The host had removed it. It was on their disallowed list, a routine scan found it, and the platform did exactly what its documentation said it would do. Everyone had agreed to that when they signed up. Nobody had read it.

That incident is the whole managed WordPress hosting vs VPS argument in miniature, and it has almost nothing to do with page speed. Both models can serve a fast WordPress site. They differ on which failures land on you, which land on someone else, and which ones stay invisible until the worst possible moment.

This post compares the two the way I’d compare them for a client: an honest profile of each, the costs that never appear on the invoice, a procedure for modelling the real number, and the decision rules I actually use. No benchmark charts, because your theme and your plugin count will dominate anything the host does.

The real question: who owns the failure at 2am

Most comparisons frame this as control versus convenience. True, but useless, because it doesn’t tell you what to buy. The framing that predicts regret is: when something breaks, whose problem is it, and how fast can they fix it?

Split the failure surface into layers and the answer gets concrete:

• Hardware and hypervisor. Theirs in both models. Your recovery time is still yours.
• OS, web server, PHP, database. Theirs on managed hosting, yours on an unmanaged VPS. This is the big split.
• Backups and restores. Theirs on managed, and tested, because their business depends on it. Yours on a VPS, and almost nobody tests them.
• Core, themes, plugins. Mostly yours either way. Some hosts auto-update, but the compatibility fallout is still yours.
• Your code and your content. Always yours. No host takes this on.

The bottom two rows never move. A managed platform will not make your bloated page builder fast, and it will not stop a vulnerable plugin from being exploited. It moves the middle three rows off your plate and charges you in money and in flexibility.

Managed WordPress hosting, assessed honestly

Where it genuinely wins

The strongest argument isn’t performance. It’s that the boring, high-consequence work gets done by people whose job it is, on a schedule, whether or not you remember.

• Restores that work. One-click recovery from a backup the platform verifies. That’s the difference between a bad hour and a bad week.
• Server-level page caching, already tuned. WP Engine, Kinsta, Pressable and Flywheel run full-page caching in front of PHP. It’s why they ban plugins that try to do the same thing badly.
• Staging that isn’t a science project. Clone, test, push back. On a VPS that’s yours to build, and a half-built staging setup is worse than none.
• Support that knows WordPress. A generic VPS provider’s ticket ends at “the server is up”. A WordPress host will look at your slow query log.
• Edge protection before PHP loads. WAF and bot filtering at the infrastructure layer, the only layer where they meaningfully help. A plugin firewall runs after the request has already cost you a PHP worker.

If nobody on your team wants to own a Linux box, that list is decisive. The rest of this section is about what you’re trading away.

Visit metering is a billing model, not a traffic report

This is the cost that bites hardest, because it doesn’t look like a technical constraint until the invoice arrives.

Most managed plans meter “visits” rather than bandwidth or CPU, and the definition is narrower than it sounds. WP Engine defines a billable visit as one unique IP address logged per day, with static assets like images, CSS and JavaScript excluded, and known and suspected bot user agents filtered out. Kinsta counts the sum of unique IPs seen within each 24-hour period across the plan, and also sells bandwidth-metered plans as an alternative. Flywheel filters IPs belonging to known bots, spammers and attackers before counting.

Three consequences follow, and they’re the ones people miss:

• Your host’s number will not match Google Analytics. Analytics counts sessions from browsers that ran its JavaScript. Your host counts IPs that hit the server. Arguing about the gap is wasted effort.
• Bot filtering is best-effort. Hosts filter what they can identify. A scraper with a browser-shaped user agent and rotating IPs looks, to the meter, like a crowd of new visitors.
• Overages are charged per block of excess visits. Rates change, so check the current figure rather than trusting any article, including this one. What matters is the shape: a viral post or a bad crawl month produces a bill you didn’t budget for.

A VPS has the opposite failure mode. Bad traffic costs you CPU and RAM, not money, right up until it exhausts them and the site falls over. One model converts load into a bill, the other converts it into an outage. Pick the one you’d rather explain to whoever owns the site.

The plugin policy is a real constraint

Managed hosts publish disallowed plugin lists and they enforce them. WP Engine documents that disallowed plugins are found by periodic scans of the site’s filesystem, the owner is notified, and the plugin is removed. That’s not a threat, it’s the operating model, and it’s why the incident I opened with was nobody’s fault.

The banned categories are consistent across vendors, and the reasoning holds up in each case:

• Caching plugins. They fight the platform’s own page cache and produce stale output that support then has to debug.
• Backup plugins. They write large archives onto the same disk the site runs on, which is a useless backup if the disk is what fails.
• Database optimisation plugins. Long-running write-heavy queries on shared infrastructure affect neighbours.
• Some security plugins, or specific features within them. Filesystem-based rule storage and high-volume traffic logging are the usual triggers. Policies vary a lot here.
• Plugins that send mail directly. Platforms push you to a transactional provider instead, for deliverability reasons that are genuinely in your interest.

