Managed IT services for small businesses · Open beta · 3 founding spots: The person who answers your ticket is the person who set it up.
SentinelGrid Managed IT takes the boring, critical, always-your-fault half of running a business off your desk. Monitoring that catches failures before your customers do. Every device inventoried. And a human on the other end of the ticket.
We're in open beta and taking three founding clients, not thirty. You get direct access to the people who build and run it, a rate that doesn't move for two years, and a published list of what isn't finished yet.
Already a client? The portal is one sign-in for tickets, your status page, documentation and signed agreements. Or just email us.
We are building a managed IT practice properly rather than reselling someone else's, and growing the team around it deliberately: vetted contractors and technicians are being brought on to support managed IT clients, and how they're vetted is published. The helpdesk and ticketing system is finished and live. Remote support and device management run in production. Monitoring, alerting and status pages are live at status.sentinelgrid.us, and you can go and look at them right now. The client portal is live at portal.sentinelgrid.us. Backups, and single sign-on in client environments, are designed but not built yet, and the build status below says exactly which is which.
That's why there are three spots instead of thirty: three is what can be supported properly while the team is built out around them. You get a lot of attention from the people who build the thing, and a lower rate while parts of it are unfinished. If you need a finished, twenty-technician MSP today, we are not that yet.
A whole shop's computers, run from one place
The fastest way to judge an IT provider is to look at something it already runs. So: every computer at Zych's Certified Auto, from the front office to the service desk to the shop floor, is enrolled, configured and kept in line by a fleet platform we wrote and operate ourselves. Here is what it does.
Zych's Certified Auto Service
In productionAn independent auto repair shop that has been fixing cars since 1976, with Windows and Ubuntu machines side by side, used all day by people whose job is not babysitting computers. The brief was simple: the computers should look after themselves, and somebody should be able to see all of them without walking the building.
- Platform
- Zych's Fleet
- Built & run by
- SentinelGrid
- In production since
- August 2026
- Current build
- 2.5.0
- Operating systems
- Windows + Ubuntu
- Check-in interval
- 30 seconds
Enrolment nobody sits through
A machine is imaged, given an enrolment code and left alone. It comes back with its accounts, its policies, its applications and the shop's wallpaper already on it. Adding a computer is not a day's work.
Policies that don't drift
Screen lock, USB storage, AutoPlay, the firewall, Remote Desktop, SmartScreen and automatic updates are decided once in the console. Change one at the keyboard and the fleet puts it back within five minutes.
Updates on the fleet's terms
Every machine reports its build on each check-in and compares it against the fleet's. Updates that need a restart go through a deferral flow so nobody loses work mid-job; the rest arrive silently.
Malware protection, verified
Endpoint protection installs unattended from a licence key typed once, and each machine reports whether the service is running rather than merely installed. The key never returns to the dashboard or lands in a command log.
Health the machines report themselves
Disk, memory, battery and restarts, plus three rows that audit the fleet's own work: the accounts it asked for against the ones that exist, the applications assigned against the ones installed, and whether the background went on.
Fewer drives to the shop
Preset commands, application delivery and file distribution run from the console, including a sound you can play through a machine to find which desk it is sitting on. Most of what used to need a visit is now a click.
What this proves, and what it doesn't
Zych's is not a managed IT beta client and is not one of the three founding spots. It is a business whose computers we run on a platform we built for it. We are showing it to you because the endpoint half of managed IT is the part everyone claims and few demonstrate: automated enrolment, enforced policy, verified endpoint protection and self-reported health, in production, on machines that have to earn money all day.
What it is not is a claim that all of that is already generalised to any client's environment. That work is the row marked In build on the build status, and Zych's is the reason that row moved at all.
Unmanaged IT doesn't fail gradually.
It works fine right up until the morning it doesn't, and then it costs you a week. These are the three that end up costing small businesses the most.
The backup nobody tested
The job has been reporting success for two years. Then you need it, and you find out it's been backing up an empty folder, or the restore takes nine days, or the encryption key left with a former employee. A backup you haven't restored from is not a backup.
