Email deliverability
Most outbound email fails before a single prospect reads it. The sending tool reports delivery, but the message lands in spam or gets silently dropped. For private equity operating teams managing multiple portfolio companies, the problem compounds. Different domains, different reputations, different failure modes. This page describes how we run deliverability as an operating workstream across a base of warmed mailboxes and sending domains, what we can control, what we cannot, and where the usual advice breaks down.
Most companies treat deliverability as a configuration step inside a sending tool. They authenticate a domain, set up a DMARC record, and assume the problem is solved. It is not. Deliverability is a continuous operating discipline that sits upstream of every outbound campaign. It involves domain reputation management, mailbox warming schedules, seed list monitoring, placement checks, list hygiene, suppression management, and the daily work of adjusting volumes and cadences when signals shift.
The sending tool is not your friend here. Its dashboard reports delivery, but delivery means the receiving server accepted the message, not that it reached the inbox. A campaign can show a high delivery rate while inbox placement sits below forty per cent. The tool has no incentive to surface this. We do, because we operate the infrastructure and we see the actual reply rates and bounce patterns that reveal placement problems before the client's pipeline goes quiet.
For private equity operating teams, the challenge is multiplied. A portfolio company may inherit a neglected domain with poor reputation, or a domain that was used for marketing blasts years ago and still carries the scars. We start every engagement with a forensic look at the domain's sending history, authentication posture, and current reputation across major mailbox providers. If the domain is compromised, we recommend a fresh sending domain and a deliberate warming path. If it is salvageable, we build a recovery schedule. Either way, the work is operational, not strategic.
We run owned infrastructure, not a rented platform. That distinction matters for deliverability. When you send through a shared platform, your reputation is pooled with every other customer on that IP range. One sender's poor practices can degrade placement for everyone. We operate dedicated mailboxes on dedicated domains, each with its own authentication records, its own sending history, and its own reputation trajectory.
The operating base is five hundred and sixty-seven warmed mailboxes across one hundred and sixteen sending domains. Each domain has per-domain SPF, DKIM and DMARC authentication. We do not send from a single domain with many aliases. We send from distinct domains, each with its own identity and its own relationship with mailbox providers. This segmentation limits blast radius. If one domain encounters a reputation issue, the rest of the fleet continues unaffected.
Warming is not a one-time event. A mailbox that sits idle for two weeks loses its sending reputation. We warm every mailbox daily, sending low volumes of innocuous messages to seed addresses that report back on placement. The warming schedule adjusts based on what the seed data shows. If Gmail starts soft-bouncing, we reduce volume to that provider and investigate. If Microsoft is placing messages in focused inbox, we maintain the ramp. This is daily operational work, not a script that runs on autopilot.
Every deliverability guide starts with SPF, DKIM and DMARC. Those are table stakes. A properly authenticated domain tells the receiving server that the message is authorised, but it says nothing about whether the message is wanted. Authentication prevents spoofing. It does not build reputation. We set up authentication correctly on every domain, and then we move on to the work that actually determines placement.
Alignment is the next layer. DMARC alignment ensures that the domain in the from address matches the domain in the return path and the DKIM signature. Misalignment is a common failure mode, especially when a company uses a third-party sending tool that signs with its own DKIM key. The receiving server sees a mismatch and treats the message as suspicious. We enforce strict alignment on every domain we operate. The from domain, the return path domain, and the DKIM signing domain are the same. This is basic hygiene, but we find it missing on a surprising number of domains that arrive in our queue.
Beyond authentication, we configure BIMI records where the domain has a verified brand mark. We publish MTA-STS policies to enforce TLS encryption in transit. We monitor TLS reporting to catch downgrade attacks and misconfigured relays. None of this guarantees inbox placement, but it removes the technical objections that mailbox providers use to filter mail. The goal is to leave no reason for a receiving server to reject the message other than recipient behaviour. And recipient behaviour is what we manage through list quality, sending cadence, and content that does not trigger spam classifiers.
We seed every campaign with addresses at major mailbox providers and a panel of smaller ones. These seed addresses do not engage. They exist solely to report where the message landed. Was it placed in the primary inbox, the promotions tab, the spam folder, or was it silently dropped. We run these checks before a campaign launches and at intervals throughout its life. The data tells us when a provider is shifting its filtering behaviour, often before the client sees any change in reply rates.
The sending tool's dashboard is not a placement monitor. It reports delivery, which is a protocol-level acknowledgement. A message can be delivered and immediately routed to spam. The tool counts that as a success. We do not. We correlate placement data with reply rates, bounce codes, and spam complaint rates to build a picture of what is actually happening. When placement degrades, we do not just reduce volume. We investigate the cause. Is it a content pattern triggering filters. Is it a list quality issue. Is it a competitor reporting the messages as spam. Each cause requires a different response.
We also monitor blocklist appearances across the major DNS blocklists. A single listing can crater placement across multiple providers. We check blocklists daily for every sending domain. If a domain appears, we triage immediately. Some blocklists are self-correcting after a few days of clean sending. Others require manual delisting requests with evidence of remediation. We handle both paths. The operating team does not need to know the difference. They need to know that someone is watching and responding.
