Docebo alternatives for a mid-market L&D team

If your Docebo renewal is ninety days out, the most useful thing you can do is not build a feature comparison. It is to work out what the platform costs you per learner who actually finished something last year, and what a replacement would cost to reach the same place. For most mid-market L&D teams the licence line turns out to be the smaller half of that number: implementation, professional services and internal admin time make up the rest, and none of them appear on the renewal quote.

Docebo does not publish pricing. Third-party benchmarks collected by trainingcost.com put typical contracts above $25,000 a year. If yours started around there, the number you renew at will not be the number you signed, because the things that drive it — active users, modules, services days — all move independently of each other. That is not a scandal. It is what a quote-only contract does, and it changes how you have to run the evaluation.

What quote-only pricing does to an evaluation

The practical effect is that you cannot do arithmetic until you are already inside a sales process. That reorders the evaluation: you shortlist on demos and feature grids, which are the parts of a platform that differentiate least, and you find out about the parts that actually differ — implementation weight, admin overhead, what your team can change without raising a ticket — after signing.

• You cannot build a comparison spreadsheet early. The one artefact that would make the decision rational is the one you get last.
• Timing becomes part of the price. A quote produced in the vendor's Q4, against a renewal date they can see, is not the same quote as one produced in February.
• Your own numbers become the negotiation. Active user counts, growth projections and go-live dates are inputs to the quote. Give them precisely, and only when you have checked them.

What "implementation" actually means

Implementation is the line most mid-market teams underestimate, because the word sounds like setup. It is the work of moving your catalogue, your people, your enrolment rules and your reports into a different system, then proving all four still behave the way compliance expects. That is a project with a plan, not a fortnight of configuration.

• Moving the catalogue: courses, learning plans, categories, and the enrolment rules that put the right people into them automatically.
• Re-testing every SCORM package. A package that reports completion correctly in one player can fail to in another, and scoring statements are where it usually breaks.
• The user feed: the HRIS or SIS connection, the field mapping, and the deprovisioning rule nobody wrote down.
• Rebuilding the group and permission model, because it never maps one-to-one between two products.
• Historical completion records — what exports, in what format, and whether issued certificates survive the move as verifiable records or as PDFs in a folder.
• Every scheduled report someone in compliance depends on, rebuilt and then reconciled against the old numbers until they agree.
• Branding, domains, email sending, and the identity integration with real test accounts.
• Admin training and a user-acceptance window with people who will actually complain.

The questions to ask about professional services

Professional services is where a configurable platform makes back what it discounts on licence, and where a team without a dedicated L&D operations function quietly becomes dependent. Ask these before signing anything — of your incumbent at renewal as much as of a new vendor.

1. What is the day rate, and who is on the day — vendor staff, or an implementation partner?
2. Which lines in the statement of work are fixed price, and which are time and materials?
3. What is the change-request process, and what has to be true before a change becomes billable?
4. How many services days are included in the renewal, and do unused days expire?
5. After go-live, which configuration changes can our own admins make, and which need a services ticket?
6. If we want a new sub-portal, a new report or a new integration in year two, is that a project or a setting?
7. Who holds the knowledge of our configuration? Is there a workbook we own, or does it live with one consultant?
8. What is the support SLA by severity, and what is explicitly excluded from it?
9. If our customer success manager changes, what carries over besides the account record?
10. What does exit look like — what exports, in what format, over what notice period, and is there a fee?

Working out your real cost per learner

The renewal quote is one input to a number you should be able to state in a sentence. Build it once, honestly, and every conversation afterwards gets shorter.

annual cost per active learner

  =  ( annual licence
     + support / success fee
     + implementation ÷ contract years
     + professional-services days billed
     + integration maintenance (your engineers' time)
     + LMS admin time (FTE × loaded salary) )

     ÷ learners who completed at least one thing this year

The denominator is the argument. Divide by people who finished something, not by seats purchased. Seats are what you were sold; completions are what you got. The gap between the two is the honest measure of whether the platform is doing its job, and it is usually the number that makes a renewal conversation short — in either direction. If that cost is low, you have a well-run programme and a good reason to stay. If it is high, you know by how much — which is the only way to negotiate without bluffing.

