Keep it running. Then make it win.
A live product will break, drift, and quietly leave money on the table; the question is who's there when it does. Product Support & Optimization is your standing team for both: fixes, updates and emergencies handled fast and for good, and evidence-led improvements that make the product measurably better, one tested change at a time.
Where it fits
Care · post-launch. This is where the Care loop closes: Radar and Health Check diagnose, this service acts. The urgent things get fixed, the promising things get improved, and both run against your real numbers.
Who it's for
Teams with a live, earning product who need expert hands on call, for the bug at 2pm and the outage at 2am, and who know the product could do more: convert more visitors, lose fewer users, earn more from the same traffic. One team covers both, without standing up in-house ops.
What you get
- Fast, clean bug fixes, reproduced, fixed, and regression-checked
- Incident response with a written postmortem and prevention steps
- Safe updates and security patches, staged and tested, with backups that are actually tested
- A prioritized backlog of improvements, ranked by impact and effort
- Designed and built changes, not just recommendations, with A/B and usability testing that proves what moves the needle
- Measurable lift, reported against your own baseline
Two modes, one team
One team closes the whole Care loop, the reactive half that fixes what breaks, and the proactive half that makes what works measurably better.
Fix, the reactive half
Bugs, incidents, patches, releases with a clean rollback. Handled fast, calmly, and for good.
Improve, the proactive half
A continuous loop of measure, hypothesize, ship, learn, run against your baseline. We don't ship taste; we ship lift.
What we do
Each kind of work is a card, pulled live from our Techniques library.
Bug Fixing
Fast, clean fixes for defects in your live product.
- What
- Functional bugs, broken flows, console/PHP errors and edge-case failures.
- How
- Reproduce, fix, test and ship — with a regression check so it stays fixed.
- Purpose
- A bug left live erodes trust and revenue every day it survives.
- When
- On demand, as issues are reported or found.
Incident Response
Rapid response when something critical breaks.
- What
- Outages, crashes, and data or access emergencies.
- How
- Triage, contain, restore — then a written postmortem with prevention steps.
- Purpose
- In an outage, minutes cost money; a practiced response saves both.
- When
- Immediately, whenever a critical incident hits.
Updates & Patching
Keeping core, plugins and dependencies current and safe.
- What
- Version updates, security patches and compatibility fixes.
- How
- Staged updates with backups and testing before they reach production.
- Purpose
- Outdated dependencies are the most common way products get breached or break.
- When
- Scheduled, plus urgent security patches as they land.
Backup & Recovery
A safety net you can actually fall back on.
- What
- Backup strategy, offsite copies and tested restores.
- How
- Automated backups, periodic restore drills, and a documented recovery path.
- Purpose
- A backup you've never tested is a guess — we make recovery a certainty.
- When
- Ongoing, with restore tests on a schedule.
Security Remediation
Closing the security gaps an audit or scan finds.
- What
- Hardening config, fixing vulnerabilities, rotating keys, locking down access.
- How
- We action security findings in priority order, then verify they're closed.
- Purpose
- Finding a risk only matters if it actually gets fixed.
- When
- As findings arrive from Radar, Health Check or disclosure.
Release Management
Getting changes to production without drama.
- What
- Deployment process, environments, rollbacks and release hygiene.
- How
- Controlled deploys with checks and a clean rollback path.
- Purpose
- Most outages start with a bad deploy — disciplined releases prevent them.
- When
- With every change that ships.
Error & Log Cleanup
Silence the noise and fix what the logs are quietly screaming about.
- What
- PHP warnings, deprecations, and bloated error/debug logs.
- How
- Read the logs, fix the root causes, then keep logs out of the webroot.
- Purpose
- Log noise hides real failures and eats disk — a clean log is a readable one.
- When
- At onboarding and on a recurring cadence.
Broken Page & 500 Fixes
Get broken pages back to a clean 200.
- What
- 500 errors, broken templates and dead links.
- How
- Reproduce, trace the cause, fix, and verify across the affected pages.
- Purpose
- A broken page in the footer or a key flow quietly costs you users.
- When
- On demand, the moment one is found.
Plugin Audit & Cleanup
Fewer, safer, current plugins.
- What
- Risky, abandoned, duplicate or unused plugins.
