Disclosure: To ensure complete honesty and independence in our analysis of when multisite setups fail, please note this article contains affiliate links that may earn us co
Multisite gets pitched constantly as the obvious answer for agencies managing several client WordPress sites — one WordPress core to update, one dashboard to log into. It’s a real advantage in specific situations and a genuine liability in others. Here’s when it’s the wrong call.
All sites in a multisite network share the same WordPress core files, the same server resources, and typically the same database. If one client’s site gets compromised through a vulnerable plugin, or a bad piece of code causes a fatal PHP error, the entire network can go down — not just the affected site. For an agency with clients in regulated industries, e-commerce stores processing payments, or any client who’d be seriously hurt by unplanned downtime, that shared blast radius is a real risk that standalone installs don’t carry.
Network-activating a plugin turns it on for every site at once. If one client’s site has a plugin conflict, a fatal error can appear on every site in the network the moment the conflicting plugin is activated. Agencies that need different plugin sets per client — which is most agencies, since clients rarely have identical needs — end up managing per-site plugin activation individually anyway, which erodes a lot of the “one dashboard, less maintenance” promise multisite is sold on.
PHP memory limits, max execution time, and database connection limits apply to the server or hosting plan as a whole, not per site within the network. A traffic spike on one client’s site (a viral post, an ad campaign, a bot crawl) can slow down or exhaust resources for every other site sharing that install. On standalone hosting, that same spike only affects the one site. If your clients include anyone running paid traffic or expecting unpredictable spikes, that shared resource pool is a liability.
Agencies eventually lose clients, and clients sometimes want to take their site and move to a different host or hand it to an in-house team. Extracting one site cleanly out of a multisite network — files, database tables, users, uploads — requires a plugin like WP Migrate DB Pro or manual export/import work, and it’s noticeably more involved than handing over a standalone WordPress install’s database and files. If client churn or eventual site handoff is a normal part of your business, standalone installs make offboarding far less friction for you and the client.
None of this means multisite is bad — it’s excellent for the right shape of project: a single brand running multiple regional or language sites, a franchise network with a shared parent company, an agency’s own internal site portfolio (staging environments, marketing microsites, internal tools) where one team controls everything and downtime risk is tolerable. The distinguishing factor is ownership and blast-radius tolerance — multisite fits when one entity controls all the sites and accepts shared risk in exchange for centralized management.
| Situation | Multisite fit |
|---|---|
| Multiple independent client businesses | Poor — use standalone installs |
| One brand’s regional/language sites | Good |
| Franchise network, shared parent company | Good |
| Clients who may churn or take their site elsewhere | Poor |
| Agency’s own internal site portfolio | Good |
For agencies who want centralized management without shared-server risk, tools like MainWP or ManageWP let you administer updates, backups, and security across many separate standalone WordPress installs from one dashboard — each site still runs on its own resources and can be isolated, backed up, or handed off independently, while you still get the “one place to manage everything” benefit multisite is usually chosen for.
[AFFILIATE CTA: MainWP]
A few concrete warning signs that a multisite network has become a liability rather than a convenience: you’ve started avoiding plugin updates because the last one broke a site you weren’t watching; a client has asked to move to their own hosting and the export process is taking days instead of hours; you’ve had a “one site’s traffic spike slowed down everyone else” incident more than once; or you’re manually tracking which plugins are safe to network-activate versus which need to stay per-site because compatibility has become unpredictable. Any one of these on its own is manageable. Multiple at once usually means it’s time to plan a migration off shared infrastructure for the highest-risk sites, even if the rest of the network stays as-is.
Migrating a single site off a multisite network isn’t instant, but it’s a known, well-documented process: export that site’s content via WordPress’s built-in exporter or a plugin like WP All Export, spin up a fresh standalone WordPress install, import the content, then manually recreate menus, widgets, and any settings that don’t travel with a standard export (theme customizer settings, some plugin configurations). Budget a half-day to a full day per site depending on complexity, and always do it on a staging copy first so the live site keeps running during the move.
Is multisite less secure than standalone WordPress?
Not inherently less secure per site, but a vulnerability that gets exploited can have network-wide consequences instead of being contained to one site.
Can I run a hybrid — some clients on multisite, some standalone?
Yes, and it’s common. Put low-risk, similar-needs sites (an agency’s own microsites, for example) on one multisite network, and put anything client-critical or likely to need independent handling on standalone installs.
Does multisite save real money on hosting?
Sometimes, since one server handles many sites, but the savings shrink once you account for the extra care needed to isolate resource-heavy sites, plus the added recovery complexity if something goes wrong network-wide.
Related reading: Multisite Plugin Management · Migrating From Single Sites to Multisite · Multisite Setup for Subdomains vs Subfolders