None of that is unreasonable. The problem is when a plugin your business depends on lands in one of those buckets, or when a host adds something to the list later. Before migrating, read the current disallowed list for that specific host and diff it against your live active plugins. Ten minutes, and it’s the highest-value thing you can do in a hosting evaluation.

You get one shape of application

A managed WordPress platform runs WordPress. If your project also needs a Node service, a Python worker, a queue, or a self-hosted analytics instance, that’s a second bill somewhere else. On a VPS those things are free in cash terms because you already own the machine. They cost you memory and attention instead. Whether that’s a good trade depends on whether you were going to need them anyway.

Running WordPress on your own VPS, assessed honestly

Where it genuinely wins

• Resources per unit of money. Hetzner, InterServer, DigitalOcean and Vultr will sell you more CPU and RAM than an entry managed plan. Whether you can use it is a separate question.
• Many sites, one bill. This is where the economics genuinely flip. Ten small sites on one adequate VPS is one server to patch. Ten managed plans is ten invoices.
• No metering surprises. Traffic spikes hit your load average, not your credit card.
• You choose the stack. Nginx or Apache, PHP-FPM pool sizing, Redis for the object cache, your own MariaDB tuning and firewall rules. If the platform’s opinions were what limited you, this is the fix.
• No lock-in. A tarball and a database dump move anywhere. Platform-specific caching behaviour, redirect rules and deployment workflows do not.

The failure that stays invisible: the backup you never restored

Here’s the one that actually ends projects. Backups on a self-managed VPS fail silently. The cron entry runs, the exit code goes nowhere, the destination bucket quietly rejects writes after a credential rotation, and nothing tells you. You find out the day you need it, which is also the day you have no other option.

The fix isn’t a better tool. Restic and BorgBackup are both excellent and neither will save you. The fix is a restore drill on a schedule, into a scratch directory, where you verify the data is actually there.

# Restore the newest snapshot into a scratch path, never over the live site
restic -r /srv/backups/wp restore latest --target /tmp/restore-drill

# Confirm the dump is real and not a zero-byte file or a truncated
# write from a run that died halfway through
ls -lh /tmp/restore-drill/db/
head -n 20 /tmp/restore-drill/db/site.sql

# Load into a throwaway database and count what came back.
# If wp_posts is empty, your backups have been failing for weeks.
mysql -u drill -p drill_db < /tmp/restore-drill/db/site.sql
mysql -u drill -p -e "SELECT COUNT(*) FROM drill_db.wp_posts;"

Adjust the repository path and database names to your own. The shape is the point: restore somewhere harmless, then check a row count that would be non-zero on a healthy site. A backup you haven’t restored is a hypothesis, not a backup. Uptime checks from something like UptimeRobot tell you the site is down; they tell you nothing about whether you can get it back.

Patching debt compounds quietly

The second invisible cost is the gap between “updates are available” and “updates are applied and the affected services have actually restarted”. Upgrading a package without restarting the process leaves the old, vulnerable code resident in memory. That’s why security scanners and package managers disagree so often.

# RHEL family (AlmaLinux, Rocky, CentOS Stream)
# What security updates are outstanding?
sudo dnf updateinfo list security

# After applying them: does anything still need a restart?
# Exit 0 means no reboot required, exit 1 means reboot required.
sudo dnf needs-restarting -r

# Debian and Ubuntu: what would unattended-upgrades actually do?
sudo unattended-upgrade --dry-run --debug

Run those on any VPS you inherited from someone else. The output is usually educational. Then there’s the WordPress layer, which stays yours regardless of hosting model:

# Do the core files match the official checksums for this version?
# A mismatch means a modified core file or an active compromise.
wp core verify-checksums

# How many plugins are behind?
wp plugin list --update=available --format=count

# Which tables are eating the disk? Usually wp_options, wp_postmeta,
# or a logging plugin nobody remembers installing.
wp db size --tables --format=table

Model the cost before you argue about it

Sticker price comparisons are how people talk themselves into the wrong answer in both directions. This procedure produces a number you can defend.

1. Count your sites. One tilts toward managed. Five or more tilts toward a VPS, because admin time is close to fixed and plan fees are per-site.
2. Estimate your metered visit count from your own logs. If you’re already on a server you control, you can approximate what a metered host would bill you before signing anything.
3. Price the VPS honestly. Instance plus block storage plus offsite backup destination plus any control panel licence. The advertised instance price is rarely the whole line.
4. Put a rate on your own hours. Whatever you’d bill a client, or pay someone else. Zero is not an honest number.
5. Add the one-off build. Standing up a hardened WordPress VPS with TLS, firewall, object cache, tested backups and log rotation is a solid chunk of focused work the first time. Amortise it over twelve months.
6. Add your recovery cost. Hours to rebuild from scratch, times your rate, times how often you honestly think it’ll happen. This term is what makes managed hosting look cheap for a single revenue-generating site.

