FMCG training is shifting from recording course completions to recording capability — who is signed off on which task, right now. The driver is operational rather than educational: lines are reconfigured constantly, audits ask who is certified on a critical control, and field sales teams turn over fast. The World Economic Forum expects 39% of workers' core skills to change by 2030, and a 2026 survey of frontline-heavy industries found employers rate their own training far higher than their employees do.
Key takeaways
- The measurable shift is from completion data to capability data — from “did they finish the course” to “are they signed off on the task”.
- Frontline learners do not reject training; they reject generic training. In a 2026 survey, 51% of those who called their training ineffective said it was too general and around 4 in 10 wanted more hands-on practice.
- Field sales and plant operators have opposite constraints but the same fix: role-specific content, assessed against a defined standard, visible to the person doing the job.
- Nearly a third of employers report losing the equivalent of a full working day to making up for skills shortfalls — that figure is the honest starting point for an ROI case.
- Most of this is configuration, not new software. Moodle's competency framework already models it; what is usually missing is the reporting layer people will actually look at.
What actually changed
For twenty years the FMCG learning stack answered one question well: did this person complete the training. Enrolments, scores, certificates, a compliance report at the end of the quarter. It was built for audit defensibility, and for that it worked.
What operations asks is a different question. Can this person run the changeover on Thursday? Not did they attend the course about it — can they do it, to standard, unsupervised, today. The completion model cannot answer that, and everybody using it knows this, which is why the answer usually lives in a spreadsheet on a supervisor's laptop.
The shift underway across FMCG is to make the right-hand column the system of record. Course completion does not disappear; it becomes one input into a proficiency judgement rather than a proxy for it.
The pressure behind the shift is not pedagogical. The World Economic Forum's Future of Jobs Report 2025 puts it at scale: employers expect 39% of workers' core skills to change by 2030, and project that 59% of the workforce will need training in that window — while 11 in every 100 workers will need training and not receive it. On the supply side, Deloitte and The Manufacturing Institute estimate US manufacturing could need 3.8 million new workers by 2033, with 1.9 million roles going unfilled if the gap persists. When you cannot hire the capability, you have to build and track it.
Three problems, stated plainly
1. The plant: readiness is invisible until it is urgent
A packing line gets reconfigured for a different SKU and the required skill set changes with it. The supervisor staffing tomorrow's shift needs to know who is signed off on that changeover — not who attended a course about changeovers in March. In most plants that knowledge is held informally, by one experienced supervisor, and it walks out of the building when they take leave.
2. The field sales team: training is generic, and they know it
Field sales in FMCG is high-turnover, distributed, and almost entirely mobile. The training that reaches them tends to be the same deck the whole country gets. The 2026 survey of frontline-heavy industries by Chegg, reported by HR Dive, is blunt about what that produces: 77% of employers believe their training is effective; only 58% of employees agree. Among employees who said it was not effective, 51% said it was too general, around four in ten wanted more hands-on practical learning, and 71% said their training had not led to any change in their pay or position.
That last figure is the one worth sitting with. If completing training changes nothing a learner can see, the rational response is to click through it.
3. The audit: the question is about people, not courses
A retailer or food-safety auditor does not ask how many people completed the hygiene module. They ask who is certified on a specific critical control, and to see the evidence. That is a proficiency question, and answering it from completion records means reconstructing an answer under time pressure. The same survey found nearly a third of employers spend the equivalent of a full working day making up for skills shortfalls.
Composite scenario
What the transition looks like in practice
A mid-sized food and beverage manufacturer runs several plants and a field sales force selling into general trade. Training exists and is well-intentioned: an LMS with a solid compliance catalogue, an annual refresher cycle, completion reporting into HR.
The operational picture is different. Line supervisors keep their own competency spreadsheets because the LMS cannot tell them who is cleared for a given changeover. Field sales completes the national product module but distributor-level conversations still vary by territory. Audit weeks pull two people off their jobs to assemble evidence that already exists, scattered.
The change is not a platform migration. It is modelling the job as a framework — duty areas with skills nested beneath them — assigning it to people as a plan, linking the skills that can be taught to courses, and leaving the ones that need watching to an assessor. Completion data keeps flowing in; it just stops being the answer on its own.
This is a composite drawn from patterns common across FMCG manufacturing and distribution. It is not an account of a specific customer, and it carries no outcome figures, because outcomes depend entirely on where an organisation is starting from.
