MSP pricing and margins FAQ

A healthy gross margin for MSP managed services is 65-75%. Below 60% means your stack costs too much or you're underpriced. Above 80% usually signals underdelivery or impending churn.

Total MRR from a client divided by total hours spent on work covered by that agreement. Support tickets, QBRs, and admin time count. Projects typically don't; they're priced separately. Most MSPs are shocked when they do this math. A $150/user client often nets $40/hour when you count everything included in the agreement.

Per user is cleaner for forecasting and scales better with remote work. Per device made sense when employees had one workstation. Now they have three. Per user wins unless you're supporting manufacturing floors or labs.

$100-200/endpoint/month for full stack. Lower end is basic RMM/AV. Higher end includes security stack, backup, vCIO, and compliance. Know your number and trend it monthly.

70-80% minimum. Higher is better for valuation and stability. Below 60% means you're running a project shop that happens to do some managed services. That's a harder business to scale and sell.

Per-user all-inclusive is the cleanest. Tiered packages (Good/Better/Best) work for sales teams. Per-device still works for infrastructure-heavy clients. Avoid hourly - it punishes efficiency and creates misaligned incentives.

MSPs should charge full price for contractors when the contractor has normal access, device support, helpdesk rights, onboarding, offboarding, and security coverage. If the support scope is lower, create a limited-access identity or app-only line item instead of guessing at a fractional user rate.

A limited-access user is a named identity with narrower support expectations: usually account setup, MFA, approved app access, and offboarding, but no managed device support or direct helpdesk access. The quote should list those limits clearly so the lower price has a matching lower scope.

Ask whether the contractor has email, MFA, VPN, managed devices, business app access, ticket rights, after-hours support, compliance scope, and a clear offboarding owner. Those answers decide the support class better than the payroll label.

Fractional user pricing usually creates renewal fights because every user becomes debatable. Support classes are cleaner: full supported user, limited-access identity, or device/app-only exception. Price the class and write the inclusions and exclusions into the quote.

Add the monthly cost to serve the client, including labor, tools, shared service overhead, and risk buffer. Then divide that cost by one minus your target gross margin. If the client costs $5,000 per month to support and you need 50% gross margin, the floor price is $10,000 per month.

Use burdened labor cost, not salary divided by 2,080 hours. Include payroll taxes, benefits, PTO, training, management time, and realistic utilization. A paid hour is not the same as a usable delivery hour.

Yes. Tools required to deliver the service belong in the cost to serve. RMM, EDR, backup, PSA, documentation, remote access, and reporting costs should be allocated by user, device, tenant, or client, depending on how the cost is incurred.

ScreenConnect is usually better for ConnectWise-standardized MSPs that need deeper technician controls, session recording, command line access, and unlimited unattended agents on support plans. Splashtop is usually better when lower starting cost, endpoint coverage, mixed-stack portability, or Mac-heavy support matters more.

Splashtop has the lower public starting price for Remote Support SOS, while ScreenConnect Standard costs more per concurrent tech but includes unlimited unattended access agents. The cheaper MSP answer depends on technician concurrency, managed endpoint count, and required security controls.

ScreenConnect publishes session limits by plan, with Standard at up to 3 sessions per tech and Premium at up to 10. Splashtop positions Remote Support around concurrent user licensing and says it does not limit connections per user, but MSPs should validate the exact plan before buying.

An MSP should keep a second remote access tool when it covers real gaps such as RMM remote control failures, Mac support, vendor-assisted sessions, after-hours access, or client-specific support cases. It should not stay if nobody owns stale agent review, technician access cleanup, audit history, and billing.

Usually yes on entry price. Splashtop's Remote Access Solo starts at $6 per month billed annually, Pro at $8.25 per user, Performance at $13 per user, and SOS at $22 per concurrent user license. ScreenConnect starts at $30 per month billed annually for One and $45 per concurrent tech for Standard, but that concurrency and support-first feature set can be cheaper once you price real technician usage.

ScreenConnect is usually the safer default for after-hours support because Standard includes Backstage, remote command line, Wake-on-LAN, session recording, and mobile guest support. Splashtop can still work, especially in SOS or Enterprise, but you should verify admin elevation, session logging, and how your desk launches the session from a ticket.

Sometimes. If one tool is your technician support layer and the other is your lower-cost remote access layer, keeping both can be rational. The mistake is paying for two tools that solve the same motion because nobody defined which one owns support and which one owns user access.

