You have worked out that your emails are not reaching the inbox the way they should, and now you want to put it right. Perhaps open rates have been sliding for months, perhaps a customer has told you their invoice went to spam, or perhaps you ran a health check and the result was worse than you were expecting. Whatever brought you here, the instinct tends to be the same, and it is a good one: fix it, quickly, before it costs any more.
Here is the part almost nobody tells you before they sell you a fix. Most attempts to fix deliverability do not hold, and they fail for a reason that has very little to do with whether the technical changes themselves were correct.
They fail because the diagnosis underneath them was incomplete. Someone changed the thing that was easiest to change, or the thing a tool happened to flag first, without establishing whether it was actually the thing suppressing inbox placement. The changes get made, everyone moves on, and a few weeks later the emails are still landing in spam.
If you take one idea from this piece, let it be that one. Fixing deliverability is not a single action, it is the outcome of understanding where the problem genuinely lives, and the businesses that recover their inbox placement are almost always the ones that resisted the urge to start fixing before they knew what they were fixing.
For anyone who has recently taken over a sending programme, this matters more than it first appears. Every email sitting in a spam folder instead of an inbox is revenue the business has already earned and simply never collected. The onboarding sequence that would have converted a trial, the renewal reminder that would have saved a subscription, the basket-recovery email that would have won back a sale, none of them can do their job from the junk folder.
Fixing deliverability is one of the highest-return things a new Head of Growth can do, because the demand already exists and the only thing standing between it and the revenue is placement. It is recovery work, not a line on a backlog.
Why most email deliverability fixes fail
When placement drops, the common response is to reach for the levers closest to hand: rewriting the subject lines, changing the send time, warming up a fresh IP address, or buying a tool that promises to sort the whole thing out. Each of these can be reasonable in the right circumstance, and each is close to useless in the wrong one, which is most of the time.
The trouble is that they all treat deliverability as though it were one thing with one cause. In reality, whether your email reaches the inbox is the product of a whole system working together, and a change made in one part of that system will do nothing for a problem that lives in another. You can rewrite every subject line in your programme and it will not move the needle at all if the reason you are in spam is a broken authentication record or a list stuffed with contacts who have not opened anything in two years. This is also why we never treat copy as the first answer, and why sending more email to "re-engage" a struggling list usually makes the situation worse rather than better.
To fix deliverability properly, you have to know which part of the system is failing. In practice, deliverability problems live in one of three layers, and the reason so many fixes come apart is that the fix and the fault ended up in different layers.
The three layers where deliverability problems actually live
- The first layer is authentication and infrastructure. This is the plumbing: whether SPF, DKIM and DMARC are correctly configured and aligned across every service that sends on your behalf, whether your sending architecture makes sense, whether your domain or IP has ended up on a blocklist, and whether the way you have split or shared your sending reputation is helping you or quietly hurting you. When this layer is broken, inbox providers cannot reliably confirm that your email is really from you, and mail they cannot verify is mail they are far more willing to filter. If you are not certain your authentication is right, our guide to SPF, DKIM and DMARC walks through what each one does and where they commonly go wrong, and it is worth checking whether your domain has landed on a blocklist before you assume the problem sits anywhere else.
- The second layer is sending behaviour and list quality. This is about what you send, how often, to whom, and how those recipients respond. Sudden spikes in volume, sending to addresses that no longer exist, a complaint rate creeping above the level providers tolerate, and large numbers of dormant contacts all sit here. This is where a great deal of quietly damaging activity happens, because none of it produces an error message to warn you. Your sender reputation is built and eroded in this layer over time, your complaint rate is telling you something specific about the quality of your list, and the dormant subscribers you keep emailing out of habit are actively dragging down placement for the people who do want to hear from you. That last point catches teams off guard, so it is worth stating plainly: inbox providers assess engagement across your whole list as a cohort, which means the dead weight on your list suppresses delivery to your best subscribers.
- The third layer is audience relevance and lifecycle design. This is the hardest to see and the one most tools ignore completely, because it is not a setting you can check. It is whether the right people are receiving the right messages at the right point in their relationship with you. A programme can be perfectly authenticated and reasonably clean and still slide, simply because it sends the same things to everyone regardless of whether they have shown any interest, and inbox providers read that flat lack of engagement as a signal that your mail is not really wanted. Relevance is a deliverability factor, not only a marketing one.
The most useful thing to understand about these three layers is that they interact. A fix in the first layer will not rescue a problem that lives in the second, and a pristine list will not save you if your authentication is failing underneath it. This is exactly why fixing before diagnosing is such a reliable way to spend money and get nowhere: you are choosing a layer to work on before you know which layer is broken.
What fixing each layer actually involves
Diagnose first is the principle, but it is fair to want to know what the fixing itself looks like, so here is the honest shape of it, layer by layer.
When the problem is in the first layer, the fix is technical and usually quick once identified. That might mean correcting an SPF record that has drifted past its lookup limit as tools were bolted on over the years, getting DKIM signing properly aligned for every platform that sends as your domain, moving a DMARC policy that has sat at p=none for two years towards genuine enforcement, or requesting delisting from a blocklist after removing whatever got you listed. These fixes are satisfying because they are concrete, and dangerous for exactly the same reason: they are the ones tools flag and teams reach for first, whether or not the real problem lives here.
