes, and turning the list-building work itself into something clients pay for — because when the agency owns data quality end-to-end, the list stops being a cost center and becomes the moat.
Why Run Apollo Sequences on ColdRelay Infrastructure
Apollo gives an agency something most of the market doesn't have: a B2B contact database with title, industry, and signal filters, plus saved personas and saved searches that make every client's ICP repeatable. That's the input side of the supply chain solved. But Apollo's sequences send from whatever mailboxes you link to it — it doesn't provision domains, configure DNS, or give each client's sending its own IP reputation. The output side of the supply chain is the infrastructure layer's job.
That's where ColdRelay fits. Instead of stretching a handful of Google Workspace seats across your whole client book, you provision mailboxes on isolated Azure tenants with dedicated IPs — SPF, DKIM, and DMARC pre-configured, 100-150 mailboxes per domain, ready in about an hour. Link them in Apollo under Settings → Mailboxes and your carefully built lists go out on infrastructure that matches their quality.
The pairing is additive, not competitive: Apollo is the data and sequencing layer, ColdRelay is the infrastructure layer beneath it. The agencies that win with this stack are the ones that treat both halves as quality problems — a verified, persona-targeted list deserves better than borrowed reputation, and dedicated infrastructure deserves better than a stale purchased CSV.
Visit Apollo →From Client ICP to Live Apollo Sequence on ColdRelay Mailboxes
Encode the client's ICP as a saved persona in Apollo
Before any infrastructure exists, translate the client's ideal customer profile into Apollo's filters — titles, seniority, industry, headcount, and the signal filters that matter for their offer (job changes, funding, tech installed). Save it as a persona and save the search. This is the master record of who the client is paying you to reach, and every list you ever pull for them refreshes from it instead of starting over.
Provision the client's mailbox pool on ColdRelay
Size the pool from the contracted volume and provision it on ColdRelay: secondary domains that echo the client's brand, isolated Azure tenants with dedicated IPs, DNS (SPF, DKIM, DMARC) pre-configured. ColdRelay supports 100-150 mailboxes per domain, and the whole pool is live in about an hour — so the list-building in step one and the infrastructure here can both finish the day the client signs.
Link the ColdRelay mailboxes in Apollo
In Apollo, go to Settings → Mailboxes and link each ColdRelay mailbox via SMTP/IMAP using the credentials exported from the ColdRelay dashboard. Once linked, Apollo can distribute sequence sends across the full pool, and every email step in a sequence sends from infrastructure the agency — not the client's primary domain — owns.
Set per-mailbox daily send limits to match the ColdRelay budget
Apollo lets you set a daily send limit per linked mailbox — set it to 2 outbound emails per day for every ColdRelay mailbox. That mirrors the per-mailbox budget of 4 sends/day total, split 2 outbound + 2 warmup. Warmup runs continuously on ColdRelay's side as part of that budget, so there's no waiting period before sequences can start and nothing extra to configure in Apollo.
Build the sequence with email, call, and LinkedIn steps and launch
Create the client's sequence in Apollo and source contacts straight from their saved persona. Mix step types — email steps carried by the ColdRelay pool, plus call and LinkedIn touches that move the conversation without spending email budget. Because contacts enter from the saved search, every new prospect who matches the ICP flows into the same sequence as the search refreshes.
The Data Supply Chain Playbook: Apollo + ColdRelay for Agencies
Sell list-building as a line item, not a hidden cost
If you're building each client a saved persona, tuning signal filters, and refreshing their search monthly, that's productized work — put it on the invoice. Agencies that itemize list-building (with the persona definition as the deliverable) anchor higher retainers and make renewals stickier, because the client can see the asset being maintained rather than assuming leads appear from nowhere.
Verify before anything touches a mailbox
Your ColdRelay pool's reputation is the asset the whole service runs on, and the list is the only thing that can damage it. Run every Apollo export through verification before it enters a sequence, drop catch-alls and unknowns for new domains, and prune hard bounces the day they happen. A sub-2% bounce rate isn't a vanity metric — it's the maintenance schedule for the infrastructure your margins depend on.