A sending domain's reputation is largely determined by how recipients treat its messages. If enough people mark a message as spam, or if enough messages bounce because the address is invalid, the domain's reputation tanks. The fastest way to kill a domain is to send to dirty lists. We treat list hygiene as a first-class operating workstream, not a pre-campaign checkbox.
Before any campaign launches, we run the prospect list through validation. We check syntax, domain validity, and mailbox existence. We remove role-based addresses, known spam traps, and addresses that have hard-bounced on previous campaigns. We suppress addresses that have unsubscribed, complained, or gone silent across multiple touches. This suppression list is maintained across campaigns and across portfolio companies when the client requests it. A prospect who unsubscribes from one portfolio company's outreach does not get contacted by another.
During the campaign, we monitor bounce rates by domain and by mailbox provider. A hard bounce rate above two per cent is a warning signal. Above five per cent and the domain is at risk of being throttled or blocked. We pause sending to problematic lists, revalidate, and resume with a reduced volume. We also monitor spam complaint rates through feedback loops with major providers. A complaint rate above one tenth of one per cent is actionable. We investigate the campaign content, the list source, and the sending cadence to identify what is driving complaints. Then we fix it before the domain takes permanent damage.
List hygiene is tedious work. It involves parsing bounce messages, updating suppression files, and cross-referencing engagement data. Most sending teams skip it because the tool does not force them to do it. We do it because we have seen too many domains burn down from neglect. A domain that takes six weeks to warm can be destroyed in a single weekend of sending to a stale list. The operating discipline is to never let that happen.
A new sending domain cannot send high volumes on day one. Mailbox providers treat unknown domains with suspicion. The correct approach is a deliberate ramp. Start with a small number of messages per day to engaged recipients. Increase volume gradually as the domain builds a positive sending history. The ramp schedule depends on the domain's age, its authentication posture, and the provider's current filtering behaviour.
We typically start a new domain at ten to twenty messages per day, sent to seed addresses and known contacts who are likely to engage. Over the following weeks, we increase volume in increments, watching placement data and bounce rates at each step. A domain might reach a few hundred messages per day after four weeks, and full volume after six to eight weeks. The schedule is not fixed. It adjusts based on what the data shows. If placement degrades at a particular volume threshold, we hold or reduce until the signal improves.
This patience is hard for operating teams who want pipeline immediately. We understand the pressure. But the alternative is to blast at volume from day one, get placed in spam, and burn the domain before it has a chance to build reputation. A burned domain is harder to rehabilitate than a new domain is to warm. We would rather spend eight weeks warming correctly than spend twelve weeks recovering a domain that was rushed. We explain this trade-off to every client at the start of an engagement. Most agree. Those who insist on speed usually come back after a domain fails.
For portfolio companies with existing domains, the ramp logic is different. We assess the domain's current reputation and sending volume. If the domain is healthy, we may be able to increase volume more quickly. If it shows signs of strain, we may need to reduce volume temporarily to let reputation recover. The key is that volume decisions are driven by data, not by a campaign calendar. We send as much as the domain can support, not as much as the sales team wants.
Most failed outbound campaigns do not fail spectacularly. They degrade quietly over weeks. Reply rates drift down. The sending tool still shows delivery. The team assumes the market is not interested. In reality, the domain's reputation has eroded and messages are landing in spam or promotions tabs. By the time someone notices, the domain is damaged and the pipeline is empty.
One common failure mode is list fatigue. The same list is mailed repeatedly with the same messaging. Engagement drops. Mailbox providers interpret low engagement as a signal that the mail is unwanted and begin filtering more aggressively. The fix is to rotate lists, refresh messaging, and suppress unengaged addresses before the providers do it for you. Another failure mode is content that triggers spam classifiers. Certain words, formatting patterns, and image-to-text ratios can cause filtering even when the domain reputation is strong. We review content before launch and adjust where necessary.
A third failure mode is infrastructure neglect. DMARC reports go unmonitored. TLS certificates expire. Forwarding rules break. These are not deliverability problems in themselves, but they create conditions where deliverability degrades. We monitor the infrastructure layer continuously. When a certificate is approaching expiry, we renew it. When a DMARC report shows unauthorised sending, we investigate. The work is unglamorous, but it prevents the slow decay that kills campaigns.
The hardest failure mode to detect is competitive spam reporting. A competitor receives your outreach, marks it as spam, and moves on. One report does little damage. A pattern of reports from the same domain or IP range can trigger filtering. We monitor complaint patterns for anomalies. If a particular domain is generating elevated complaints, we investigate whether the list includes competitors and adjust suppression accordingly. This is not a perfect defence, but it is a signal we track.
No provider can guarantee inbox placement. Mailbox providers control their own filtering algorithms, and those algorithms change without notice. A domain that places well today may struggle tomorrow because a provider updated its model. We can monitor, adjust, and remediate, but we cannot promise a specific placement rate. Any provider who does is either misleading you or defining placement in a way that excludes the failures.
