Section 1
Your price structure is the evidence, because nothing else can be checked
A 40-person accounting firm loses email for most of a Tuesday. The office manager, who is not technical and did not want this job, gets three managed service providers on the phone by Thursday. All three describe themselves in almost identical language: certified engineers, 24/7 monitoring, proactive rather than reactive, we become your IT department. She cannot verify a single claim in any of it. She has no way to test whether your patching discipline is tighter than the next firm's, or whether your alerts get acted on at 2am or acknowledged and forgotten. She will not find out for at least a year. That is information asymmetry: one side of a trade knows the quality of the goods, the other does not, and has to decide anyway. Economists study what happens next under the heading of signalling. A signal is any action that is genuinely cheaper for a high-quality provider to take than for a low-quality one. That cost gap is the entire mechanism. If a weak provider can do it just as easily, it carries no information. This is why certification lists do so little. Any provider can buy the exam and put the badge in the footer. It costs the good firm and the bad firm the same, so it separates nobody. The same goes for the word proactive and for testimonials nobody can trace. Price structure is different. What you commit to including, what you name as out of scope, and what you refuse to do below a certain size are all expensive for a sloppy provider to imitate. Publish a scope boundary you cannot hold and you get eaten alive inside ninety days. That asymmetry is what makes structure readable.
Section 2
Quote-only pricing forces the one comparison you lose
Assume you publish nothing. The buyer still has to compare. They just build the comparison themselves, out of whatever material you gave them. What they have is three proposal PDFs. Each describes scope in its own vocabulary, so the scope sections do not line up. Each has one thing in common: a monthly figure. When a non-technical buyer holds three documents that disagree on everything except the presence of a number, the number becomes the comparison. Not because they are unsophisticated, but because it is the only field that appears in all three and means the same thing in all three. Under that frame you lose more often than you should, for structural reasons rather than sales-skill ones. The lowest bidder is very often the one who scoped least. They excluded the server work, or capped after-hours response, or left project labour out. The buyer cannot see the exclusions because they cannot read scope. So you are not being compared to a rival. You are being compared to a smaller product wearing the same job title. There is a second cost, easier to ignore because it never shows up in your CRM. Quote-only offers filter at both ends. Buyers who could comfortably afford your rate never call, because with no anchor at all they assume an unknown firm is either out of reach or not worth the discovery process. Buyers who could never afford you do call, and you spend four to six hours on a site walk, an audit and a proposal before the number does the disqualifying that one published sentence could have done in week one. You pay for that sorting in delivery hours, after the fact, which is the most expensive time to pay for anything.
Section 3
Per-user, per-device, per-outcome: what each one signals and where each one bleeds
The unit you price in is not an accounting detail. It tells the buyer what you think you are selling, and it decides which client behaviours make you money. **Per-device.** You bill per endpoint. It signals that you manage machines, and buyers accept it because the count is objective. Where it bleeds: your revenue is tied to a number the client is actively trying to reduce. Every consolidation and every shift to BYOD shrinks your invoice while the support load stays flat, because the work followed the person, not the box. It also puts you on the wrong side of the incentive: you earn more when your client runs more hardware, the opposite of the advice you should be giving. **Per-user.** Bills per person, devices bundled. It signals that you support people, and it matches how the buyer already budgets. Where it bleeds: users are not interchangeable. The partner with a laptop, a tablet, a phone and a habit of travelling costs you several times what the front-desk hire costs, and pays the same. Per-user also underprices infrastructure. A manufacturer with 25 staff, three servers and an ancient line-of-business application destroys a rate built on office clients. **Per-outcome.** A flat fee tied to a result rather than a count: a per-site fee, an uptime commitment, a fixed monthly regardless of seats. It reads as confidence, which is what makes it a signal. Where it bleeds: you absorb all the variance. Every unbudgeted incident lands on your margin. If you cannot state your cost to serve per client, per-outcome pricing is a bet placed without knowing the odds. And outcome definitions that felt obvious at signature get argued over in month nine. Most mature providers land on a hybrid: a per-user base plus separate lines for servers or sites. That is usually correct, and every unit you add makes you harder to compare, which cuts both ways.
Section 4
Publish the structure, not the number
The move is not to put a rate card on your homepage. Very few providers can hold one published rate across an accounting firm, a clinic and a machine shop, and pretending otherwise creates proposals that contradict your own website. Publish everything around the number instead. That means five things. First, named tiers, two or three, each with a one-line description of who it is for. Second, an in-scope and out-of-scope list per tier. The out-of-scope list matters more, because exclusions are what the buyer cannot construct alone. Third, a stated minimum: the smallest engagement you take, in seats or monthly value. Fourth, the price movers: servers on site, a compliance regime such as HIPAA or CMMC, multiple locations, after-hours coverage, legacy systems. Fifth, what is billed separately: projects, hardware, licensing, remediation. None of that requires publishing what any client pays. It gives the buyer enough to place themselves. A three-location clinic reads the price movers and understands, before speaking to anyone, that they are not a base-tier account. That is the objective. The deeper effect is on the comparison frame. Whoever sets the terms of comparison usually decides who wins, and in an unpackaged market nobody sets them, so they default to the monthly rate. Publishing structure moves the buyer's question from which of these is cheapest to which of these matches our situation. You will still lose deals under the second question. You will lose different ones, earlier and cheaper. Worked examples of how this reads on a live page sit at https://bizgrowthaxel.com/msp/. One caution on where advice comes from. Many pricing templates circulating in MSP peer groups originate with vendors whose own revenue is per-seat. That does not make them wrong. It does mean the default suits their billing model, so check it against yours first.