Then check what your contract meters, because registered users, monthly active users and purchased seats produce three different bills from the same population. LearnUpon, another quote-based product, is benchmarked as per-active-user. TalentLMS publishes its prices outright, so you can model it before anyone calls you — and it is a deliberately lighter product, which is the trade.

Be fair about the depth — then count what you would rebuild

Docebo is an established enterprise LMS with extensive configuration, white-labelling and a large partner ecosystem. That is not a courtesy; it is the reason the price is what it is. Depth has to be built, documented, supported across time zones, and kept working across a decade of customer estates that all diverge from one another. Somebody pays for that. If you use the depth, the renewal is the cheap answer and this post is a budgeting exercise.

So audit it. This is a boring afternoon and it decides the entire decision.

• How many of the automation rules configured in year one still run, and does anyone know what they do?
• How many custom reports are opened by a human being in a given month?
• Which integrations carry data that someone actually reads or acts on?
• How many portals or branded domains are live, as opposed to configured?
• Which modules appear on the invoice that nobody has logged into this quarter?
• How many admin hours a month go into keeping the thing fed?

Then count the rebuild, before you fall in love with a shortlist. Content packaged as SCORM travels; content authored natively in one tool generally gets rebuilt. Completion history exports as a file, not as a working system, so if a regulator can ask you for a 2023 record, you need to know where that record lives afterwards. None of this is a reason to stay. It is a reason to put a real number on switching instead of assuming it is zero.

Where Lurno fits, and where it does not

We build Lurno, so read this section as the interested part of the post. It is here because the shape of the product answers a specific version of the problem above, and it is useless for other versions.

Lurno is multi-tenant. The structure it is built around is an organisation containing other organisations: nested sub-organisations, an internal branch tree where a role granted at a branch cascades down it, per-organisation custom domains with automatic TLS, and white-label branding. That fits a training provider running separate branded academies for client companies, or a group whose business units need real separation rather than folders. Tenant isolation is enforced in the database — separation enforced underneath the platform — not only in application code. Prices for the first three tiers are published with their member caps on the pricing page, so you can model the licence line before speaking to anyone.

Three things to know before that model is worth anything. SCORM 1.2, SCORM 2004, xAPI and LTI 1.3 are in development — they are modelled in the product and the runtime is being built, so if standards conformance gates your rollout, that is a conversation to have before a pilot rather than after one. There is no SAML or OIDC single sign-on today. What ships is partner sign-on (silent SSO); SAML/OIDC, MFA and passkeys are on the roadmap and should not be counted as available. There is no self-serve signup or checkout — payments are designed and not built, and accounts are created by invitation, so onboarding involves us either way.

The side-by-side, without the sales copy wrapped around it, is on the Docebo comparison page; the corporate L&D view covers the rest of the platform.

Before you take the renewal call

1. Pull last year's real figures: licence, services days billed, admin hours, active learners, completions.
2. Ask for the renewal quote in writing with the same line items as the original contract, so you can diff the two.
3. Ask which lines are indexed, to what index, and whether the uplift is capped.
4. Run the depth audit and write down the five things you would genuinely lose by moving.
5. Get quotes from two platforms that publish pricing, so you hold a unit cost to compare against, even if you never intend to move.
6. Ask the exit question — export format, notice period, fee — while you still have a year left, not while you are leaving.
7. Decide your walk-away number before the call rather than during it.

The point of this work is not to leave. It is to be able to say what the platform costs per learner, and to say it in the first ten minutes of the call. A vendor with a good product has no reason to fear that conversation, and one who does has told you something useful.