Step two is worth doing properly, because it’s the number nobody has. On Nginx with the standard combined log format, this gets you close:

# Unique client IPs in the log. A rough upper bound on metered "visits".
awk '{print $1}' /var/log/nginx/access.log | sort -u | wc -l

# Closer to how hosts count: ignore static assets, since most platforms
# exclude images, CSS and JS from billable visits.
awk '$7 !~ /.(css|js|png|jpg|jpeg|gif|svg|webp|woff2?|ico)$/ {print $1}' 
  /var/log/nginx/access.log | sort -u | wc -l

# What is actually hitting you? User agents by request count.
# If the top entries are crawlers, that is your overage risk.
awk -F'"' '{print $6}' /var/log/nginx/access.log | sort | uniq -c | sort -rn | head -20

Field positions assume the combined log format; adjust the column numbers if you’ve customised log_format. The third command is the interesting one. If crawlers dominate your traffic, visit-metered hosting will charge you for whatever the host’s filtering misses, and that’s a risk you can quantify before signing up rather than after.

The middle ground worth looking at first

This gets presented as two options. There are four, and the middle two get skipped far too often.

• Managed VPS. InterServer, Liquid Web and others sell VPS plans where the provider handles OS patching, firewall and backups while you keep root. Read the scope: “managed” often stops at the operating system and leaves the WordPress stack to you.
• Control panel on your own VPS. DirectAdmin, cPanel or CyberPanel gives you WordPress-aware tooling, staging and one-click TLS on infrastructure you still own. You pay a licence and give up some configuration purity.
• Server management platforms. Cloudways, RunCloud, GridPane and SpinupWP provision and manage a stack on a VPS you rent from a provider of your choice. Managed-style workflows without the per-visit meter.
• A CDN in front of anything. Putting Cloudflare or a similar edge ahead of a modest VPS absorbs a large share of the traffic that would otherwise push you onto a bigger plan. Cheapest performance win available to either model, and frequently the actual answer to “we need better hosting”.

If you’re leaning toward a VPS mainly because managed pricing looks steep, look at options two and three before committing to a bare Linux box.

How I’d decide between managed WordPress hosting vs VPS

These are the rules I apply, in order. The first match usually settles it.

• Nobody on the team can or wants to run Linux. Managed, and it isn’t close. An unpatched VPS is worse than the most restrictive managed plan.
• One site, and it generates revenue. Managed. The recovery term dominates everything else in the cost model.
• Five or more sites and someone competent to run them. VPS, or a management platform on top of one.
• You need services WordPress isn’t. VPS. Don’t pay for two platforms to avoid learning one.
• A plugin you can’t remove is on the host’s disallowed list. VPS, or a different managed host. Do not migrate hoping for an exception.
• Traffic is spiky and crawler-heavy. Lean VPS, or pick a bandwidth-metered plan over a visit-metered one. Metering punishes exactly this traffic shape.
• You’re mostly annoyed at the current bill. Neither, yet. Put a CDN in front, cut the plugin count, fix the object cache. Hosting is often not the bottleneck.

Arguments that don’t survive contact

• “A VPS is faster because you get dedicated resources.” Only if it’s configured well. A default LEMP install with no object cache and an untuned PHP-FPM pool loses to a managed platform’s edge cache every time.
• “Managed hosting handles security, so I’m covered.” It handles infrastructure security. The most common route into a WordPress site is a vulnerable plugin, and that’s yours in both models.
• “I’ll just move if I outgrow it.” Migration cost is real and grows with the site. Redirect rules, caching behaviour, deployment workflow and email configuration all need rebuilding.
• “The VPS is only a few dollars a month.” The instance is. Offsite backup storage, monitoring and your own hours are not, and they’re most of the number.
• “Managed hosts back everything up, so I don’t need my own backups.” Retention windows are finite and account access can be lost. Keep an independent copy the host doesn’t control. This applies to both models.

Frequently asked questions

Is a VPS actually cheaper than managed WordPress hosting?

For one site, usually not once you price your own time honestly. For several sites, usually yes, and the gap widens with each site you add. The crossover sits somewhere around three to five sites for most people, depending on what your hours are worth and how much the sites change.

Why does my managed host report more visitors than Google Analytics?

They measure different things. Analytics counts browser sessions that executed its JavaScript, so it misses anything blocking scripts and misses non-browser clients entirely. Your host counts requests reaching the server, deduplicated by IP per day. Bot filtering closes some of the gap but never all of it. The two numbers are not supposed to match.

Can I run other applications alongside WordPress on managed hosting?

Generally no. Managed WordPress platforms run WordPress and nothing else. A background worker, a separate API, a queue or a self-hosted analytics instance needs a VPS or a second platform. Factor that second bill into the comparison.

