Hello!
Sorry in advance for quite a long post but I am unsure how to explain everything without writing it all down here… so thank you and please bare with me! :)
I’ve just started a new job as junior IT sysadmin and am trying to figure out whether my worries are normal or if this is just what some smaller companies are like and everything is ok.
During the hiring they’ve told me they have no internal IT department and relies only on 2 external IT people: -1 consultant who sometimes helps with sharepoint -1 sysadmin who manages everything but also works fulltime for another company They’ve also mentioned they had recently migrated to M365 and that the migration has caused all kinds of issues with permissions, access rights, and overall administration. It sounded very messy but also interesting, so I still accepted the offer thinking I will have to deal mostly with M365.
Now I’ve started this job and got to know that comany’s CEO is apparently quite controlling and wants to know and see everything. He is also a Global admin in M365 and has additional high privilege roles assigned. I’ve also learned that the expectations now seems much bigger than what I would normally think that only a junior sysadmin alone should do since they expect me to: -help with sharepoint administration and it’s structure -manage and redesign existing M365 permissions and access -communicate with all departments to understand their workflows, requirements and software that they use -review entire companys IT infrastructure -potentially introduce company wide AI and security policies -work with and administrate MS Dynamics and PowerBI (I told them that I have no or almost no experience with those) -work with integrations between sales platforms and internal systems -participate in and maybe even lead future CRM migration (no more info) -help with creating a document management system because they currently don’t have it -potentially introduce on prem servers in future
At this moment I have only sharepoint permissions in M365, which are nowhere near enough for many tasks I’m being asked to help with not to mention that I have no admin access in my computer either. I’ve requested necessary access to actually do tasks they’re expecting from me from the external sysadmin. I asked for global admin and local admin rights, providing detailed info in an email for why I need them. I’ve also cc’d my manager (that’s not even related to IT) to document everything. The external sysadmin just sorta ignored me by only replying that they wish to meet up sometime later so I still have no access. I honestly don’t get how I’m supposed to manage systems if I cannot even access them or see what’s inside normally…
I also asked my manager a fairly direct question the other day: ‘If the company needs all of this why did you hire specifically just a junior sysadmin instead of an experienced IT manager or some senior sysadmin?’ The answer I got was that they want to ‘grow a person internally alongside the company needs’… Then I pointed out that someone with more experience would make less mistakes and be able to set everything properly. The response then was basically that they are not afraid of mistakes because they’re an RnD company where mistakes are normal. They also said that they didn’t want someone who would come in and ‘do everything their own way’… like wtf…? That answer then left me confused and speechless because things like permissions, security, infrastructure and stuff already have best practices for a reason!
So…is this a normal situation for smaller companies that are building their internal IT for the first time? Is this an actual growth opportunity and I’m just worrying for nothing…? Would you guys be concerned and think about exiting already? I’m simply very confused on what’s the right thing to do…
This sounds pretty normal for a small company that doesn’t really understand its I.T. needs yet.
Aim to automate as much as you can, you’ll be busy enough without having to perform the boring, but necessary, repetitive tasks like handling user on-boarding, off-boarding, password resets, and permission changes. I’ve generally found that it’s best to do a job manually first, noting down everything, then aim to automate it the second time, because, if there was a second time, there’ll be a third, fourth, fifth, and so on ad-nausiam.
Try to also automate testing. A script that can sanity check permissions, paths, and the like will save you many painful hours of debugging as you can test every change, rather than working out that the quick fix you did last Tuesday has broken the CEOs access to a critical slide deck today, and needs to be fixed 5 minutes ago.
Document the why, the how, and the what of everything you do. It doesn’t need to be fancy, a set of text files is more than enough, but I’ve been using Obsidian, or vimwiki on Linux, and it will save your bacon when you need to figure why something is the way it is, how you can fix it, and what to change. If you do use a wiki, keep a daily journal, but link the entries to pages about the relevant services. Don’t store the wiki on sharepoint, or you risk not having access when you need to fix sharepoint.
Set up an internal monitoring service that can notify you if a service goes down, or one of those tests you automated starts to fail. You will seem like a wizard when you’re already fixing a problem before anyone else notices.
Seen as the CEO has admin access, as, presumably, does the external IT person, enable audit logging on everything. That way, when something breaks, you’ll know who did it, and what they did, even if it was you.
As others have said, this will be hard, but it’s a massive opportunity to grow your skillset rapidly, in an environment that sounds like it may tolerate some mistakes.
Good luck.
One other thing: get all requests in writing, or confirm them back to the requester in an email if they insist on giving verbal instructions. This will save a lot if miscommunication and misunderstanding. Instigate a ticketing system if you can, and only process tasks through that. Add the tasks yourself if you have to. Users may start off hating it, but they’ll come to appreciate being able to see how their request is progressing, and what else you’re being asked to do.