Yes, Splashtop SOS can raise a Windows session to admin and interact with UAC, which matters for installs and privileged fixes. That does not make every Splashtop plan equal, so test the exact tier and workflow your techs will use.

Pick ScreenConnect if technician-led support, concurrency, and deep session control are the priority. Pick Splashtop if budget, named-user access, and lighter-weight support workflows matter more. If your team cannot explain the difference between remote access and remote support, fix that first.

Segment the client before raising price. If the gap is caused by scope creep, tool changes, or added users, bring renewal options: increase price, reduce scope, move project work out of the monthly fee, or clean up the environment to reduce support load. If the client is a bad fit, use tighter boundaries or plan a controlled offboarding.

Kaseya high watermark pricing tied billing to a prior peak usage level instead of only current usage. If an MSP's endpoint or seat count dropped later, the bill could still reflect the higher commitment until the contract or billing terms changed.

Kaseya said Datto RMM, Autotask, and SaaS Protection are moving to committed minimum quantity plus variable consumption. MSPs are billed for the committed minimum or calculated usage, whichever is higher, so the minimum still needs to match the real client base.

MSPs should audit active endpoint counts, committed minimum quantities, product-level license pools, last 90 days of usage, and client-billed counts. The goal is to catch stale licenses, underbilled clients, and renewal commitments before they turn into margin leaks.

Datto is better when you want appliance-first BCDR with predictable restore behavior and flat-fee economics. Acronis is better when you want backup plus security and endpoint management in one platform and can handle more configuration.

Yes, if you already live in the Kaseya stack and want the shortest path to recovery certainty. No, if support friction, bundle pressure, or platform lock-in now cost more than the restore benefits.

Not automatically. Acronis can start with a lower entry point because it prices per workload or per GB, but add-on packs and DR can close the gap. Datto is usually easier to budget if you want flat-fee pricing.

It is a real migration, not a toggle. Plan for a parallel run, fresh agent deployment, restore validation, and a client-by-client cutover because existing recovery history does not move cleanly.

The best MSP backup solution depends on the recovery model and billing shape you can support. Datto is strongest for appliance-backed BCDR, Acronis for flexible backup plus cyber protection packaging, and Axcient for cloud-first BCDR with pooled fair-use storage.

Compare Datto and Acronis by recovery design, not only by feature count. Datto is simpler to position when clients need appliance-backed local recovery. Acronis is a better fit when you want backup, DR, and security services in one service-provider platform.

Axcient can be better when you want Direct-to-Cloud, BYOD, BYOC, or hybrid BCDR and can manage fair-use storage limits. Datto is usually better when the client needs a packaged appliance-backed recovery story with fast local restore.

MSPs should compare backup pricing by protected workloads, protected data size, retention, RTO, RPO, recovery testing, contract terms, and overage rules. A cheap backup quote can still lose margin if storage growth or recovery labor is not priced.

Acronis and Axcient are common Datto BCDR alternatives for MSPs, but they solve different margin problems. Acronis fits broader cyber protection packaging. Axcient fits cloud-first or mixed deployment models with public fair-use storage rules.

Dropsuite is usually better when the MSP wants a current SaaS backup and archive standard across Microsoft 365 or Google Workspace. SkyKick can still be better when old SkyKick backup history, migration workflows, or ConnectWise-aligned vendor strategy make switching risky.

Choose Dropsuite over SkyKick when you are starting fresh, standardizing Microsoft 365 backup, or cleaning up client takeover backup scope. Confirm retention, archive add-ons, deleted-user handling, and restore testing before moving clients.

The migration tail is the old backup history that may need to stay accessible after moving a client from SkyKick to another backup tool. MSPs should confirm export options, minimum license requirements, deleted-user handling, and who pays for old-data access.

MSPs should run sample restores across Exchange, OneDrive, SharePoint, Teams, and deleted-user scenarios, then record the workload, restore target, technician, date, and result. A green backup status is not the same thing as client-ready restore proof.

SaaS backup margins leak through archive add-ons, retention exceptions, offboarding labor, restore tests, distributor billing cleanup, and vendor migration work. Put those rules in the package instead of hiding them in support time.

NAKIVO is usually the better fit when the MSP wants a simpler, lower-friction backup stack for mixed environments and can keep recovery scope narrow. Veeam is usually the better fit when the MSP wants deeper service-provider licensing, monthly usage reporting, and more control over backup architecture.

NAKIVO often looks cheaper on the license line because its MSP pages emphasize no contracts, no upfront fees, and flexible per-workload pricing. But MSPs still have to pay for storage, immutability, monitoring, restore tests, and support labor, so the real cost can flip fast.