Refresh from saved searches instead of reselling stale data
The cheap-agency failure mode is buying a list once and mailing it for six months. With saved searches, freshness is a button: re-run the client's persona monthly, pull only net-new contacts, and lean on signal filters — job changes, funding rounds, new tech installed — so every batch has a reason to be contacted now. Data recency is the difference between an agency that owns its supply chain and one that resells someone else's.
Spend call and LinkedIn steps to conserve email budget
At 2 outbound emails/day per mailbox (4/day total with 2 warmup), email capacity is the scarce resource — so stop spending it on touches that don't need it. Use Apollo's call and LinkedIn steps for the nudges and follow-ups, and reserve email steps for the messages that carry the offer. A sequence that closes a meeting in 3 email touches instead of 6 effectively doubles the client's capacity without adding a single mailbox.
Typical Agency Benchmarks (Apollo + ColdRelay)
| Metric | Benchmark | Notes |
|---|---|---|
| Inbox placement rate | 95%+ | Dedicated IPs and isolated tenants, fed by verified persona-targeted lists rather than bulk-purchased data |
| Outbound capacity per mailbox | 2/day | 4 sends/day total per mailbox — 2 outbound + 2 warmup, with warmup running continuously |
| Bounce rate on verified Apollo exports | <2% | Verification before sequencing plus same-day pruning of hard bounces; unverified exports routinely run 5-10% |
| Time from client signature to live sequence | Same day | Persona and saved search built in Apollo while the ColdRelay pool provisions in ~60 minutes |
| List-building line-item revenue per client | $500-1,500/mo | Persona maintenance, monthly search refreshes, and verification billed as a deliverable, not absorbed as overhead |
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). DNS, dedicated IPs, and isolated Azure tenants are included — and because your whole client book totals into one bill, multi-client agencies reach the cheaper tiers quickly.
Apollo is billed separately on its own subscription, with database access, export credits, saved personas, and sequence features gated by plan tier — priced per its current plans.
The two bills map to the two halves of the supply chain: Apollo's cost scales with data access and seats, ColdRelay's scales with sending capacity. For agencies charging list-building as a line item, that split makes the unit economics legible — data cost per client on one side, infrastructure cost per client on the other, and the margin on each visible in a spreadsheet.
| 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
Does ColdRelay replace Apollo?
No — they're complementary layers. Apollo provides the contact database, saved personas, and sequences (email, call, and LinkedIn steps); ColdRelay provides the infrastructure those sequences send from — the domains, mailboxes, and dedicated IPs. Agencies use them together: Apollo as the data and sequencing layer, ColdRelay as the infrastructure layer underneath.
Can I link ColdRelay mailboxes directly in Apollo?
Yes. Export the mailbox credentials from the ColdRelay dashboard and link each mailbox in Apollo under Settings → Mailboxes via SMTP/IMAP. Once linked, set each mailbox's daily send limit in Apollo to 2 outbound emails per day to match the ColdRelay budget — 4 sends/day total per mailbox, split 2 outbound + 2 warmup.
Does list quality really affect the infrastructure side?
Directly. Hard bounces and spam complaints are the main inputs to sending reputation, so the list is the one thing that can degrade an otherwise healthy mailbox pool. That's why the supply chain matters: verified exports from a tuned Apollo persona keep bounce rates under 2%, which keeps the dedicated IPs and isolated Azure tenants performing at the 95%+ placement they're built for.
Do I need a warmup period before launching Apollo sequences?
No. ColdRelay mailboxes run continuous warmup as part of the per-mailbox budget — 4 sends/day total, split 2 outbound + 2 warmup — with no waiting period before sequences can start. Provision the pool, link the mailboxes in Apollo, set the per-mailbox limits to 2 outbound/day, and launch the same day the client's persona is built.