What happens if I install a disallowed plugin?

It depends on the host, but the documented pattern at WP Engine is that periodic filesystem scans detect it, you get notified, and the plugin is removed. Check the specific host’s current policy and enforcement mechanism before migrating, and check it against your live plugin list rather than what you assume is installed.

Do I still need a caching plugin on a VPS?

You need caching. Whether it’s a plugin is a design choice. The strongest setup is full-page caching at the web server or CDN layer plus a persistent object cache in Redis or Memcached, which handles the repeated database queries page caching can’t touch. A page cache alone leaves logged-in traffic and admin requests hitting the database on every load.

Is managed VPS hosting a real middle ground or just marketing?

It’s real, but the scope varies enormously and the word isn’t standardised. Ask specifically: who applies OS security patches, who restarts services afterwards, who owns backups and where they live, whether restores are tested, and whether the WordPress stack is in scope or the coverage stops at the operating system. Get the answers in writing before comparing prices.

How hard is migrating from managed hosting to a VPS?

Moving files and the database is the easy part. The work is everything the platform did invisibly: redirect rules living in the host’s dashboard, cache purging behaviour, transactional email, TLS renewal, and any deployment workflow tied to the platform. Plan a parallel run with low DNS TTLs rather than a cutover, and keep the old plan alive until the new one has survived a full traffic cycle.

The one thing worth remembering

Managed WordPress hosting vs VPS isn’t a performance question, and it isn’t really a price question either. It’s a question about which failures you’re equipped to own.

Managed hosting converts operational risk into a predictable bill and a set of constraints you have to live inside. A VPS converts that bill into flexibility plus a standing obligation to do the boring work: patch, restart, and prove your restores actually restore. Both are defensible. What isn’t defensible is buying a VPS because it looked cheap and then skipping the work, which is the most common failure in this whole category and the one that ends with a rebuild from a backup nobody ever tested.

Run the restore drill. Whatever you choose.

Need a second opinion on your hosting decision?

Most of the hosting work I get asked for is one of these. If it looks like your situation, I can help:

• Cost modelling before you commit. Working out from your own access logs what a visit-metered plan would actually bill you, against a realistically priced VPS including backup storage and hours.
• Migration in either direction. Managed to VPS or VPS to managed, planned as a parallel run with DNS TTL staging so there’s no blind cutover.
• Building the VPS properly the first time. Nginx or Apache with PHP-FPM sized to the actual RAM, Redis object cache, TLS with automated renewal, firewall and Fail2ban, log rotation that doesn’t fill the disk.
• Backups you’ve actually restored. Offsite destination, retention policy, and a scheduled restore drill that fails loudly instead of silently.
• Plugin policy audits. Checking your live plugin list against a specific host’s disallowed list before migration, and finding replacements for anything that fails.
• Monitoring that answers the right question. Grafana and Prometheus dashboards covering PHP-FPM saturation, database load and cache hit ratio, not just whether the site returns 200.

Send me something concrete and I’ll tell you what I see: an access log sample, your current plan and traffic numbers, your active plugin list, or the output of the commands above. That’s usually enough for a straight answer before either of us talks about scope.

The post Managed WordPress Hosting vs VPS: Who Owns the Failure? appeared first on John Nessime.

Then someone points out that a nightly job has been copying a de-identified extract into that account for eight months, the de-identification script never removed the admission dates, and the account was spun up before anyone thought about the Business Associate Addendum. Nothing was misconfigured. Every control you built worked exactly as designed. And you have been out of compliance the entire time.

That is the shape of most real failures here. Not a breach, not a misconfiguration, but a mismatch between the boundary your tooling checks and the boundary your obligation actually follows. This post covers HIPAA compliance on AWS organised by those gaps: where the contract stops, where encryption stops being a control, what “six years” genuinely applies to, and which parts of the estate people forget are in scope at all.

Eligible is not compliant, and the difference is the whole job

AWS does not sell HIPAA compliance. It sells HIPAA eligible services, which is a genuinely different thing. Eligible means AWS has built the service so it can lawfully handle electronic protected health information and has agreed to cover it under a Business Associate Addendum. Compliant describes an entire system: your architecture, your key management, your access reviews, your policies, your staff, your vendors.

Under the shared responsibility model, AWS secures the infrastructure. You secure everything you build on it. Nothing about signing the BAA transfers a single obligation off your side of the line. An unencrypted RDS instance, an overly broad IAM policy or an application that logs a patient identifier into stdout is your problem in exactly the same way it would be in a rack you own.

People know this in the abstract. Where it bites is in the specifics below.

Gap one: the BAA is a contract boundary, and nothing enforces it

This is the one I would fix first, because it is invisible to every security tool you own.