The account that never got closed
Someone left eighteen months ago. Their email is off, but their VPN login, their admin account on the file server, and their access to the billing portal are all still live. Nobody knows, because nobody owns the list.
The single point of failure with a pulse
One person knows how the server room is wired, which service to restart, and where the licence keys are. It isn't written down. When they're on holiday, or they quit, the business finds out exactly how much of its operations lived in one person's head.
One flat monthly line. Six disciplines.
This is roughly what a competent in-house IT hire would cover, without the salary, the benefits or the recruiting, and without everything depending on one person.
Monitoring & alerting
Continuous checks on servers, endpoints, sites and network gear, with alerts routed to us rather than to an inbox nobody reads.
- uptime & service checks
- disk, memory, temperature
- certificate & domain expiry
- alert triage before escalation
Backups: on the roadmap
Designed, not built, and not something we sell yet. Until it ships, your backups stay your responsibility, and the agreement says that in writing.
- encrypted, off-site copies (planned)
- restore tests, documented (planned)
- we will not call a backup working until we have restored from it in front of you
Endpoint management
Every machine known and accounted for. Patching, encryption enforcement and a standard build are on the roadmap or in build, labelled below.
- full device inventory
- OS & third-party patching (roadmap)
- disk encryption enforcement (in build)
- standard build & imaging (in build)
Identity & access
MFA across your tools and a written joiner–mover–leaver process instead of a spreadsheet and good intentions. Single sign-on is on the roadmap.
- sso / federated login (roadmap)
- mfa rollout & enforcement
- onboarding & offboarding runbook
- quarterly access review
Network & infrastructure (roadmap)
On the roadmap, not included yet: firewall configuration, server upkeep and network documentation. We do not provide a VPN or private network service today.
- firewall & segmentation (roadmap)
- server & hypervisor upkeep (roadmap)
- network documentation (roadmap)
Helpdesk & IT support
A ticket queue, a response target, and the same technician who already knows your setup. No call centre and no reading from a script.
- email & portal ticketing
- agreed response targets
- remote support sessions
- monthly ticket review
Most MSPs resell a stack. We build and run one.
SentinelGrid started as a software studio and still is one. The managed side runs on the same discipline: infrastructure we operate ourselves, automation we write ourselves, documentation we maintain because we're the ones who have to read it at 2am.
That distinction shows up on the days when the fix isn't in a vendor's knowledge base. When the answer is "someone needs to write a script for that" or "this needs a proper identity provider in front of it," we're already the people who do that work, not a reseller waiting on a support ticket of their own. If you want the longer version, read why we built managed IT for small businesses and why we run our own uptime monitor.
Building our own stack also means some things cost you less with us:
- Business email at no extra cost. Every client gets email on their own domain: send up to 300 a day and receive as much as you need, included with your plan.
- Fewer app subscriptions. When a paid app is doing a simple job, we can replace it with tools we build and run for you.
- On site wherever you are. We support you remotely by default and come on site when hardware needs hands.
- Agreements you can read first. Pricing, response targets and terms are published before you sign anything.
Four levels of "not your problem"
Published beta rates, held for 24 months from your start date. The audit confirms your seat and device counts; nothing else moves the number. No per-incident billing and no surprise line items.
What "beta" changes about these: the coverage below is what each plan becomes. The beta rate is about 20% under the list rate we'll publish when the beta ends, and it buys the parts that are built today (helpdesk, remote support, monitoring) plus everything that ships during your 24 months, at no increase. You'll know which is which before you agree to anything, because we'll walk the build status with you during the audit.
