Load Profile
News site hosting — A newsroom's worst day is a cache-hit-ratio problem
Your best day is your hardest day: a story travels, ten thousand readers land inside an hour, and the desk still has to file the follow-up while it is happening.
The short answer
Run WordPress behind a full-page cache on Nitro WordPress resources: a surge is thousands of anonymous readers requesting a handful of URLs, so it should cost you bandwidth rather than PHP — and the figure that decides which one it costs is your cache hit ratio, not your core count.
Below: how to read hit ratio on a live masthead, the three template patterns that quietly leak it, why the admin slows down before the public site ever does, and how to rehearse a spike on staging instead of discovering the answer during one.
By the HostingFast team · Reviewed 24 August 2026
NVMe
Storage under every article
Free
Domain for year one
99.9%
Uptime target we monitor
Flat
Renewal matches order
A news site is two workloads sharing one domain. The public side is a large number of identical requests for a small number of URLs, and every one of them is cacheable. The private side is a dozen journalists inside the admin, where nothing is cacheable by definition and every action competes for the same PHP workers.
Size the server as though those are one workload and you will buy the wrong thing. The spike is absorbed by the cache; the newsroom is carried by workers, database speed and disk. Nitro WordPress is specified at a million monthly visits with 3 cores and 6 GB behind it, and that split is precisely why the two halves are quoted separately.
Read the hit ratio before you read anything else
Take a live article URL and request it twice with curl -sSI, then look for the cache header in the second response. A miss on the repeat call means every reader on the big day executes PHP for a page that has not changed since it was published, and the spike becomes a compute event nobody sized for.
Then go looking for the pages that never cache. On a news template there are usually three: the homepage, if it carries a live-updating block; the section fronts, if pagination arrives as a query string the cache is not told to ignore; and anything the advertising stack rewrites per request. Each one is a leak measured in workers.
Do this on a quiet Tuesday. Hit ratio is the one number you cannot usefully collect for the first time in the middle of an incident.
What actually degrades during a surge is the desk
Editors are logged in, so nothing they touch is served from cache. Autosave fires on a timer for every open post, the media library generates thumbnails on upload, and the revisions table grows all day. When the public spike lands, both sides queue for the same workers, and the report that reaches you is "the admin has gone treacly" rather than "the site is down".
Two cheap corrections. Lengthen the autosave interval, and cap revisions rather than letting them accumulate for years — WP-CLI will clear the backlog on a schedule and keep the table small. Both reduce uncached write pressure at the exact moment there is none to spare.
Watch the effect in the panel's resource graphs rather than by feel. If the worker count sits flat against its ceiling for minutes at a time, that is the thing to fix.
The build we would run for a masthead
Run cached WordPress on Nitro WordPress resources — the big-story day is the business model, so the platform has to treat it as unremarkable.
In our line-up that is the Nitro WordPress plan: SSL, migration and mail already included, renewal charged at the order rate, backups landing every four hours with Imunify360 in front, and an upgrade path so this decision never has to be re-run from zero.
Order an annual plan and the first year of your domain registration costs you nothing.
Rehearse the spike instead of hoping through it
Clone the site to a staging copy, warm the cache, then load test a cached article and an uncached search URL separately. The two curves answer different questions: the cached one tells you about your network, the uncached one tells you how many concurrent readers the PHP layer holds before latency starts climbing.
Run the same pair of tests during the trial. Migrate a real masthead in free, repeat the identical measurements on both platforms, and keep the one whose uncached curve stays flatter for longer. That is the only comparison a publisher actually needs.
Why we print the method
A hosting company telling a publisher to buy a bigger box is the least useful advice in this market. The ratio, the header check and the two admin settings are the parts that matter, and they work on any provider you care to try them on.
On this tier backups land every four hours with Imunify360 in front — a cadence that suits a newsroom rather than a brochure site, where a day-old copy would cost you a day of filed copy.
- Hit ratio is a number, so go and measure it
- A cached surge costs bandwidth, not workers
- Free migration makes the side-by-side test free
- Renewal rate printed beside the order rate
Why HostingFast
Standard on every plan
Sized for the desk, not just the readers
3 cores and 6 GB with 200 GB of NVMe behind them, because the uncached half of a news site is the half that struggles.
Backups every four hours
On this tier the restore point is measured in hours, which is the right unit when a day of filed copy is the asset.
Staging you can load test
Clone the masthead, warm the cache and push traffic at it — rehearsal is a great deal cheaper than discovery.
PHP X-Ray on CloudLinux Pro
When a template change slows the admin down, the profiler names the function instead of leaving you to guess at it.