The AWS BAA is self-service through AWS Artifact, at no extra cost. You can accept it for a single account, or, if you are in the management account of an AWS Organization, accept it once so that existing and future member accounts are covered. That organization-level option is the one worth using, because the per-account version quietly rots: someone creates a new account for a proof of concept, nobody repeats the Artifact step, and six months later that account is running something real.

The second half of the boundary is the HIPAA Eligible Services Reference that AWS publishes. Only services on that list may create, receive, process, maintain or transmit ePHI under the BAA. The list is long, it changes, and some entries carry carve-outs where the service is eligible but a specific feature is not. Reading a service name on the list and assuming every feature inside it is covered is the kind of mistake that only surfaces during an audit.

Here is the part worth internalising: there is no AWS control that stops you putting PHI into a non-eligible service. No API error, no Config rule out of the box, no GuardDuty finding. The eligible services list is a contractual construct. Your infrastructure has no idea it exists.

Turning a contract boundary into a technical one

The mechanism that actually helps is Service Control Policies on the organizational unit that holds your PHI accounts. SCPs set the ceiling on what any principal in those accounts can do, including the root user, so they work as a guardrail rather than a suggestion.

Start with the easy one. Pin the accounts to the Regions you have actually assessed, because data residency assumptions fall apart the moment someone launches something in a Region you never reviewed:

{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Sid": "DenyUnapprovedRegions",
      "Effect": "Deny",
      "NotAction": [
        "iam:*",
        "organizations:*",
        "route53:*",
        "cloudfront:*",
        "support:*",
        "sts:*"
      ],
      "Resource": "*",
      "Condition": {
        "StringNotEquals": {
          "aws:RequestedRegion": ["us-east-1", "us-west-2"]
        }
      }
    }
  ]
}

The NotAction list matters. Global services are backed by endpoints in specific Regions, so denying them wholesale by Region locks you out of IAM and breaks Route 53 and CloudFront. Those entries are exemptions, not an allow-list.

The harder one is restricting which services can be used at all. The same NotAction pattern works, with the services you have approved for PHI listed as the exemptions and everything else denied. It is effective and it is blunt: every new service anyone wants becomes a change request against the policy, and if you forget a dependency you find out through a failure in production. I would only reach for it on a dedicated PHI OU where the workload is well understood, not across a general-purpose organization.

Whichever route you take, write down the approved service list somewhere a human reviews on a schedule, and diff it against the AWS reference periodically. That review is itself a compliance artefact.

Gap two: encryption is a checkbox, the key policy is the control

Almost every guide to HIPAA compliance on AWS tells you to encrypt at rest and in transit. Almost none of them explain why it is worth doing properly rather than minimally, so teams enable default encryption with an AWS-managed key, watch the Config rule turn green, and move on.

The reason to care is the Breach Notification Rule. It applies to unsecured PHI, meaning PHI that has not been rendered unusable, unreadable or indecipherable through a method HHS has specified. HHS guidance points at NIST-validated encryption. If PHI is encrypted to that standard and the decryption keys were not compromised alongside it, an incident involving that data generally does not trigger the notification machinery at all. No individual letters, no HHS portal submission, no press release for a large incident.

Read that second condition again, because it is where the architecture decision lives. The safe harbour depends on the keys not being compromised with the data. If your encryption key is one an attacker inherits automatically the moment they compromise a role in the account, you have encryption but you may not have the argument.

What that means in practice

• Use customer managed KMS keys for anything holding PHI, not AWS-managed keys. Only a customer managed key gives you a key policy you can write, and only a key policy lets you deny decryption independently of the resource policy.
• Separate the key administrators from the key users. The people who can schedule deletion of a key should not be the people whose application role uses it every second.
• Use a distinct key per data domain rather than one key for the whole account. Blast radius and audit trail both improve, and you get the ability to revoke access to one dataset without touching another.
• Constrain key usage with the kms:ViaService condition so a key that exists to encrypt RDS storage cannot be used to decrypt something a role dragged into Lambda.
• Turn on key rotation and leave it on. It costs nothing operationally and it is the kind of thing an assessor asks about by reflex.

Pull the current key policy before you assume it says what you think:

aws kms get-key-policy 
  --key-id alias/phi-rds 
  --policy-name default 
  --output text

# Find storage that slipped through unencrypted
aws ec2 describe-volumes 
  --filters Name=encrypted,Values=false 
  --query 'Volumes[].{Id:VolumeId,AZ:AvailabilityZone}' 
  --output table

aws rds describe-db-instances 
  --query 'DBInstances[?StorageEncrypted==`false`].DBInstanceIdentifier' 
  --output text

The RDS query is the important one, because RDS encryption cannot be enabled in place. If that command returns anything, the fix is a snapshot, an encrypted copy of the snapshot, a restore, and a cutover. Plan for downtime or a replication strategy. This is the single most common “we will fix it later” item I see, and later gets expensive.