- Monitoring & alerting, 7 days a week. P1 outages: best-effort response within 2 hours, any hour
- Patch reporting
- Monthly health report
- Full asset & access documentation
- Managed patching & remediation (roadmap, not included yet)
- Helpdesk for staff
- Identity & network management (roadmap, not included yet)
- Everything in Watch
- Helpdesk for your whole team: tickets by email or the portal
- Remote help for everyday IT problems
- Endpoint management & encryption
- Staff onboarding & offboarding
- Everything in Support
- Managed patching & remediation (roadmap, not included yet)
- Endpoint management & encryption
- Staff onboarding & offboarding
- MFA management
- Single sign-on (roadmap, not included yet)
- Quarterly review & roadmap
- Network & server infrastructure (roadmap, not included yet)
- Everything in Manage
- Server & hypervisor management (roadmap, not included yet)
- Firewall management (roadmap, not included yet)
- Zero-trust access (roadmap, not included yet)
- Identity provider operations (roadmap, not included yet)
- Documented disaster recovery plan (roadmap: needs the backup pipeline)
- Priority response target
- Named technical lead
- Technology planning & budgeting
- Infrastructure audit & written risk listWeek 1. Yours whether you sign or not.
- Free
- Onboarding (weeks 2–4)Stabilise, instrument, document, hand over. Post-beta this is a one-time fee equal to one month.
- Free during beta
- A "user"A person with a work login. Their primary laptop or desktop is included.
- Included
- Additional server or hypervisorBeyond what the plan includes. Support includes none; Watch counts servers as devices.
- $99 / month
- Projects & out-of-scope workOffice moves, migrations, new systems. Quoted and approved before we start.
- $95 / hour
- On-site visitsRemote is the default; on-site when hardware needs hands. Two hours a month included on Operate. Travel outside Central Florida is quoted with the visit.
- $95 / hour
- Third-party licences & hardwareMicrosoft 365, backup storage, firewalls and so on. Always approved by you first.
- At cost
- BillingMonthly in advance, net 14, by invoice. Month to month, 30 days' notice, no exit fee.
- USD
Worked examples: a 30-person company with no servers on Support is $1,350 a month. A 6-person office with one server on Manage is $474 a month. A 12-person firm with two servers and a firewall on Operate is $1,548 a month. A 20-device shop that already has an IT person, on Watch, is $239 a month. Not sure which fits? The audit answers that, and it's free either way. Full detail on the pricing page.
Four weeks from "we should sort this out" to sorted
You get the deliverable from step one whether or not you continue. That's the entire risk.
Audit & inventory
We map what you have: machines, servers, accounts, subscriptions, licences, network gear, and the undocumented thing only one person knows how to restart. You get a written inventory and a prioritised risk list. Free, yours to keep, no obligation to continue.
Stabilise
We fix what's already broken or one bad day away from it. We tell you in writing whether your existing backups have ever been restored from. Admin accounts inventoried and locked down. Stale access removed.
Instrument & document
Monitoring agents deployed, alert routing configured, helpdesk opened to your team, and runbooks written for the things that will eventually break. Your documentation lives in your account, not ours.
Hand over
We take ownership. From here, alerts come to us and tickets come to us. Tickets are taken 24 hours a day, seven days a week, and we work them every day, weekends included. On-site visits run Monday to Friday, 3pm to 8pm Eastern, and at weekends by arrangement. A P1 outage gets a best-effort response within two hours at any hour. You get a monthly report and a quarterly review of what keeps generating tickets, because the goal is fewer of them each quarter, not more billable hours.
You're handing someone the keys. Here's who, and on what terms.
A managed IT provider gets admin on everything you own. We think that means the contract, the security practices and the person behind them should all be public before you ever talk to us.
Who you're dealing with
SentinelGrid is a software company and managed IT provider with headquarters in downtown Orlando, Florida, a location in Sanford in Seminole County, and service hubs in Jupiter, Florida, and Laurel Park, North Carolina. There is no sales layer and no account manager between you and the people doing the work: whoever scopes your environment is accountable for supporting it. Contractors and technicians are being brought on under a published vetting process, and you are told by name who has access to your systems. Every agreement names SentinelGrid as the counterparty, and the Terms of Service, Privacy Policy and Managed IT Service Terms are published and versioned.
