It is 09:40 on a Monday. A technician has a disk alert open in the monitoring tool, the ticket open in the service desk, and the documentation tool open in a third tab because that is where the client’s domain admin password lives. She copies the hostname from the first tab into the second, reads the password out of the third and types it into a remote session. Three products, one logo on the invoice, and the joining done by hand. If that is your Monday too, this post is about why it happens, and about what disappears when the three tabs are one record.
Because they were bought, not built. Look at the history of any large vendor in this category and you will find the RMM came from one company, the service desk from another and the documentation tool from a third. Each arrived with its own database, its own permission model and its own release cycle. Joining them up afterwards is genuinely hard engineering, so mostly what gets joined up is the login page.
That is not a criticism of the engineers involved. It is a structural constraint. A product that already has ten years of its own users, roles and tables cannot be rebuilt on somebody else’s schema without breaking every customer it has. So the vendor adds single sign-on, writes a connector that copies records between the two databases on a schedule, and calls the result a platform. It explains something every technician already knows: no matter how many products you buy from the same logo, the same handful of things stay stubbornly manual.
Each of these is a seam: a place where two tools meet and a person does the joining. The copy-paste is the visible part. Underneath it is a permission check nobody made, because the person who copied the password had the right to read it in the vault but the vault has no idea what they did with it next. Seams cost minutes a day per technician, which is bad enough. What they really cost is the answer when something goes wrong, because the story is split across systems that do not know about each other.
Every vendor in this category uses the word. It usually means single sign-on plus a scheduled sync. Here is the difference between the common version and one system.
The right-hand column is not a list of features. It is a list of consequences. One identity provider with multi-factor means one place to add someone and one place to remove them; deactivate them and every session ends immediately, not eventually. Permissions granted by action through your groups mean that changing someone’s group once changes their access in every module at the same moment. One release train means every module ships at the same version on the same day, so there is no compatibility matrix to check and no integration breaking in the gap between two vendors’ releases.
None of the following is a feature anyone chose not to build. They are what becomes possible when the tools share a database, a permission model and a release, and impossible when they do not.
Route tickets on what the monitor saw. Intake rules match on the hostname, the failing check’s name, the host group, even the exit code the check returned, not just the words in an email. A disk alert on a SQL server goes straight to the database queue. How routing on the check works.
Connect without seeing the password. The credential is pulled from the vault and injected into the session server-side. A junior technician can work on a domain controller without ever being shown domain admin. The thirty-second version.
No overnight device sync. The device list in the service desk is fetched live from the monitoring platform. There is no nightly job to drift and nothing stale to reconcile. Why the sync lies.
Turn a fix into documentation. Write the resolution once and publish it to the knowledge base from inside the ticket, as yourself, already linked to the client and the machine.
Attachments land somewhere governed. A file emailed into the service desk becomes a tenant-owned document in the vault, scanned on the way in, not a loose blob in storage nobody audits.
The credential opens the host. The host is on the ticket.
We put the seams into minutes because that is how a technician feels them. The Stack Seams Score words its results the same way: device names copied into tickets, about ten minutes per technician per day; a password looked up in one tool to type into another, about five minutes and an audit gap each time; a nightly sync nobody trusts, one reconciliation a month. Multiply the first two by the size of your bench and the number of working days, and the login-page kind of integration has a payroll cost before you count the outage the client found first.
One platform means one platform. The modules are not standalone products with an optional platform bolted on; they genuinely run on Guardian Plus, and it comes with every deployment. You cannot take one module and run it beside a competitor’s. If you want to keep another vendor’s service desk and buy only our monitoring, we are not your answer, and we would rather say so here than on the third call. In exchange you get everything on this page, and that should be a conscious choice rather than a surprise.
Ten yes/no questions, two minutes. Your score, the three most expensive seams in your stack, and what to do about each.