NAKIVO can be a fit for Microsoft 365 backup when the client wants a simpler SaaS backup stack and the MSP is comfortable owning storage and restore validation. Veeam usually fits better when the MSP wants that same reporting model across a broader service-provider platform.

Veeam is usually better when the MSP wants control over backup architecture, repositories, recovery design, and service-provider operations. Axcient is usually better when the MSP wants a more packaged BCDR model with Direct-to-Cloud, appliance, hybrid options, and public fair-use storage rules.

MSPs should price Veeam beyond the license. Include repositories, object storage, immutability, backup servers, proxies, monitoring, patching, restore testing, usage reporting, and engineer time for recovery runbooks.

MSPs should measure protected data before quoting Axcient, compare it with the fair-use pool, and name what happens when storage, Virtual Office usage, or data exports exceed policy limits. Fair use should be a client term, not an invoice surprise.

MSPs should show which systems are protected, when the last restore test ran, what type of restore was tested, whether the client accepted the RTO and RPO, and what labor is included during a live incident.

Cove is usually better when the MSP wants a cloud-first backup service with included private-cloud storage, simpler server or Microsoft 365 packaging, and recovery testing close to the product. Veeam is usually better when the MSP wants more control over repositories, storage targets, service-provider reporting, and custom recovery design.

Compare Cove and Veeam by storage responsibility, not just license price. Cove includes private-cloud storage in its public packaging, while Veeam makes the MSP design and price repositories, object storage, retention, immutability, growth, egress assumptions, and recovery testing.

Cove packages automated recovery testing closer to the backup service, which can make restore proof easier to sell and explain. Veeam can support strong restore proof too, but the MSP must design the test cadence, repositories, reports, recovery targets, and runbooks that prove recovery works.

The quote should include protected workloads, protected data size, growth assumptions, retention, restore-test cadence, RTO, RPO, Microsoft 365 scope, local-copy rules, fair-use or storage assumptions, emergency labor, exclusions, and price-change triggers.

Wasabi is usually better when backup data is steady, retention is longer than 90 days, and monthly downloads stay at or below active storage. Backblaze B2 is usually better when restore testing or disaster recovery could create larger download months because it includes free egress up to 3x average monthly storage.

Wasabi's free egress policy works best when monthly downloads are less than or equal to active storage. If a client regularly downloads more than they store, the use case may fall outside Wasabi's published guideline, so MSPs should document restore testing and mass recovery assumptions before quoting.

Backblaze B2 includes free egress up to 3x average monthly storage, then charges a posted overage rate for additional egress. That makes B2 easier to quote for larger restore tests or disaster recovery months, but MSPs still need to price the labor around the restore.

MSPs should confirm whether the bucket needs bucket-level compliance, object-level Object Lock, governance mode, compliance mode, legal hold, and backup application support. Wasabi Object Lock must be enabled when the bucket is created, while Backblaze B2 can enable Object Lock on an existing bucket but cannot disable it afterward.

An MSP backup storage quote should include active data size, growth assumptions, retention, deleted-data behavior, restore test cadence, full recovery download assumptions, immutability mode, egress boundaries, and who pays for recovery labor. Storage price alone is not enough.

Veeam is usually better when the MSP wants control over backup architecture, repositories, VM-heavy recovery design, and service-provider reporting. Acronis is usually better when backup belongs inside a broader cyber protection package with endpoint, SaaS, storage, DR, and security services in one MSP platform.

Both can support restore proof, but neither proves recovery by itself. MSPs should document the workload, recovery point, restore target, technician, elapsed time, result, RTO and RPO acceptance, and client evidence for each test.

The quote should include protected workloads, storage and retention assumptions, Microsoft 365 scope, restore-test cadence, evidence records, emergency labor, after-hours rules, compliance reporting, and change triggers. Vendor capability is not the same as a priced recovery promise.

Acronis, Axcient, Cove, NAKIVO, MSP360, and Datto can all be Veeam alternatives for MSPs, depending on whether the MSP wants cyber bundle packaging, packaged BCDR, backup-first SaaS operations, simpler licensing, storage control, or appliance-backed continuity.

Acronis is usually better when an MSP wants backup tied to a broader cyber protection platform with DR, security, and RMM options. Cove is usually better when the MSP wants a backup-first service with included cloud storage, flat server or user pricing, and clear restore proof.