A company on the record
Vitalcare at Home Ltd, registered in England and Wales — the same due diligence you would run on any supplier.
One rate, order and renewal
A publishing budget carries a fixed line item easily; it absorbs a year-two jump nobody flagged rather less easily.
Prices Side by Side
Our numbers next to the big names
Typical intro and renewal pricing across the market, lined up against ours — including the renewal figure most comparison charts skip.
| Feature | HostingFastMost popular | Typical household-name host | Typical bargain host | Typical loss-leader deal |
|---|---|---|---|---|
| Starting price / mo* | $2.42/mo | $4–$6 | $2–$4 | $1–$3 |
| Renewal price / mo | $2.42/mo | $10–$15 | $8–$12 | $4–$6 |
| Cheapest plan renews at its sign-up price | ||||
| SSL as standard | ||||
| Migration done for you | ||||
| NVMe drives on the cheapest tier | ||||
| Daily backups from the first tier | ||||
| Real humans on support, 24/7 |
*Our number is the lowest-priced plan we sell on an annual term, pulled live from the same catalogue that powers our pricing page — it physically cannot fall out of date. The other three columns show the ranges shared hosts in each bracket typically advertise: introductory offers that mostly demand a one-to-four-year commitment, then jump once it expires. We no longer name rivals or print their prices, because a figure we cannot verify on the day you read it has no business on this page. Line us up against whichever host you are really weighing — and give the renewal row the hardest look.
Quick Start
From order to online
- 1
Collect hit ratio on a quiet day
Request three real article URLs twice each and record which came back cached. That single table tells you what a spike will cost you.
- 2
Fix the uncacheable pages first
Homepage widgets, query-string pagination and per-request advert rewrites are where the ratio leaks. Each one you close removes workers from the critical path.
- 3
Load test staging, then compare hosts
Push traffic at a cached and an uncached URL on both platforms during the refund window, and keep the flatter curve.
Built In
Loaded onto every plan
- Staging copies for rehearsing a change before it ships
- LiteSpeed cache built into the server, not bolted on by plugin
- Full SSH with Git and Composer, not a restricted shell
- WebP conversion handled server-side at no extra cost
- Webmail plus IMAP, POP and SMTP for whichever client the desk uses
- Softaculous on board for one-click application installs
- Per-site PHP version switching from the control panel
- Renewal billed at the rate you ordered at
- Free migration carried out by our engineers
- Money-back cover: 30 days on hosting plans, 7 on reseller
Frequently Asked
What people ask us most often
How do I read cache hit ratio without installing anything?
Two curl requests against the same article and a look at the response headers on the second one. That answers the question for a single URL, which is usually enough to expose the pattern. For a fuller picture, sample a handful of URLs across article, section front and homepage — those three behave differently and only one of them is normally the problem.
How many PHP workers does a newsroom actually need?
Enough that the queue never saturates while editors are filing. The way to find out is to watch the worker graph during an ordinary publishing afternoon rather than during a spike: if it is already flat against the ceiling on a normal day, readers are not what will break you. Uncached admin traffic sets that number.
Does a CDN remove the need for server-side caching?
No, and assuming it does is a common way to get hurt. An edge network will happily cache your HTML once the origin has produced it, but every purge, every logged-in editor and every uncacheable URL still lands on the origin. Server-side caching decides what the origin costs; the edge decides how far the copy has to travel.
What should I measure the day after a spike?
Three things: how many requests were served from cache versus origin, whether the worker queue ever saturated, and how long the admin took to save a post at the peak. Those numbers turn a stressful afternoon into a capacity plan — and they only exist if you were collecting them before it started.
Keep reading
Hosting With Cron Jobs
Scheduled work that has to fire on time — imports, feeds and the jobs a newsroom quietly depends on.
cPanel Hosting Plans
The panel most publishing teams already know, with nothing unfamiliar bolted on top of it.
Uptime (Glossary)
What an uptime figure covers, what it excludes, and how it is actually monitored.
WordPress Hosting
Managed WordPress with LiteSpeed, Redis object caching and staging copies included.
Web Hosting
cPanel hosting on NVMe with SSL, free migration and a year-one domain included.
Changing hosts? Run through our checklist first.
A straightforward sequence for a switch your visitors never feel: which files move first, how to shift email across without losing a single message, the right moment to repoint DNS, and the two mistakes behind almost all the downtime we get asked to rescue.
Build for the big day.
Cached pages served from NVMe, staging to rehearse on, and humans answering at the hour a story actually breaks.
View WordPress Hosting plans