Most content about managing multiple inboxes is either someone with three mailboxes acting like they cracked the code, or a tool vendor trying to sell you something. Nobody actually walks through what happens when you scale past 20, 30, 50+ inboxes across dozens of client campaigns and things start breaking in ways you never anticipated. So here's what actually broke for us and what we changed, because the problems are never the ones you expect.

We run a seven-person lead gen agency with 31 active client campaigns right now, hovering around $91K a month in revenue. Been doing this about five years. We manage somewhere between 52 and 58 inboxes at any given time depending on who's onboarding and who's churning. The number fluctuates because we spin up new domains for new clients and sunset old ones on a rolling basis.

HOW THE WORKFLOW ACTUALLY RUNS

The day-to-day looks like this: we build lists from Apollo and LinkedIn Sales Navigator, enrich with Prospeo first and then Clay if we need firmographic stuff layered on, verify everything through ZeroBounce, load into Saleshandy for sending, and track results in a combination of Saleshandy dashboards and a messy but functional Google Sheet that one of my team members built and I honestly don't fully understand anymore. That's the skeleton. The meat is all the stuff that goes wrong between those steps.

WHAT BROKE FIRST: DNS AND INBOX HEALTH MONITORING

When you have eight inboxes it's easy to just check them manually. Open each one, send a test, make sure DKIM and SPF are passing, whatever. When you have 50+, that becomes a full-time job and nobody wants that job. What happened to us about 14 months ago was we had a client campaign running on four inboxes and two of them had their DKIM records silently fail after a domain registrar migration. We didn't catch it for 11 days. By then those inboxes had sent maybe 900 emails with failed authentication and the domain reputation was cooked. We had to burn those domains entirely and spin up new ones, which meant three weeks of warmup before we could send again. The client was not happy. We almost lost them.

What we changed: we built a monitoring script (one of my guys is decent with Python) that checks DNS records for every active domain every morning at 6 AM and posts failures to a Slack channel. Took about two days to build and it's saved us probably four or five times since then. If you don't have someone who can code that, you can use dmarcian or Postmark's tools to monitor, but honestly the custom script is better because it checks exactly what we need and nothing else. The key numbers here are: we check SPF, DKIM, and DMARC for every domain every 24 hours. Any failure triggers an alert. Someone on the team has to acknowledge the alert within two hours during business hours or it escalates to me.

THE BIGGER ISSUE: WARMUP COORDINATION

When you're running 50+ inboxes you always have some percentage in warmup at any given time. For us it's usually six to ten inboxes warming up simultaneously. We use Saleshandy's built-in warmup but the problem isn't the warmup tool, it's tracking which inboxes are ready and which aren't. We had a situation last March where someone on my team started sending a campaign from an inbox that was only nine days into warmup. They sent 45 emails on day one from that inbox. Open rates were fine but we got flagged by Google and that inbox was basically throttled for the next month. The daily send limit we follow is: days one through seven, send nothing but warmup only, about 20 to 30 warmup emails per day. Days eight through fourteen you can do five to eight real sends mixed with warmup. Days fifteen through twenty-one ramp to fifteen to twenty real sends. After day twenty-one you can go up to 30 to 35 per inbox per day max. We never go above 35 from a single inbox. Ever. Even if the inbox feels healthy.

What we changed: we built a spreadsheet (I know, I know) that tracks every inbox with its creation date, warmup start date, current phase, daily send limit, and which client campaign it's assigned to. One person on the team owns this sheet and updates it every Monday. It's not glamorous but it works. We looked at Mailscale and Hypertide for inbox management and honestly Mailscale is fine for spinning up inboxes but it doesn't solve the operational tracking problem. You still need something to tell you which inbox is in which phase. We ended up using Mailforge for bulk inbox creation because the pricing made more sense at our volume, around three bucks an inbox per month versus some competitors charging six to eight, but the management layer is all manual on our end.

THE LIST QUALITY PROBLEM THAT ACTUALLY COST US A CLIENT

About eight months ago we had a client in the insurance space, decent sized account, paying us $4,200 a month. We were pulling lists from Apollo, doing basic verification, and sending. Bounce rates were creeping up to around four and a half to five percent which I knew was high but we were busy and I let it slide. Then one week it spiked to 7.2% and two of the sending domains got blacklisted on Spamhaus. We couldn't recover those domains. The client saw their reply rates drop from around 2.8% to under 1% and they churned within six weeks. That account loss cost us over $25K in annual revenue and it was entirely preventable.