- How
- Review active vs installed against vulnerability data; remove or replace.
- Purpose
- Every extra plugin is attack surface and weight you don't need.
- When
- At onboarding and at each maintenance pass.
Database Cleanup
Trim the database back to healthy.
- What
- Orphan tables, expired transients, bloated autoload and duplicate post types.
- How
- Safe queries and cleanup with a backup first; measured before and after.
- Purpose
- A bloated database silently slows every page and every query.
- When
- At onboarding and periodically.
File & Disk Cleanup
Reclaim the space leftover junk is eating.
- What
- Legacy migration files, old backups, stray archives and logs.
- How
- Audit disk usage, remove safely, and set quotas/alerts so it doesn't recur.
- Purpose
- A full disk is how sites suddenly 503 — we stop it before it fills.
- When
- At onboarding and whenever disk climbs.
User & Access Cleanup
Make sure only the right people have the keys.
- What
- Test, stale and over-privileged accounts, and admin roles.
- How
- Audit users and roles, remove or downgrade, and document who has what.
- Purpose
- A forgotten admin account is an open door.
- When
- At onboarding and on review.
SSL & Domain Management
Certificates valid, domains pointed, nothing expiring on you.
- What
- SSL issuance and renewal, DNS records and domain configuration.
- How
- Issue and renew certificates, manage DNS, and verify across every domain.
- Purpose
- An expired certificate or misrouted domain takes the whole site down.
- When
- At setup, on renewal, and whenever domains change.
Monitoring & Alert Setup
Put the early-warning system in place.
- What
- Uptime, disk, performance and error monitoring with alerts.
- How
- Configure monitoring (and Radar) so problems page us, not your users.
- Purpose
- You can't fix fast what you find out about late.
- When
- At onboarding — the foundation for everything else.
Migration & Hosting
Move hosts or servers without losing a thing.
- What
- Host and server migrations, environment setup and cutover.
- How
- Stage on a copy, verify, then cut over with backups and a rollback plan.
- Purpose
- Migrations are where data and uptime get lost — we make them boring.
- When
- When you outgrow or leave a host.
Caching Setup
Make the site fast for everyone, cheaply.
- What
- Page cache, object cache, CDN and edge configuration.
- How
- Configure and tune caching to your stack, then verify nothing breaks.
- Purpose
- Caching is the highest-leverage speed win there is.
- When
- At onboarding and after major changes.
Performance Tuning
Fix the things that make a product feel slow.
- What
- Heavy assets, slow queries, render-blocking resources and runaway cron.
- How
- Profile, then fix the worst offenders and measure the gain.
- Purpose
- Speed is conversion, SEO and patience — all at once.
- When
- On demand and after an audit flags it.
Redirects & SEO Hygiene
Keep links, slugs and rankings from quietly breaking.
- What
- 301 redirects, non-standard slugs, broken links and indexing basics.
- How
- Map and implement redirects, fix slugs, and verify nothing 404s or drops.
- Purpose
- Broken links and lost redirects bleed traffic and rankings.
- When
- After migrations and restructures, and on review.
Content & Data Updates
The everyday edits that keep a product current.
- What
- Content changes, data fixes, and small copy and asset updates.
- How
- Make the change, check it in context, and ship it.
- Purpose
- A live product needs constant small upkeep to stay accurate.
- When
- On demand, as requests come in.
Analytics Review
Read what your numbers are already telling you.
- What
- Funnels, drop-offs, event data and the screens people quietly abandon.
- How
- Audit the analytics setup, fix the gaps, then read behavior at scale.
- Purpose
- You can't improve what you never measured properly.
- When
- At onboarding, then continuously.
Conversion Optimization
Turn more of the traffic you already have into customers.
- What
- Signup, checkout and lead flows that lose people partway through.
- How
- Find the drop-off, form a hypothesis, ship the change, measure the lift.
- Purpose
- A point of conversion is worth more than a month of extra traffic.
- When
- Continuously, against your baseline.
A/B Testing
Prove a change works before it becomes permanent.
- What
- Competing variants of a flow, layout, or piece of copy.
- How
- Split traffic, run to significance, keep the winner and document the loser.
- Purpose
- Opinion is cheap; a measured result settles the argument.
- When
- On changes big enough to be worth proving.