When the problem is in the second layer, the fix is behavioural, and it takes longer to show results because reputation is earned back over weeks of consistent sending rather than corrected in an afternoon. It usually means suppressing the contacts who have not engaged in a defined window rather than continuing to mail them out of habit, cleaning out addresses that hard bounce, smoothing out volume so providers see a steady sender rather than a spiky one, and making unsubscribing genuinely easy so frustrated recipients leave quietly instead of hitting the spam button. Counter-intuitively, the most effective fix in this layer is almost always sending to fewer people, not more. Cutting your dormant segment shrinks the list and lifts placement for everyone left on it, which is a trade any commercial team should take.
When the problem is in the third layer, the fix is structural. It means segmenting so different audiences stop receiving identical treatment, rebuilding lifecycle journeys around what recipients have actually done rather than around your send calendar, and in some cases separating transactional and marketing streams so a struggling newsletter cannot drag down the password reset email sent the same day. These fixes take the most thought and deliver the most durable results, because they change the engagement signals providers see rather than patching around them.
The order of operations matters as much as the fixes themselves. Authentication gets corrected first, because nothing else you do will be read fairly while providers cannot verify your mail. Behaviour comes next, because a clean, engaged list is the raw material every other improvement depends on. Structure comes last, built on the foundation the first two created. Run in that sequence, the fixes reinforce each other. Run out of sequence, each one is undermined by the layer beneath it.
Diagnose first, then fix, in that order
None of this makes fixing complicated once you know what you are fixing. It means the sequence matters. You establish what is genuinely wrong, across all three layers, and only then do you fix the specific things that are wrong, working in the order of how much each one is costing you. Reversing that sequence is the single most common reason deliverability fixes fail to hold, and it is worth being honest that a lot of the fixing that goes on in the wider market is really guesswork dressed up as action.
This is where the quality of the diagnosis matters enormously, and where a lot of what is available to UK businesses falls short. Much of it is either an automated tool report, which is decent at checking the first layer and effectively blind to the second and third, or it is a service run out of a US practice with no real feel for the environment UK senders actually operate in. That environment has its own texture. A large share of UK business email is received through Microsoft 365, which filters on different signals to Gmail, so a fix that was only ever validated against Gmail can miss half the picture. UK GDPR and clear consent shape what a healthy list even looks like in the first place. And when something does break, being able to reach someone working your hours, rather than waking up to a problem that has been live and untended all night, is not a small thing.
A proper review reads all three layers together and ranks what it finds by commercial impact, so you are not handed a flat list of forty technical observations and left to work out for yourself which two actually matter. With Digistrat, the review and the fix are a single engagement rather than two separate purchases, which matters more than it might sound. When a diagnosis and its remedy are bought separately, the findings tend to sit in a report waiting for a second procurement cycle while the revenue keeps quietly leaking. Our Deliverability Review and Fix is deliberately built as one service with two phases for that reason: we review first, then scope the fix from what the review actually found, so nothing stalls in the gap between knowing and doing.
What a fix that actually holds looks like
A deliverability fix is not a switch you flip once and forget. It is a point-in-time correction to a system that keeps moving, and it will drift back if nothing is watching it. Inbox providers change their requirements, as Gmail, Yahoo and Microsoft all did recently and will do again. Reputation rises and falls with the way you send. Lists decay quietly as subscribers lose interest and move on. A domain that was placing perfectly well in March can be struggling by June without anyone laying a finger on it, simply because the conditions around it shifted.
This is the second failure mode worth naming, sitting alongside fixing the wrong layer. The first is putting the fix in the wrong place; the second is making the fix once and assuming the job is now finished. The teams whose inbox placement stays healthy are the ones who treat monitoring as part of the fix rather than an optional afterthought, because monitoring is what turns a problem you would otherwise hear about from an angry customer into a problem you catch and correct before it ever reaches your revenue. Knowing that your complaint rate has ticked up this week, or that an unfamiliar sending source has appeared on your domain, or that Microsoft has started treating your mail differently, is the difference between a quiet adjustment and a full-blown emergency. This is the job our ongoing monitoring does, and it is why we treat it as the natural continuation of a fix rather than as a separate idea bolted on at the end.
Where to start fixing your email deliverability
If you suspect your emails are not landing where they should, the honest first step is not to start changing things. It is to find out what is actually wrong, across all three layers, before you spend a single hour or a single pound fixing the wrong one. That is the difference between fixing email deliverability once, properly, and paying to fix it again in six months.
The quickest way to get a first read is our free email health check, which takes under a minute and will tell you whether there is something here worth looking at more closely. If there is, the sensible next move is a proper conversation about it. Book a free check-up with our team and we will walk through your sending situation together, tell you honestly whether you have a problem worth investigating, and explain what fixing it properly would involve.
Book a free check-up 15 minutes · no obligation