Also switch on EBS encryption by default in every Region you use, so the next instance somebody launches from a console wizard is not a new exception:

aws ec2 enable-ebs-encryption-by-default --region us-east-1
aws ec2 get-ebs-encryption-by-default --region us-east-1

Gap three: you have logs, but you may not have evidence

The Security Rule requires audit controls: mechanisms that record and examine activity in systems containing ePHI. It also requires you to regularly review records of information system activity. Both of those are about having and using the records.

Now the correction, because this one is repeated everywhere and it is wrong in a way that costs money. You will read that HIPAA requires six years of audit logs. It does not. The six-year requirement sits in the documentation standard, and it applies to the policies, procedures and records of actions, activities and assessments that the Security Rule requires you to keep, retained for six years from creation or from when the document was last in effect, whichever is later. There is no clause anywhere in the Security Rule that names a retention period for CloudTrail events.

What this actually means is more demanding, not less. You have to decide your log retention period, write it into a policy, justify it against your risk analysis, and then keep that policy for six years. And an assessor will hold you to the number you wrote. Setting a CloudWatch Logs retention of thirty days while your policy claims one year is a finding. Storing seven years of everything because a blog told you to, when your policy says two, is not compliance, it is just a bill.

So: pick a period you can defend, make the infrastructure match it exactly, and treat any gap between policy and configuration as a defect.

Making logs into evidence

Retention is only half of it. The other half is being able to show that the records were not altered. CloudTrail has log file validation for exactly this, and it is off unless you turn it on:

aws cloudtrail update-trail 
  --name org-phi-trail 
  --enable-log-file-validation

# Later, prove a window of logs is intact
aws cloudtrail validate-logs 
  --trail-arn arn:aws:cloudtrail:us-east-1:111122223333:trail/org-phi-trail 
  --start-time "$(date -u -d '90 days ago' +%Y-%m-%dT%H:%M:%SZ)"

With validation enabled, CloudTrail writes signed digest files alongside the log files, and validate-logs checks them. The difference between “here are our logs” and “here are our logs, and here is a cryptographic check that nothing was modified or deleted” is the difference between an assertion and evidence.

Put the archive bucket in a separate account that the workload accounts cannot write to or delete from, and apply S3 Object Lock in compliance mode for the retention window you committed to. Object Lock in compliance mode cannot be shortened or bypassed by anyone, including the root user, which is exactly the property you want and exactly the property that will hurt if you set the period carelessly. Test it in governance mode first.

For the review obligation, a query interface matters more than raw storage. Athena over the CloudTrail bucket is the cheap default. If you want alerting and dashboards on top of access patterns, this is a natural place for a platform such as Grafana, Datadog or Splunk, and any of them will hold access records for you. Just remember that if those records contain PHI, that vendor needs a BAA too. See the subprocessor section below.

Gap four: backups, snapshots and the parts of scope people forget

The Security Rule’s contingency plan standard is not optional decoration. It requires a data backup plan, a disaster recovery plan and an emergency mode operation plan, plus testing and revision procedures. Most teams have the backups. Far fewer have the tested restore, and the tested restore is the part that gets asked about.

Three things routinely go wrong here.

1. Copies leave the boundary. A cross-Region snapshot copy lands in a Region you did not assess. A cross-account copy for the DR account lands somewhere outside the OU your SCPs protect. The data is still PHI. The controls did not travel with it.
2. Re-encryption changes the key, not just the copy. Copying an encrypted snapshot to another account requires a key the destination can use. It is easy to end up with a shared or less restrictive key protecting your backups than protects production, which inverts the risk model.
3. The restore is never rehearsed. A backup you have never restored is a hypothesis. Schedule a restore into an isolated account, record the elapsed time, and file the result. That record is your evidence for the testing requirement, and it is the single easiest compliance artefact to produce for free.

While you are inventorying, remember the places PHI ends up without anyone deciding it should: application logs that include request bodies, database slow query logs capturing parameter values, support tickets with screenshots attached, CSV extracts in an analyst’s bucket, and non-production environments seeded from a production dump. That last one is the classic. If your staging database is a copy of production, staging is in scope, and staging is almost never built to the same standard.

Gap five: the business associate chain does not stop at AWS

Your BAA with AWS covers AWS. It covers nothing else in your stack.

Every vendor that can create, receive, maintain or transmit PHI on your behalf is a business associate and needs an agreement. In a typical AWS estate that means the error tracker holding stack traces, the log aggregation platform, the APM tool, the transactional email provider, the customer support desk, the CI system if it ever touches a production dataset, and any AI or analytics service you have wired in.

Build the inventory as a table with three columns: vendor, what PHI it can see, and whether a signed agreement exists. The third column is usually where the surprises are. Some vendors sign readily, some only on higher-priced tiers, and some decline entirely, at which point you have an architecture decision rather than a procurement one.