Compare Acronis and Cove by protected data size, retention, growth, local copy rules, and recovery movement. Acronis solution-based storage is pooled by workload at the customer level. Cove includes cloud storage in its standard pricing model, but fair-use and service scope still need to be documented.

Both can support restore proof, but the motion differs. Acronis fits broader DR designs across Acronis Cloud, Azure, or hybrid recovery. Cove puts automated recovery testing close to the backup service. MSPs should define the test cadence, evidence, RTO, RPO, and labor rules either way.

The quote should name protected workloads, storage assumptions, retention, restore test cadence, Microsoft 365 edge cases, recovery labor, after-hours rules, reporting evidence, and client approvals. Vendor name alone is not a backup scope.

MSP360 is usually better when an MSP wants white-label backup software and more control over storage vendors. Acronis is usually better when backup is part of a broader service package with disaster recovery, security, and endpoint operations. The final choice should follow the restore proof and quote scope.

Compare storage by target provider, ownership, retention, immutability, egress behavior, growth, and restore testing. MSP360 leads with bring-your-own-storage control, while Acronis supports packaged storage models and direct backup to S3-compatible storage.

Both MSP360 and Acronis can support restore proof, but the MSP has to define the test. A real proof package should name the workload, restore type, result, RTO, RPO, labor boundary, and client evidence. Vendor capability is not the same as a signed restore test.

The quote should include protected workloads, storage target, retention, immutability mode, restore test cadence, reporting evidence, Microsoft 365 scope, storage growth, overage rules, after-hours recovery labor, and vendor support ownership. Product name alone is not a service scope.

Azure CSP billing rolls customer usage through an Azure plan, subscriptions, resource groups, resources, and meters. The MSP uses Partner Center, Cost Management, and reconciliation data to map spend back to the client agreement and invoice the client cleanly.

MSPs avoid Azure consumption spikes by setting budgets, cost alerts, owner tags, approval limits, and pass-through language before work starts. Dev/test resources, storage growth, backups, and project environments need named owners and monthly review.

MSPs should include Azure reservations or Savings Plans only when the client approves the commitment, expected utilization, owner, term, and change risk in writing. Do not hide reserved capacity assumptions inside a flat managed services fee.

For many MSPs, the safest Azure billing model is pass-through consumption plus a clearly priced management fee. Cost-plus and percentage markup can work, but only when the client accepts variable bills and the agreement explains spike handling.

MSPs should set budget threshold alerts, anomaly alerts where available, and escalation rules by client, subscription, and key resource group. The alert recipients and approval limits should be listed in the client agreement or quote.

The platform is $99 per user per month month-to-month, or $89 per user per month with a 12-month commitment. Both options are billed monthly. The 14-day trial starts at signup and does not require a credit card.

Scopable is $99 per user per month month-to-month, or $89 per user per month with a 12-month commitment. Both options are billed monthly. The 14-day trial starts at signup and does not require a credit card.

Scopable is $99 per user per month month-to-month, or $89 per user per month with a 12-month commitment. Both options are billed monthly. The 14-day trial starts at signup and does not require a credit card.

Scopable is $99 per user per month month-to-month, or $89 per user per month with a 12-month commitment. Both options are billed monthly. The 14-day trial starts at signup and does not require a credit card.

Do not compare Huntress public endpoint pricing with guessed Sophos pricing. Compare total service cost: endpoints, identities, integrations, Microsoft coverage, after-hours response, cleanup labor, reporting time, partner margin, and any work left with the MSP.

Use partner pricing, not public guesses. Then add the MSP work around the vendor cost: endpoints, users, cloud identities, onboarding, policy tuning, after-hours communication, reporting, partner margin, and separately scoped remediation or cleanup work.

Price the full managed endpoint service, not only the license. Include endpoints, servers, operating systems, modules, onboarding, policy tuning, exclusions, EDR or MDR escalation, reporting, partner margin, and separately scoped incident cleanup.

Price ESU as a project plus the management around it: device inventory, enrollment, activation, update checks, reporting, and the retirement deadline. Do not bundle that labor into ordinary support.

MSPs should separate AI license costs from service work. Price readiness, rollout, governance, training, automation, and ongoing support from delivery cost, tool cost, risk buffer, and target margin. A practical starting point is a $2,500-$6,000 readiness project, a $750-$2,500 monthly governance retainer, or a $10-$30 per user monthly add-on when deliverables are defined.