Building the ROI case honestly
Most published ROI figures for training are unusable, because they come from someone selling training. So rather than offer a number, here is the arithmetic — five cost lines that a capability model reduces, each one measurable inside your own operation before you spend anything.
The one external anchor worth using is the survey figure above: nearly a third of employers lose the equivalent of a full working day to skills shortfalls. Start there, then measure your own.
| Cost line | What to measure in your operation |
|---|---|
| Supervisor reconstruction time | Hours a week spent answering “who is cleared for this” and maintaining the shadow spreadsheet, × loaded hourly cost × number of supervisors × 52 |
| Rework and scrap attributable to capability | Your existing rework rate, filtered to incidents where the operator was not yet signed off on the task |
| Line downtime waiting for a qualified operator | Downtime hours logged with a “waiting for trained personnel” reason code × contribution per line hour |
| Audit preparation | Person-days spent assembling training evidence × number of audits a year × loaded daily cost |
| Time to productive for a new starter | Days from induction to unsupervised work × daily contribution × annual intake — the line that moves most in a high-turnover function |
Set against that: the platform, the effort to model the framework once, and assessor time — which is the honest cost most business cases leave out, because someone qualified has to watch the work being done and record a judgement, and that time is real.
If those five lines do not add up to more than that cost in your operation, the case is not there yet. That is a legitimate answer, and it is better found with your own numbers than with someone else's.
Where the completion model still wins
- Regulatory training where attendance itself is the requirement
- Broad, low-stakes awareness content across a whole population
- Cheap to run — no assessor time, no framework modelling
- Reporting everyone already understands
- Adequate when roles are stable and tasks rarely change
Where the capability model earns its cost
- Roles where the standard is observable performance, not knowledge
- Operations that reconfigure often, so required skills shift with the work
- Audits that ask who is certified on a named control
- High turnover, where time-to-productive is a real cost line
- Anywhere a supervisor already maintains a spreadsheet the LMS should own
Need something custom-built?
Most of this is configuration rather than new software — Moodle's competency framework already models duty areas, skills, ratings and proficiency. What is usually missing is a reporting layer people will actually look at. We build and run Moodle for manufacturers and distributors, including the framework modelling and the reporting on top of it.
Model one role before you model the catalogue
Pick a single role with a clear standard — one line, one territory — and take it end to end: framework, plan, course links, assessor ratings, reporting. A framework built for the whole organisation before anyone has used one is the most common way these projects stall.
Frequently asked questions
What is a skills matrix in an FMCG context?
A structured chart of a job: duty areas across the top, and the specific tasks that make up each duty listed underneath, with each person's status against every task. In FMCG it typically covers hygiene and food safety, line operation, quality and batch records for plant roles, and territory, category and negotiation skills for field sales.
How is competency-based training different from course-based training?
Course-based training records what someone has completed. Competency-based training records what they have been assessed as able to do, to a defined standard. A course can be one route to a competency, but an assessor watching the task performed is another, and in operational roles it is usually the one that counts.
Do we need to replace our LMS to do this?
Usually not. Moodle's competency framework already supports frameworks, learning plans, rating scales, proficiency and course links. The gap in most deployments is that the feature is switched off or that its default reporting screens are lists rather than something a supervisor will use.
How do you train a field sales force that is never in one place?
Short, role-specific content on the device they already carry, with the assessment tied to observable behaviour rather than a quiz score. The survey evidence is consistent that generic content is the main complaint, so segmentation by territory, channel and category matters more than production value.
What is a realistic ROI for this kind of programme?
Anyone quoting you a single figure is guessing. The measurable lines are supervisor time spent reconstructing status, rework attributable to unqualified work, downtime waiting for a qualified operator, audit preparation, and time to productive for new starters. Measure those five in your own operation before and after — that is the only ROI number that will survive scrutiny.
How long does it take to model a framework?
It depends far more on how well the job is already documented than on the software. Organisations that already maintain job task analyses or a training matrix are largely transcribing something that exists. Organisations starting from scratch are doing job analysis first, and that is the real work.
Where to start
The organisations getting value here are not the ones that bought the most sophisticated platform. They are the ones that picked a single role, modelled it honestly, decided which skills can be evidenced by a course and which need somebody watching, and then made the result visible to the person doing the job.
If you want to see what that looks like on screen, we wrote about the skills matrix dashboard we built for exactly this, and about running an LMS for a deskless, multi-plant manufacturing workforce.
Want to map one role and see what the reporting would tell you?
Book a Free Demo