One structural move that reduces this surface considerably: keep everything that does not need PHI out of the PHI accounts entirely. Your marketing site, your docs, your status page and your public API gateway for non-clinical traffic do not belong in a regulated account. Running them on ordinary infrastructure, whether that is a separate AWS account, a straightforward VPS from a host like InterServer, or a static site behind Cloudflare, shrinks the estate you have to assess, evidence and defend. Fewer things in scope is the cheapest compliance win available.

For tracking the paperwork side, compliance automation platforms such as Vanta, Drata or Secureframe pull evidence from AWS on a schedule and keep the vendor register current. They are genuinely useful for the collection and reminder burden. They do not design your architecture, and I have seen teams treat a green dashboard in one of those tools as though it were an assessment. It is not. It is a checklist that knows what you told it.

What is changing, and why “addressable” is a bad thing to build on

Since it was adopted, the Security Rule has split implementation specifications into required and addressable. Addressable never meant optional. It meant you assess whether the specification is reasonable and appropriate, and if not, you implement an equivalent alternative or document why neither is necessary. In practice, a lot of organisations turned the documented justification into the deliverable and skipped the control.

HHS published a Notice of Proposed Rulemaking in the Federal Register in January 2025 that would remove that distinction, making implementation specifications required with limited exceptions, and would explicitly require encryption of ePHI at rest and in transit and multi-factor authentication, again with limited exceptions. The comment period closed in March 2025.

Be precise about the status, because a lot of vendor content is not: this is a proposed rule and it is not final. The expected timeline for final action has slipped more than once, and the requirements could still change or be withdrawn. Nobody should be telling you a compliance deadline as though it were settled.

What is worth taking from it is the direction of travel. If your current position depends on having documented that encryption or MFA was not reasonable and appropriate, that position is fragile regardless of what the final rule says. On AWS specifically, encryption at rest and MFA are both cheap and both already best practice. Building the architecture on an addressable deferral is an unforced risk.

Troubleshooting the findings you will actually hit

“An assessor asked which accounts are in BAA scope and nobody could answer”

Sign in to AWS Artifact from the management account and check the organization agreements tab to see whether the BAA was accepted at the organization level or per account. If it is per account, list your accounts, work out which hold PHI, and confirm each one individually. Then move to the organization-level agreement so this question has one answer forever.

“Config says the bucket is encrypted but we cannot prove who read the objects”

Bucket encryption and object-level access logging are unrelated. CloudTrail management events do not record S3 object reads by default. You need CloudTrail data events for that bucket, or S3 server access logging, and both cost money proportional to request volume. Enable data events selectively on the buckets that hold PHI rather than account-wide.

“We enabled an SCP and production broke”

Almost always a Region deny catching a global service endpoint, or a service allow-list missing a dependency the workload calls indirectly. Check CloudTrail for AccessDenied events with an explicit deny from an SCP, and look at the service name in the event rather than the one you expected. Attach new SCPs to a test OU with a representative workload before the PHI OU.

“Snapshot copy to the DR account fails with a KMS error”

The destination account cannot use the source key. The source key policy has to grant the destination principal permission to use it, and the copy has to specify a key the destination can decrypt with. Fix it by granting explicitly on a key you control, not by falling back to an AWS-managed key, which is the tempting shortcut and gives up the key policy control you needed.

“CloudWatch Logs retention was never set”

New log groups default to never expiring, which is both a cost problem and a policy mismatch. Audit them with aws logs describe-log-groups and look for groups with no retentionInDays value, then set the period your policy specifies with aws logs put-retention-policy.

Common mistakes

• Treating the signed BAA as the finish line rather than the prerequisite. It is the thing you need before the first byte of PHI arrives, not evidence that anything is configured correctly.
• Assuming a service is fully eligible because its name appears on the list, without reading the feature-level carve-outs next to it.
• Quoting “six years” as a log retention requirement, then either overspending on storage or writing a policy that contradicts the actual configuration.
• Using AWS-managed KMS keys for PHI, which leaves no key policy to write and no independent revocation path.
• Seeding staging or test environments from production data and then holding those environments to a lower standard.
• Forgetting that the risk analysis is a required, recurring, documented activity, not a one-off spreadsheet from the year you launched.
• Signing a BAA with AWS and none of the ten other vendors that can see the same data.
• Letting a compliance automation dashboard stand in for an architecture review.

Best practices worth the effort

• Isolate PHI in its own accounts and its own OU. Account boundaries are the strongest isolation AWS offers, and they make the scope question answerable in one sentence.
• Accept the BAA at the organization level. It removes an ongoing manual step that fails silently.
• Customer managed keys, one per data domain, with split admin and usage roles. This is what makes the breach safe harbour argument defensible rather than theoretical.
• Write the retention period down first, configure second. Policy and infrastructure should agree exactly, in both directions.
• Ship audit logs to a separate account with Object Lock and CloudTrail validation enabled. Immutability and integrity are what turn logs into evidence.
• Keep PHI out of everything that does not need it. De-identify early, tokenise where you can, and route non-clinical traffic through infrastructure that is not in scope.
• Define everything in Terraform or OpenTofu. A reviewable, version-controlled definition of your controls is worth more to an assessor than any screenshot, and it stops drift being invisible.
• Rehearse the restore and the breach response. Both are required, both are tested by asking for the record, and both are cheap to evidence if you actually do them.

