How long does deliverability work take?

The records are usually a day of work plus DNS propagation. Reputation is the slow part. Gmail restores mitigation eligibility for a bulk sender once the user-reported spam rate stays below 0.3% for seven consecutive days, and a domain that has fallen to Low or Bad needs weeks of clean, authenticated sending at a steady volume before it recovers.

Can I fix email deliverability myself?

Often yes. Publishing SPF, DKIM and DMARC from your vendor's own instructions, finding the exact bounce code and reading the headers of one real failing message solves a large share of cases. Get help when authentication already passes and mail still filters, when a blocklist is involved, or when DMARC has to move to enforcement without blocking your own invoices.

Is a DMARC monitoring tool enough on its own?

A tool tells you who sends as your domain and whether they align, which is the hard half of the data. It will not decide which senders to keep, fix a vendor's DKIM signing, or write a delisting request that holds. Start with reports at p=none, read two weeks of them, then decide whether the remaining work is yours or someone else's.

Email deliverability consultant: what they do, when to hire

Say a 25-person logistics company switches invoicing software. Two weeks later the accounts team notices that half the invoices never get a reply, a customer says they found theirs in spam, and one supplier's mailbox is rejecting them outright. Nothing on the sending side looks broken: the software reports delivered, and a test to a colleague's Gmail arrives in a second.

That gap between "we sent it" and "they read it" is what deliverability work is. Most of it is not mysterious. It is DNS, then reputation, then list behaviour, in that order, and a fair amount of it you can check yourself in an afternoon. The rest is where an outside pair of eyes earns its keep, because the evidence lives in bounce strings, message headers and DMARC reports that nobody on a 25-person team has had a reason to read before.

The situations that justify bringing someone in

Outside help is worth paying for when the problem has an order of operations and getting the order wrong costs you weeks. Reputation recovery is the clearest case: raise volume before authentication is clean and you spend the recovery window digging the hole deeper. Blocklists are the second, because a delisting request that does not name the cause and the fix tends to get declined, and on some lists the entry comes straight back.

Use this to work out which bucket you are in.

What you are seeing	What it usually turns out to be	First move
Mail started going to spam right after a new platform began sending	The platform sends as your domain with no aligned `SPF` or `DKIM`	Do it yourself: the vendor publishes the exact DNS records
One recipient domain bounces, everyone else gets the mail	A filter or an address block on that side, or a forwarding chain	Your sending provider first, then get help if it spreads to other domains
Gmail rejects with `550 5.7.1` or `550 5.7.26`, repeatedly	Low domain reputation, or mail arriving unauthenticated	Get help: the fix has a sequence and a waiting period
Your domain or sending address appears on a blocklist	A compromised mailbox, a purchased list, or a noisy neighbour on shared hosting	Get help: a delisting only holds once the cause is gone
A migration moved `MX` and the old system still sends	Two SPF records, or DKIM signing on a domain that no longer matches the From line	Get help if mail is already failing, otherwise plan the cutover

Two things are not deliverability problems, and an honest answer on the first call says so. Open rates that fell while replies stayed steady are usually a measurement change rather than a delivery change. And one message that a single person cannot find is a support ticket, not a project.

What the first diagnosis covers, item by item

This is what we work through on a first pass for a client domain. The order matters, because each item explains the next one.

Every system that sends as your domain. The mail platform, the CRM, the invoicing software, the online store, the helpdesk, the scanner in the corner, the marketing tool somebody signed up for two years ago. This list is almost always longer than the person who hired us expects, and the forgotten entry is usually the one failing.

SPF, as published and as evaluated. One record, not two, and under the lookup ceiling. RFC 7208 section 4.6.4 requires that the include, a, mx, ptr and exists mechanisms plus the redirect modifier total no more than ten DNS queries per evaluation, and an implementation that goes over that limit must return permerror. Three or four vendor includes is enough to cross it, and when it crosses, the record fails for every recipient at once.

DKIM, signing and aligned. Google's sender guidelines ask for keys of at least 1024 bits and recommend 2048, and require the domain in the From line to line up with the SPF domain or the DKIM domain. A platform that signs with its own domain instead of yours passes DKIM and still fails DMARC, which is the single most common surprise in this work.

