Build, Buy or Commission: Choosing an Onboarding Training Solution
Three documents are open on your desk. A catalogue subscription offering 480 ready-made courses at a price per learner that looks too reasonable. A platform licence with an authoring tool and a promise that your team can build whatever it likes. And a commissioned quote with a storyboard in which a new hire walks through a rendered atelier and picks the right leather for a made-to-order request.
You are choosing between three genuinely different purchases, and the vendor of each describes the other two unfairly. What follows compares what custom onboarding training costs against what the alternatives deliver, written by an agency that builds the expensive option, so the cases where commissioning is the wrong answer have their own section.
One number to hold onto. In a June 2026 survey of 1,000 employers and 1,005 employees reported by HR Dive, 77% of employers thought their training effective while only 58% of employees agreed, and 51% said training was too general or disconnected from their actual work. That gap is what you are buying your way out of.
The Three Options, Stated Plainly
There are three:
Buy. Licence finished content someone else made. Choose from a catalogue, assign it, report on completions.
Configure. Licence a platform with templates and an authoring tool. Your team produces content inside somebody else's structure.
Build. Commission content designed against your brief: your narrative, your assessment logic, your art direction, produced once and owned.
Most large maisons end up with all three, which is the correct answer. The mistake is rarely the choice between columns but applying one to content that belonged in another.
Buy: The Catalogue Is Better Than Its Reputation
Off-the-shelf content is unfashionable in this industry and frequently the right call.
For anything true regardless of which maison you work for, a catalogue course is cheaper, faster and usually better made than what you would produce yourself. Data protection. Fire safety. Anti-harassment. The rules do not change because the logo on the door is Italian.
The catalogue also solves speed: you can assign a course this afternoon to 2,000 people in nine markets and have a completion report by Friday. Custom production cannot compete, and pretending otherwise is how agencies lose credibility.
Where it fails is predictable. The moment content encodes something specific to your maison, the catalogue offers an open-plan office where a smiling advisor greets a client at what is clearly a bank counter. Your new hire on avenue Montaigne knows within eight seconds that this was not made for her, and that costs more than the licence saved: it teaches her training is a box to tick.
Configure: The Middle Path Has a Hidden Payroll Line
Configurable platforms are sold on an attractive premise: buy the engine, keep creative control, avoid agency fees.
For structured, high-volume, frequently changing content this genuinely works: product sheets that refresh every season, market-specific rules, quiz banks, short reinforcement sequences. A good platform lets a regional trainer in Seoul push a corrected price list on Tuesday without a production cycle.
The hidden cost is not the licence. It is that somebody inside your organisation now has a second job as a production studio: scriptwriting, instructional design, art direction, translation management, accessibility, version control across markets, and retiring stale content. That work does not disappear because the tool has templates: it moves onto your payroll, usually onto the person already running induction.
There is a quieter problem. Templates assume you already know what good looks like. Comité Colbert's June 2025 study with MAD, surveying 31 luxury maisons, found only 42% had a competency framework at group level and 29% at regional level. If your definition of a competent second-month advisor is written nowhere, an authoring tool will help you produce a great deal of content that assesses nothing in particular.
Commission: What the Money Actually Buys
When people quote a figure for a custom build they mean the production. The production is not the expensive part.
What you are paying for is the design decisions taken before anyone opens a 3D engine: what the learner must do afterwards, which moments are worth simulating, how failure is handled, what gets assessed, and how it degrades on a five-year-old tablet in a stockroom. The mechanics are in how a serious game is put together.
It also buys a shape of organisation. A custom build is either made by one team or assembled along a chain: lead agency, subcontracted studio, offshore development, translation vendor. In the first, the art director and the instructional designer argue over the same scene in the same room, and a market's correction lands in days rather than at a contract boundary. Ours is one team of about 50 in Paris, which is why a programme ships in under two months, launches in up to 19 languages at once, and belongs to the client outright with unlimited usage.
Commissioning earns its price in three situations. When the content is the brand rather than about the brand, so heritage, gesture and the selling ceremony are the material itself. When the population is large enough that cost per learner collapses: the Dior onboarding experience has reached more than 100,000 people across 100 countries, and at that scale the unit economics stop resembling a production budget. And when the content is stable for two or three years, so the asset outlives the invoice.