Frequently asked questions

Is AWS HIPAA compliant?

Not on its own, and the phrasing is the problem. AWS offers HIPAA-eligible services and will sign a Business Associate Addendum, which means you can build a compliant system on it. Compliance is a property of your whole environment, including configuration, policies, vendors and staff. No provider can sell it to you as a finished product.

How do I sign a BAA with AWS?

Through AWS Artifact in the console. It is self-service and there is no additional charge. Accept it for an individual account under account agreements, or from the management account of an AWS Organization under organization agreements so all current and future member accounts are covered. It should be accepted by someone with authority to bind your organisation, and it must be in place before any PHI reaches AWS.

Does HIPAA require six years of CloudTrail logs?

No. The six-year requirement is a documentation retention rule covering the policies, procedures and records the Security Rule requires you to maintain, kept for six years from creation or from when they were last in effect. The audit controls standard requires the mechanism to record and examine activity but does not name a retention period for the logs themselves. You set that period in your own policy, justify it, and make the configuration match.

Which AWS services can I use with PHI?

Only those on the AWS HIPAA Eligible Services Reference, and only in accounts covered by your BAA. Check the list before adopting anything new, read the feature-level exclusions noted against individual services, and re-check periodically because entries are added over time. Nothing in AWS will stop you using a non-eligible service with PHI, so this has to be an explicit process on your side.

If encrypted PHI is exposed, do I still have to report a breach?

Generally no, provided the encryption meets the standard in HHS guidance and the decryption keys were not compromised along with the data. The Breach Notification Rule applies to unsecured PHI, and properly encrypted data does not meet that definition. This is why key management, not just enabling encryption, is the part that determines whether the protection is real. You still document the incident and the assessment.

Are the new HIPAA Security Rule requirements in force?

Not at the time of writing. The proposals to make all implementation specifications required and to mandate encryption and multi-factor authentication came from a Notice of Proposed Rulemaking published in January 2025. The comment period has closed, but no final rule has been issued and the timeline has moved. Treat any specific compliance deadline you see quoted with suspicion and check the current status directly.

Is a HIPAA-compliant AWS environment expensive to run?

The controls themselves are mostly cheap. KMS keys, CloudTrail validation, Config rules and account separation cost very little. The real costs are log storage volume, CloudTrail data events on busy buckets, running non-production environments to production standard, and staff time on risk analysis and evidence collection. Reducing what is in scope is the most effective cost lever, which is another reason to keep non-clinical workloads out of the regulated accounts.

The one thing to take away

HIPAA compliance on AWS fails at the seams, not at the controls. Your encryption works. Your IAM policies are tight. What goes wrong is that the obligation follows the data into an account nobody added to the BAA, a Region nobody assessed, a staging database seeded from production, a vendor nobody signed an agreement with, or a retention period nobody wrote down.

So build the boundary technically rather than trusting it contractually. Isolate PHI into its own accounts, wrap those accounts in guardrails that make the contract boundary enforceable, own your keys so the encryption means something legally, and make your logs provable rather than merely present. Then write the whole thing down, because in this domain the documentation genuinely is part of the control.

None of that is exotic engineering. It is ordinary AWS work applied to a boundary that no dashboard draws for you.

Working on a healthcare workload on AWS?

This is the kind of work I do. If you are building or inheriting a PHI environment on AWS, I can help with:

• Scope and boundary review: mapping which accounts, Regions, services and vendors actually touch PHI, and finding the ones nobody knew about.
• Account and OU design with enforceable guardrails: SCPs, organization-level BAA coverage, and Region and service restrictions that hold without breaking your workloads.
• KMS key architecture: customer managed keys per data domain, split administration and usage, and key policies written so the breach safe harbour argument stands up.
• Audit logging that produces evidence: centralised CloudTrail with log file validation, an isolated archive account with Object Lock, and retention that matches your written policy exactly.
• Backup, restore and contingency testing: encrypted cross-account copies that stay inside your boundary, plus rehearsed restores documented as compliance artefacts.
• Terraform or OpenTofu modules for the whole control set, so your posture is reviewable, repeatable and does not drift between audits.

If you would rather start with something concrete than a discovery call, send me a redacted account structure, an SCP that is causing trouble, or the output of a Config or Security Hub run, and I will tell you what I would look at first.

The post HIPAA Compliance on AWS: The Gaps That Pass Every Security Check appeared first on John Nessime.