- Business
- SentinelGrid Managed IT
- Locations
- Headquarters in downtown Orlando, Florida; a location in Sanford, Florida; service hubs in Jupiter, Florida, and Laurel Park, North Carolina, USA
- Serving
- On-site and remote, wherever you are
- Governing law
- Florida
- support@sentinelgrid.us
- Response
- One business day, by a person
Agreements you can read first
Plain-English contracts with a one-line summary under every heading. Both are blank templates you can read before the audit, and both say in writing what's built and what isn't.
Security practices, published
Passwordless single-use sign-in links, HMAC-signed sessions, no public sign-up anywhere, admin surfaces reachable only over a private Tailscale network, secrets in a vault, everything deployed from version control, remote sessions logged by name.
Regulated data, handled properly
If you handle health records, card data, criminal-justice information or student records, there's a signed addendum for it that allocates responsibilities, and we keep that data out of our systems until it's signed.
Your data stays yours
Network diagrams, runbooks, inventories, credentials, the audit itself: all yours, kept current, handed over in a usable format on exit. We are your data processor and act only on your instructions.
Incidents: told within 72 hours
If something happens to your data on our watch, your primary contact hears from us without delay and within 72 hours of confirmation. You get what happened, what was affected and what we did, and we help with any notice you have to make under Florida's FIPA, HIPAA or PCI DSS.
Leave whenever you like
Month to month, 30 days' notice, no exit fee. On the way out you get documentation and credentials, we remove our agents, and we help whoever comes next. We will never withhold your own data over a billing dispute.
Existing clients: the Client Portal
One sign-in for raising and tracking tickets, your private status page, your documentation and the agreements you've signed. Sign-in is a single-use emailed link, so there is no password to remember or leak.
Could a scammer send email as your business?
Find out in seconds. Type your domain and get a free letter grade for how easy it is to fake, with exactly what to fix. No sign-up.
What's ready and what isn't
Updated as things ship, so you can see what you'd be buying into.
This is the short version. The full build status has every row, including legal review and team build-out. If a row moves, it moves because it shipped.
Take the free audit. Decide after.
Tell us what's breaking and we'll come back with what we'd cover, what it would cost, and a realistic timeline. No pitch deck, no discovery call about your discovery call.
The questions we get asked
Where are you located, and do you come on site?
SentinelGrid has its headquarters in downtown Orlando, Florida, a location in Sanford in Seminole County, and service hubs in Jupiter, Florida, and Laurel Park, North Carolina, and we are not tied to them. We go on site wherever a client is, and we support clients remotely too. Almost everything a managed IT provider does is remote by design: monitoring, patching, helpdesk and identity management don't depend on where you are. On-site visits are for jobs that need hands on hardware.
It's a beta, so what am I signing up for?
Nothing binding, at first. Joining the beta list gets you a conversation and a free written audit of your infrastructure. You see the scope and the price before anything is agreed, and if you walk away at that point you keep the audit document. That's the whole downside.
Who works on our systems?
Only people named to you in advance. Managed IT is the newest thing our team does, and we are actively bringing on contractors and technicians to support managed IT clients, and every one of them goes through the same vetting process before touching a client system: identity verified, references checked, a technical assessment, a signed contractor agreement with confidentiality and access limits, per-engagement access that is revoked when the work ends, and fingerprint-based background checks for CJIS work. Three beta spots is what can be supported properly while that team is built around them. You are told by name who has access to your environment, before they have it, and you can say no.
What happens if SentinelGrid goes away?
Every small MSP should be asked this. Your documentation, inventory and credentials live in your possession, not locked in our tooling, so another provider can pick them up and be useful on day one. That's a deliberate design choice and it's the same reason there's no lock-in contract.
Which parts are working today?
Helpdesk and ticketing is finished and live. That's the portal our clients raise tickets in once they're onboarded. Remote support and device management run in production. Monitoring and alerting is live, and you can see it yourself at status.sentinelgrid.us. The client portal is live at portal.sentinelgrid.us. Backups, and single sign-on in client environments, are designed but not built. The build status table above is kept accurate. If a row says "planned", it means we haven't built it, not that we're being modest.
Why would I buy something unfinished?