For a small SMB, a Copilot readiness review often starts around $1,500-$2,500, while a rollout package with access review, pilot setup, policy, training, and 30-day review can land around $3,500-$5,000. License costs should stay separate from service pricing, and MSPs should use current distributor pricing before quoting.

The margin is usually in assessment, rollout, policy work, training, governance, reporting, and recurring advisory support, not license resale. AI can also improve managed services margin by reducing internal delivery time, but MSPs should only charge a separate AI fee when the client receives a named deliverable.

MSPs should avoid vague AI help, guaranteed productivity claims, unlimited AI support, and custom agent work without a separate scope. If the service does not define the tool, data, decision owner, support boundary, and acceptance criteria, it is not ready to quote.

Yes. Work IQ APIs use consumption-based billing through Copilot Credits. MSPs should set policy-level monthly limits, per-user ceilings, alerts, and named budget owners before agents run, especially if a third-party agent grounds in client Microsoft 365 data.

Quote Work IQ API work as a small rollout project, not free admin time. Include use-case review, licensing and billing setup, security group design, spend policies, alert owners, pilot support terms, and a monthly governance review if the client moves to production.

MSPs should separate AI usage from AI service work. Define included usage, set client-approved overage thresholds, assign alert owners, and show pass-through usage separately from readiness, rollout, governance, and support labor.

Use three thresholds: a watch threshold for unusual usage, an approval threshold before spend exceeds the client budget, and a stop threshold where access or expansion pauses until the client makes a decision.

Yes. Any increase to AI usage caps, token allowances, Copilot Credit budgets, agent access, or premium model usage should require written client approval. Otherwise the MSP owns the invoice argument after the usage happens.

Use Auvik when the network is its own risk domain: topology, SNMP depth, config history, multi-site visibility, and discovery cleanup matter. Use NinjaOne when the MSP is RMM-first and needs practical network monitoring beside endpoint management, patching, remote access, automation, and technician workflow.

NinjaOne network monitoring can be enough for smaller client sites, RMM-standardized MSPs, and environments where endpoint operations drive the service model. It is a weaker default for clients that need deep topology validation, config-change visibility, or formal multi-site network documentation.

Auvik is worth serious evaluation when the MSP inherits poorly documented networks, supports multi-site clients, needs config history, or wants network maps that can support discovery, remediation, and QBR planning. The MSP still needs to price cleanup and validation work separately.

Quote discovery, collector or agent setup, SNMP credentials, NetFlow or syslog setup, map validation, alert tuning, config-backup checks, documentation handoff, and recurring review. Monitoring software does not erase that work. It exposes it.

Kaseya 365 can turn an RMM choice into a bundle decision because endpoint management, security, backup, documentation, PSA, quoting, or operations modules may sit in the same commercial plan. MSPs should model endpoint counts, user counts, contract terms, renewal dates, and which products can or cannot move through FLEXSpend before choosing Datto RMM or VSA 10.

Committed Minimum Quantity billing means Datto RMM billing is based on the committed minimum or actual license usage, whichever is higher. MSPs need to reconcile endpoint counts, stale agents, add-on usage, client churn, and renewal timing before assuming tool cost will drop when client revenue drops.

Atera markets per-technician pricing with unlimited endpoints. ConnectWise is quote-based, and the final bill can depend on RMM package, endpoint count, PSA usage, remote access, add-on modules, services, and contract terms. MSPs should compare total operating cost, not only license price.

ConnectWise does not publish a simple public Automate pricing table for every buyer. MSPs usually need a quote, which means they should model endpoint count, modules, implementation help, contract length, renewal language, and exit cost before treating the price as comparable to a published SaaS plan.

Atera MSP pricing is built around paying per technician rather than per endpoint. That can work well for lean teams managing many devices, but MSPs still need to check which tier includes the reporting, audit log retention, automation, remote access, API, and billing features they need.

Atera and Syncro both use technician or user-based pricing models that can protect endpoint-heavy MSPs from per-device RMM bills. The cheaper option depends on technician count, plan tier, AI or reporting add-ons, Microsoft 365 needs, and whether the lower plan includes the workflow the MSP actually needs.

Windows 365 is easier to quote because it is billed per user per month, but it is not always cheaper. AVD can cost less for pooled or variable usage, but only when the MSP actively manages Azure compute, storage, networking, autoscale, and approvals.

AVD quotes should name Azure compute, storage, networking, FSLogix profiles, images, app testing, printing, backup, monitoring, security policy, after-hours support, and monthly cost review. The risk is not only the VM price. It is the operating model around the VM.