The thing I had completely backwards was thinking verification alone was enough. It isn't. What we do now is a three-step process: first we enrich through Prospeo which gets us valid emails for about 80 to 82% of the contacts on a typical list, then we run everything through ZeroBounce for verification, which costs us around $350 to $400 a month at our volume, and then we run the catch-alls and unknowns through Scrubby which is specifically built for that. Scrubby catches stuff that standard verification misses, especially catch-all domains which are like 30 to 40% of B2B emails depending on the vertical. After implementing this three-step process our bounce rates dropped from that four to five percent range to consistently under 1.8% across all campaigns. Some months it's as low as 1.1%.

The cost breakdown on enrichment and verification alone is roughly: Prospeo runs us about $99 a month on the plan we're on which has been solid for the volume we push through. ZeroBounce is around $350 to $400 a month. Scrubby is $49 a month. And Clay is $149 a month but we use Clay more for data enrichment and conditional logic than for finding emails. Total enrichment and verification spend is somewhere around $650 a month. Sounds like a lot until you realize one blacklisted domain costs you way more than that in lost campaigns and client trust.

SENDING INFRASTRUCTURE

The actual sending side is less complicated than people make it. We use Saleshandy for almost everything. Tried Lemlist for about three months in early 2024 and it was fine but the multi-inbox management in Saleshandy is just better for our use case. Lemlist felt more designed for someone running three to five campaigns, not 31. The main thing with sending infrastructure that people mess up is they think about it in terms of tools when they should think about it in terms of rules.

Our rules are simple and we don't break them:

Max 35 sends per inbox per day. Each campaign gets two to four inboxes assigned depending on the client's volume needs. We rotate sending across inboxes within a campaign using Saleshandy's rotation feature. Minimum four-minute delay between sends, we actually set it to five to seven minutes usually. No sending on Sundays. Saturdays are optional and vertical dependent, we send Saturdays for real estate and recruiting clients but not for enterprise SaaS. Every campaign gets a four-step sequence: initial email, follow-up at day three, second follow-up at day seven, breakup at day fourteen. We used to do five and six step sequences and the data showed steps five and six generated almost zero positive replies and occasionally generated angry ones.

HIRING AND PROCESS

This part is less about automation and more about not losing your mind. When it was just me and two other people we could keep everything in our heads. Once we hit five people things started falling apart because everyone had slightly different processes. Someone would skip verification. Someone would set warmup wrong. Someone would forget to update the inbox tracker.

What actually fixed this wasn't a tool, it was documentation. We spent two weeks building SOPs for every single step. List building SOP, enrichment SOP, verification SOP, campaign setup SOP, inbox creation and warmup SOP, weekly health check SOP. Each one has screenshots, exact settings, and the specific numbers I mentioned above. When someone new joins they spend their first week just reading SOPs and shadowing. It's boring but it means by week two they can actually run a campaign without supervision.

The weekly health check is probably the most important ritual we have. Every Monday morning one person spends about 90 minutes going through every active campaign and checking: bounce rate must be under 2%, reply rate, we flag anything under 1% for review, inbox health via our DNS monitoring script, warmup status for any inboxes in rotation, and whether any domains need to be sunset. If a domain has been sending for more than four months we start watching it more closely. After six months we usually retire it and spin up a replacement. Domains are cheap. Reputation is not.

CRM AND REPORTING

We use Folk as our CRM which is maybe an unusual choice but Salesforce was overkill for what we needed and HubSpot's pricing at our scale was annoying. Folk is like $30 per user per month and it does what we need. Positive replies from Saleshandy get tagged and moved into Folk where our closers, two of the seven people, take over. We track meetings booked, show rate, and pipeline generated per client per month. Average across all clients right now is about 14 to 18 qualified meetings per month per client, with a show rate around 72%. Some clients get more, some get less, depends heavily on the vertical and the offer.

STUFF I WISH SOMEONE HAD TOLD ME

Catch-all domains are not your enemy but they will hurt you if you don't handle them. Roughly 35% of the B2B emails we encounter are on catch-all domains. If you just send to all of them your bounce rate will creep up. If you skip all of them you lose a third of your addressable market. Scrubby is the answer here, it validates catch-alls specifically and has saved us more headaches than I can count.

Don't cheap out on domains. We buy from Namecheap and Porkbun, usually .com or .io, and we set up proper DNS records before we even think about creating inboxes. The whole setup per domain takes about 15 minutes if you know what you're doing. SPF, DKIM, DMARC, custom tracking domain. Skip any of those and you're building on sand.

LinkedIn Sales Navigator is worth the $99 a month even if you never send a single LinkedIn message. The search filters are just better than anything else for building targeted lists. We export from Sales Nav, enrich through Prospeo and Clay, verify, and send. That's the core loop.

Anyway this got way longer than I planned. There's probably stuff I forgot but these are the main things that broke and what we did about them. Managing 50+ inboxes isn't really about finding the right tool, it's about building systems that catch problems before they become expensive. Most of the damage we've taken over five years came from things we didn't notice, not things we did wrong on purpose.