DMARC, and whether anyone reads the reports. A record at _dmarc with a reporting address is the only way to see who sends as you:

Type: TXT
Host: _dmarc
Value: v=DMARC1; p=none; rua=mailto:dmarc@example.com; adkim=r; aspf=r

Two weeks of those reports usually ends the argument about which system is broken.

Reputation, from the receiver's own numbers. Postmaster Tools rates a sending domain High, Medium, Low or Bad, and the domain dashboard only shows messages sent from the exact domain used for DKIM and SPF authentication, so a separate sending subdomain needs its own verified view. Gmail's sender guidelines say to keep the user-reported spam rate below 0.1% and never let it reach 0.3%. A bulk sender above 0.3% is not eligible for delivery mitigation, and eligibility returns once the rate stays under 0.3% for seven consecutive days.

The thresholds at the other mailbox providers. Microsoft published requirements for high-volume senders into Outlook.com and its other consumer services, effective May 2025: SPF passing, DKIM passing, and DMARC at a minimum of p=none aligned with one of the two. Non-compliant mail was first routed to Junk, and Microsoft later said it would reject it, which shows up as 550 5.7.515.

List and complaint behaviour. Where the addresses came from, how fast the list grew, whether unsubscribes are honoured quickly, and whether marketing and transactional mail leave on the same domain. One purchased list can undo a quarter of careful sending.

Blocklists, and the cause behind the listing. Spamhaus is explicit that the cause of a CSS listing has to be identified and corrected before removal, that an address can be relisted immediately if the problem is still detectable, and that a removal request should state what caused the issue, what was fixed, when it was fixed, and what prevents a repeat. It also says there is never a charge for removing one of its listings, which is worth knowing before you pay anyone who implies otherwise.

What you can check yourself before calling anyone

Do this first. If it resolves the problem you have saved yourself a project, and if it does not, you arrive at the first conversation with the evidence already in hand.

Find the exact bounce string. "It bounced" is not diagnostic. 550 5.7.1, 550 5.7.26 and 550 5.7.515 point at three different fixes.
Send one message to an address at the provider that is filtering you, then read that message's headers. The authentication results tell you what happened to real mail, which is not always what a DNS checker predicts.
Count the references in your SPF record, and confirm there is exactly one record starting v=spf1.
Publish DMARC at p=none with a reporting address and wait two weeks. It costs nothing and it answers the hardest question, which is who else sends as you.
Verify your domain in Postmaster Tools so that reputation and spam-rate history exist when you need them. The data starts when you verify, not retroactively.

Questions to ask before you hire anyone

This field has careful practitioners and it has people who sell a dashboard as a diagnosis. Four questions sort them quickly.

What will you have looked at before you recommend anything? You want to hear the inventory of senders, the published records, the headers of a real failing message and two weeks of DMARC reports. If the answer is a score out of 100, keep looking.

Will you move DMARC to enforcement, and on what evidence? Going to p=reject before every legitimate sender is aligned blocks your own invoices. The right answer involves report data and a staged ramp, not a date on a calendar.

Who touches the DNS, and who keeps the credentials? The records belong in your zone, under your control, documented so the next person understands them. Work that lives inside a vendor's portal is work you cannot audit or take with you.

What happens if the problem is our sending behaviour? Sometimes the records are fine and the list is the problem. A consultant who cannot tell you that is not going to fix your spam rate.

What drives the cost

Scope, not hours. A single domain sending through one platform is a short, bounded job. The work grows with the number of independent systems sending as you, because each one is its own alignment problem. It grows with the number of DNS zones and with how many people control them. It grows when reputation has already fallen, because recovery is measured in weeks of clean sending rather than in effort. And it grows when the sending behaviour has to change and not just the records. A blocklist entry caused by a compromised mailbox adds an incident on top of the project.

Anyone who quotes you before seeing the list of senders is guessing. How we scope work is on how we work.

How Guanacos Tech helps

We are an independent consultancy with Google-certified engineers, and this is the problem we get called about most: invoices in spam, a Gmail rejection that will not go away, a domain on a blocklist, a migration that left two systems sending as you. We run the diagnosis above, hand you the findings whether or not you hire us for the fix, and leave the records in your own DNS, documented, under your control. If that is where you are, email deliverability consulting for small business is what we do, and the 30-minute call is for working out whether you have a records problem, a reputation problem, or neither.

What an email deliverability consultant actually does, and when you need one