Be honest about the mechanism. Sailer and Homner's 2020 meta-analysis found gamification produced a cognitive effect of g = 0.49 and a behavioural effect of g = 0.25, with the size depending heavily on which design elements are used. A custom shell around weak instructional design buys a beautiful thing that changes nothing.
When Commissioning Is the Wrong Answer
Four situations where you should not commission a custom build, from a company that sells them.
Your headcount is too small. A custom build costs the same whether 80 people play it or 8,000. Below a few hundred learners the arithmetic rarely works, and you would get more from a good facilitator and a well-written manual.
The content is generic compliance. If a regulator wrote the source material, a custom narrative adds cost and risk without adding accuracy. Buy it and spend your budget on reinforcement.
Your maison does not have distinct codes. Some retailers compete on price, range and availability rather than ritual. If nobody internally can name three gestures specific to your floor, a commissioned build will invent them, and invented codes are worse than none.
The content changes faster than it can be produced. Anything rewritten every eight weeks belongs in a configurable tool. Custom production suits stable material: brand foundations, the selling ceremony, values, the induction narrative.
A fifth, less comfortable one: if nobody senior owns the programme internally, a custom build will be delivered flawlessly and then quietly stop being assigned when its sponsor changes role. We turn briefs away on all five counts, and an agency that never does is selling rather than advising.
A Decision Test You Can Run Before Lunch
Take each block of your onboarding content and ask three questions.
Does this encode something only our maison does? If yes, it is a candidate for a custom build. If it would be equally true at a competitor, licence it.
How many people, over how many years? Multiply learners by the life of the content before you look at any quote, then check it against what a departure costs. McKinsey's 2024 research put a single frontline retail departure at roughly $10,000, as set out in build the financial case first.
How stable is it? Content that survives two years is an asset. Content rewritten quarterly is an operating cost, and should be tooled as one.
Comité Colbert and Bain reported in September 2025 that European luxury groups spend on average 3.1% of revenue on technology, 63% of it on running what already exists and only 37% on change. The question is rarely whether you can afford a custom build, but which block of content deserves that 37% this year.
Then ask every shortlisted vendor the same four questions: who owns the source files, what a market-level change costs after go-live, whether the output is SCORM-compatible and LMS-ready, and what happens on a low-bandwidth connection in a stockroom. Those answers separate suppliers faster than any demo; the technical half sits in the integration questions your IT team will ask.
Buy the Boring, Build the Distinctive
The strongest onboarding programmes we see are not the most expensive. They are the ones where somebody drew a line down the middle of the content map.
On one side sits everything true everywhere: statutory training, tools, policies, the payroll portal. Licence it, assign it, forget it. On the other, the small number of things that would be wrong at any other maison: how a client is greeted, how a piece is presented, what discretion means here. That side deserves production values, because it is what a new advisor will still be performing in three years.
Buyers get into trouble when they invert this, spending the change budget to make the fire-safety module beautiful while the selling ceremony is taught by a PDF and a colleague with ten minutes between clients.
Frequently Asked Questions
How much does custom onboarding training cost?
Any agency quoting a figure before seeing your brief is guessing. Cost is driven by the number of interactive scenes, the amount of 3D or original illustration, assessment complexity and language count, not by duration. Fix the budget first and ask what design each vendor delivers inside it. That shows who is designing and who is decorating.
Is off-the-shelf onboarding content ever enough on its own?
For a small organisation with no distinctive service ritual, sometimes. For a luxury or premium retailer, almost never, because what differentiates you is by definition not in anybody's catalogue. The realistic pattern is catalogue content for statutory material, custom production for brand codes and the client relationship.
How long does a custom onboarding build take?
Emraude delivers programmes in under two months, which surprises buyers expecting a year. The variable is rarely production speed but how quickly your side supplies source material, agrees what competence looks like and gets validation from the markets. The timeline is a decision-making question, not a technical one.
Can we start off-the-shelf and move to custom later?
Yes, and often the sensible sequence. Run catalogue content for a cycle, measure where new hires struggle in their first ninety days, then commission custom production against the two or three gaps you have evidenced. You will buy a smaller, sharper build than you would have specified at the start.
Emraude designs a gamified onboarding programme for luxury and retail maisons, and will tell you when your content belongs in a catalogue instead. To test the comparison against your own content map, request a demo.