Usability Testing
Watch real people use the thing.
- What
- Confusing flows, unclear labels and steps that stall users.
- How
- Task-based sessions with real users, then fixes ranked by how often they hurt.
- Purpose
- Low-traffic products still get evidence, just qualitative instead of statistical.
- When
- Before big bets, and when the data says something is wrong but not why.
Onboarding & In-App Guidance
Get new users to value faster.
- What
- First-run experience, walkthroughs, empty states and in-app guidance.
- How
- Map the path to first value, then remove or explain everything in the way.
- Purpose
- Most churn happens before a user ever sees what the product is for.
- When
- When activation lags signup.
Retention & Journey Mapping
Find where and why users quietly leave.
- What
- Lifecycle stages, repeat usage, and the moments people stop coming back.
- How
- Map the journey against real cohort data, then fix the weakest stage.
- Purpose
- Keeping a user is cheaper than winning a new one.
- When
- Once there's enough usage history to read.
Accessibility Enhancements
Make the product usable by everyone.
- What
- Contrast, keyboard navigation, screen-reader support and heuristics.
- How
- Evaluate against WCAG and usability heuristics, then fix in priority order.
- Purpose
- Accessibility widens the audience and improves the product for everybody.
- When
- At audit, and with every new interface.
UX Writing
Words that remove hesitation instead of adding it.
- What
- Labels, microcopy, error messages and calls to action.
- How
- Rewrite for clarity, then test the version that converts better.
- Purpose
- Copy is the cheapest, fastest lever on conversion there is.
- When
- Alongside any flow we optimize.
Whatever breaks, there's a defined way we fix it. Whatever works, there's a defined way we improve it.
On call, not on hold
Support is only worth its name when someone answers. Critical issues get a fast, human response, triaged, contained, and resolved, with a written postmortem and prevention steps. AI speeds the diagnosis, so the fix starts sooner. No ticket black holes, no waiting on a queue.
Evidence, not opinion
Optimization without data is decoration. Every change starts from a number, a drop-off, a slow step, a screen people quietly abandon, and ends with a measured result against your own baseline. AI reads behavior at scale, surfaces where the next win is hiding, and drafts the experiments worth running. We don't ship taste; we ship lift.
How it runs
The support side runs as an ongoing retainer, a set of hands always on your product, or per-issue when you just need a thing fixed. The improvement side runs as a continuous loop or focused sprints: measure, hypothesize, ship, learn, repeat.
Ongoing (retainer) / per sprint. From $150 / mo.
Backed by 10,000+ screens, 90+ products shipped, and 35M monthly people.
Questions, answered
Critical issues — outages, security, anything blocking users — get a response in under two hours. Everyday fixes are scheduled and shipped on a predictable cadence. Exact response times are set in your plan.
Either. Most teams take an ongoing retainer, a set of hands always on the product, starting at $150 / mo. When you just need one specific thing fixed, we also work per-issue. The improvement side runs as a continuous loop or as focused sprints against a specific goal.
No. Some teams start with fixes only, others come for the growth work. It's one service and one team because the two feed each other: the fixes keep the product trustworthy, and the improvements make it worth more. Most teams end up running both.
We triage, contain and restore as fast as possible, then deliver a written postmortem with the root cause and the steps that stop it recurring. You get the site back and the reason it broke.
By evidence, not opinion. We start from your data, drop-offs, slow steps, abandoned screens, and from findings handed over by Radar or a Health Check, then rank everything by expected impact against effort.
We build them. The recommendation and the implemented change are the same engagement, not a slide deck handed over at the end.
Every change is measured against your own baseline, with A/B or usability testing where it makes sense. If a change doesn't move the number, we say so and move on. You see lift, not promises.
No. High-traffic products get statistical A/B tests; lower-traffic ones get qualitative testing, heuristic and behavioral analysis. The method scales to your numbers, the evidence-led approach doesn't change.
WordPress, headless and custom builds are all supported. We work on your stack, your hosting and your tools, the support model doesn't change, only the specifics.
No, this service stands on its own. But they pair naturally: Radar and Health Check diagnose, and this is where those findings become fixes and improvements.
Start
Tell us what's breaking, what you'd rather never break, and the number you want to move.
Related services
Product Support & Optimization is one part of how we keep products healthy.