We also cannot fix a fundamentally bad list. If the prospect data is stale, purchased, or scraped from sources with poor hygiene, no amount of warming and authentication will produce good results. The list is the input. Deliverability is the pipe. A clean pipe cannot clean dirty water. We will tell a client when their list is the problem and recommend they invest in better data before we continue sending. Some clients appreciate this candour. Others do not. We would rather lose an engagement than burn domains on lists that cannot perform.
We cannot override recipient behaviour. If people do not want your email, they will mark it as spam, delete it, or ignore it. Deliverability engineering can get the message to the inbox, but it cannot make the message relevant. That is a content and targeting problem. We can advise on patterns that correlate with higher engagement, but the operating team and the portfolio company leadership own the value proposition. The infrastructure delivers the message. The business delivers the reason to care.
Finally, we cannot operate in regulatory grey areas. We require that every campaign complies with CAN-SPAM, GDPR where applicable, and the mailbox provider's own terms of service. We will not send to lists that were scraped without consent, we will not use deceptive subject lines, and we will not forge headers. These practices might produce short-term results, but they destroy domains and expose the client to legal risk. We are not willing to take that risk, and we advise our clients not to either.
Deliverability does not exist in isolation. It is the foundation layer under cold email infrastructure, outbound calling, and the lead generation work we run for portfolio companies. A well-engineered sending infrastructure with strong deliverability practices feeds clean, placed messages into the top of the pipeline. The calling team then qualifies the replies and the interested prospects. The automation layer routes leads, schedules follow-ups, and writes dispositions back to the client's systems.
When deliverability fails, the entire programme suffers. The calling team has fewer replies to work. The automation layer processes less volume. The portfolio company's pipeline thins out, and the operating team cannot tell whether the problem is the market, the message, or the infrastructure. By treating deliverability as a dedicated operating workstream, we make the infrastructure layer visible and accountable. The operating team knows what is being sent, where it is landing, and what is being done when placement degrades.
This visibility matters for portfolio company boards. When an operating partner reports on a lead generation programme, they need to explain not just the output but the inputs and the assumptions. Our deliverability reporting provides the data to do that. Placement rates by provider, bounce rates by domain, complaint rates, and remediation actions taken. The board can see that the programme is running on a sound technical foundation, and the operating partner can focus the conversation on pipeline quality and conversion, not on whether the emails are being delivered.
Questions
Ignore the sending tool's delivery rate. Track inbox placement rate by major provider, measured through seed addresses. Track hard bounce rate, which should stay below two per cent. Track spam complaint rate, which should stay below one tenth of one per cent. Track reply rate as a trailing indicator of placement health. Track domain reputation scores through tools like Google Postmaster and Microsoft SNDS. And track blocklist appearances. These metrics tell you whether the infrastructure is healthy. Delivery rate tells you nothing useful.
A new domain typically requires six to eight weeks to reach full sending volume. We start at ten to twenty messages per day and increase gradually while monitoring placement and bounce signals. The schedule is not fixed. If a provider shows signs of filtering, we hold or reduce volume until the signal improves. Rushing this process burns the domain and forces a restart. A burned domain is harder to rehabilitate than a new domain is to warm, so the patience is worth the wait.
No. And any provider who says they can is not being honest. Mailbox providers control their own filtering algorithms, and those algorithms change without notice. We can monitor placement, adjust volumes, remediate reputation issues, and maintain strong authentication and list hygiene. These practices maximise the probability of inbox placement, but they do not guarantee it. We report what is actually happening, not what we wish were happening.
Deliverability is not sold as a standalone line item. It is part of the infrastructure and operations workstream that underpins our outbound programmes. The cost is bundled into the engagement, which is scoped based on the number of domains, mailboxes, and campaigns the client needs. We do not charge per email sent. We charge for the operating work of maintaining the infrastructure, warming the mailboxes, monitoring placement, and managing hygiene. An engagement typically starts with an assessment of the client's current sending posture and a recommended scope. Pricing is discussed after that assessment.
A sending tool sends email. It does not manage deliverability. Its dashboard reports delivery, not inbox placement. It does not warm mailboxes daily, monitor blocklists, investigate complaint anomalies, or adjust volumes by provider based on seed data. Those are operating tasks that require someone to do them every day. If your team has the capacity and the expertise to run that workstream, you may not need us. Most operating teams we work with would rather focus on the portfolio company's commercial strategy and leave the infrastructure operations to a team that does nothing else.
Related
Next step
Deliverability is not a configuration step. It is not a feature of a sending tool. It is a daily operating discipline that determines whether your outbound programme produces pipeline or produces silence. We run this discipline across a base of warmed mailboxes and sending domains, with monitoring, hygiene, and remediation work that continues for the life of the engagement. We do not promise inbox placement, because nobody can. We do promise that when placement degrades, someone will notice, someone will investigate, and someone will fix it. For private equity operating teams managing multiple portfolio companies, that is the difference between a programme that reports activity and a programme that reports results. Contact zach@lawlessllm.com to discuss your current sending infrastructure and where the gaps might be.
Start a conversation