Section 5
Design the rules so the wrong buyer disqualifies himself
Mechanism design means writing rules so that people pursuing their own interest produce the result you wanted, without you policing anything. Applied to pricing, the goal is straightforward: a poor-fit buyer should work that out from your published material and leave, before either of you has spent anything. Four disclosures do most of the sorting. A published minimum. One sentence: our smallest managed agreement is X seats, or engagements start at X per month. It removes the entire category of enquiry that ends in an apologetic email after two meetings. An onboarding or remediation fee, stated as a range. Buyers who want you to inherit a decade of deferred maintenance for free will see the range and leave. Buyers who understand that cleanup is real work read it as evidence you have done this before. A stated coverage boundary. What is included during business hours, what happens outside them, and what after-hours work costs. Vagueness here is not generosity. It is a future dispute you have agreed to have. A co-managed tier. If a prospect has an internal IT person, a we-become-your-IT-department framing asks them to fire a colleague. Naming co-managed as its own tier keeps a deal you would otherwise lose to positioning rather than capability. Then the one commitment that works as a genuine costly signal: a response-time guarantee with a real remedy attached, such as a service credit. A provider without staffing depth cannot afford it, which is what makes it informative. Note the trap. If it works, rivals copy it, and an arms race of richer guarantees compresses everyone's margin without separating anyone. Adopt it as a lever you price and revisit, not a promise you make once and forget.
Section 6
What to do this quarter
This is a two-week exercise, not a project, and most of the work is arithmetic you already have the data for. Start with your last twenty closed agreements. For each, compute monthly revenue divided by users and by devices, then set it against your delivery hours for that client. You are looking for the spread between your best and worst effective rate. Without that number, tiers are guesswork and per-outcome pricing is reckless. Next, write your exclusions before your inclusions. Inclusions come easily and mean little, since every competitor lists the same ones. Exclusions are hard to write, which is why they carry information. Build two or three tiers. Not five. Past three options the buyer stops choosing and starts deferring, and a deferred decision is a lost deal that never gets marked lost. Write the price movers as plain conditions, not a formula. Servers on site. More than one location. A regulated environment. Coverage outside business hours. The buyer needs to recognise themselves in it, not compute anything. Publish a floor rather than a rate. A floor is a boundary you can hold across every client type. A rate is a number your own proposals will contradict. Add a next step smaller than a sales call: a short scoping questionnaire, or a self-assessment. Then measure the right thing. The metric is not lead volume, and if the structure works your enquiry count may fall. Track scoping calls to issued proposals, and proposals to closed deals. If both improve while raw enquiries drop, the structure is doing the sorting you used to do by hand. Give it two quarters, because deal cycles here are long enough that one quarter is mostly noise.
Section 7
When structure stops separating you, and what this lens cannot see
Every signal decays through adoption, and this one is no exception. If every provider in your metro publishes tiers, scope boundaries and a floor, structure stops separating anyone and becomes table stakes. Buyers then re-anchor on whatever is still unpublished: transition plans, contract remedies, or a reference they can actually call. Plan to re-earn separation rather than assuming you can hold it. The first-mover advantage is real and temporary. There is a sharper break available too. The tier ladder most MSPs build assumes a stable relationship between ticket volume and cost to serve. If AI-assisted support collapses the cost of tier-1 work, ticket-driven tiering needs repricing rather than adjusting. That is not a prediction. It is a reason to build tiers you can revise annually rather than tiers welded to a five-year contract template. Then the honest limit. Signalling models assume a market with enough providers that comparison is the buyer's central problem. Plenty of MSPs do not operate in one. In a two-provider town the game is not comparison at all. It is a choice between the two firms everyone already knows, settled by who someone's brother-in-law trusts and a twenty-year history no published tier will overwrite. There, publishing structure mostly tells your one competitor where your boundaries sit. The same blind spot shows up inside referral networks anywhere. One person vouching for you closes deals no structure would have won and no comparison frame influenced. That dynamic is idiosyncratic and invisible to the lens used here. Treat all of this as guidance on where effort is likely to pay, not a prediction about any specific deal. Local reality outranks the model every time they disagree.
Section 8
The fitness test
You are ready to restructure and publish your managed services pricing if most of the following is true. You know your cost to serve for each existing client, at least approximately, in delivery hours. You have enough clients, roughly fifteen or more, that patterns are visible rather than anecdotal. You are losing deals at the proposal stage, which means buyers are comparing you and the comparison frame is the problem. You can describe your ideal client in one sentence. And you have the financial room to turn down a bad-fit deal in the next ninety days, because a published floor you cave on is worse than no floor. You are not ready if any of the following is true. You have fewer than about ten clients and are still discovering what you sell, so publishing structure freezes a definition you have not settled. You cannot state your delivery cost per client, which makes any tier boundary a guess and per-outcome pricing reckless. Most of your revenue sits in one or two accounts, so a published floor is a negotiation you cannot afford to win. Or almost all of your business arrives through referral and you rarely face a competitive proposal, so nobody is comparing you and the comparison frame is not your constraint. Fix distribution first. One more disqualifier, worth naming plainly. If you publish structure and then quote outside it whenever a prospect pushes, you have not built a signal. You have built a document proving your commitments are negotiable, which is worse than the quote-only position you started from. Structure works because it is costly to hold. Remove the cost and you remove the information.