Because the price reflects it, the rate is locked for two years while the rest ships, and the alternative for most small teams isn't a polished MSP. It's nobody, or whoever is best with computers doing it on top of their real job. If what's already live (a helpdesk, remote support, monitoring) fixes more than it costs, that's a reasonable trade. If it doesn't, the audit will show that and we'll say so.
We already have someone who does our IT.
Most of our target clients do, usually one overloaded person, or whoever is most comfortable with computers. The Watch plan exists for exactly that: monitoring, patch reporting and documentation handled underneath your person, so they get their week back and stop being the single point of failure.
How small is too small?
Not very. A five-person office with a couple of servers and a pile of laptops is a completely reasonable client. Below that, a one-off audit and cleanup usually makes more sense than a monthly agreement, and we'll tell you that rather than sell you a plan you don't need.
What does managed IT cost?
It's published, above: Watch is $149 a month flat for up to 10 devices ($9 per extra device); Support is $45 per user per month with a $195 minimum; Manage is $79 per user per month with a $395 minimum; Operate is $129 per user per month with a $995 minimum and up to 3 servers included ($99 per extra server). The audit and onboarding are free during the beta, project work is $95 an hour and quoted first, and licences are passed through at cost. The audit confirms your seat and device counts, and that's the only thing that moves the number. Once agreed, your rate is held for 24 months, including through everything that ships during that time. No per-incident billing, no surprise line items.
Do you lock people into long contracts?
No. The 24-month beta rate is a commitment from us to you, not a cage. The engagement is month to month with 30 days' notice. Your documentation, inventory and credentials stay in your possession throughout, and we'll hand over cleanly to whoever comes next if you ever move on.
What do I sign?
One document, written in plain English with a one-line summary at the top of every section. Businesses we already know sign the Simple Client Agreement, which runs nine pages. Everyone else signs the Founding Client Agreement, which adds schedules for your plan, response targets, build status and contacts. Both are blank templates you can read before the audit; the real one is signed in the client portal, which records who signed, when and from where. Both say in writing what's built and what isn't, hold your rate for 24 months, and let you leave on 30 days' notice with your documentation in hand. If you handle regulated data there is an addendum for it: HIPAA business associate agreement, PCI DSS responsibility matrix, CJIS security addendum, FERPA student-data addendum. And when it's time to leave, a termination & hand-over notice that walks both of us through giving everything back.
How do I sign in as an existing client?
Use the Client Portal. It's one sign-in for tickets, your private status page, your documentation and the agreements you've signed. Sign-in is a single-use link emailed to the address we enrolled, so there's no password to remember. If your link doesn't arrive, email support@sentinelgrid.us and a person will sort it out.
What if something goes wrong and you can't fix it?
Then we say so, and we bring in someone who can: vendor support, a specialist, or a hardware supplier. Part of what you're paying for is someone who owns the problem until it's closed, including owning the escalation.
On site wherever you are. Headquartered in downtown Orlando.
Almost all of managed IT is remote by design, and when a job needs hands on the hardware we come to you, wherever you are. Our headquarters is in downtown Orlando, Florida, with a location in Sanford in Seminole County and service hubs in Jupiter, Florida, and Laurel Park, North Carolina, so these places each have their own page.
Lake Mary
Offices and professional services on the International Parkway corridor.
Sanford
Downtown, trades and light industrial, where computers live on a shop floor.
Oviedo
Practices and small firms whose IT was set up by someone who has since left.
Altamonte Springs
Medical, dental and retail on the 436 corridor, holding regulated data.
Winter Park
Small professional firms where confidentiality is the product.
Orlando
Businesses too small to interest a large MSP, too dependent to keep winging it.
Winter Springs, Casselberry, Longwood, Heathrow, Maitland and Apopka are covered too, and so are Jupiter and Palm Beach County, Hendersonville and Henderson County and Asheville, North Carolina, and everywhere else. All locations.
Stop being the IT department.
Three beta spots. Worst case, you get a free map of your own infrastructure and walk away.