What the ADDIE model is
The ADDIE model is a way of ordering the work of training design, also called instructional design, into five stages, where the output of each stage is the input to the next: analysis, then design, then development, then implementation, then evaluation. The name is formed from the initial letters of those five stages in English.
The ADDIE model is not a training style or a teaching method. It is an order of work that says what is done before what, and its output is a programme ready to be delivered. The question it answers is “how do we build this programme?”, not “which method do we use inside it?”. We found no published measurement on which to base a claim that the ADDIE model performs better than other training design models.
Why the ADDIE model insists on an order
A training programme can be built from its end: a topic is chosen, content is requested, and only later does anyone look for who needs it and how anyone will know that it helped. Built in that reversed order, a programme has no measure, because no measure existed when the content was written.
The value of the ADDIE model lies in giving each stage a written output that is handed to the next. The second stage does not begin until a document from the first is in hand, and that requirement on its own can prevent the waste that arises when a stage starts without one.
Analysis, the first stage of the ADDIE model
Analysis is training needs analysis at its three levels, together with the check that comes before it: is it that the employee does not know, knows but cannot do it, or knows and can but does not?
What the ADDIE model adds to that is one point: analysis is a stage with an output that is handed over, not a decision taken in a meeting. Its output is a document stating which gap, for whom, and by which indicator its closing will be recognised. Until those three are written down, the first stage has not ended, however long the discussion of it has run.
Design in the ADDIE model: the measure is written before the content
The output of the design stage is a plan that carries four things:
- Objectives written as observable behaviour. “The trainee prepares a correct accrual entry from a source document” is an objective that can be checked. “The trainee understands the importance of accuracy” is not an objective, because it cannot be checked. Describing competencies calls for the same observable terms.
- How attainment of each objective will be measured, written before the material and not after it. Material written first produces a measure that asks about whatever the material contains; a measure written first produces material that serves what the programme is meant to achieve. Whether a given measure supports the decision it is meant for is the question of test validity.
- The sequence of topics and the method for each part: explanation, practice on a case, or role play training, and whether each part is delivered in person or in another form, as in blended learning.
- The distribution of time: whether the programme is delivered in a single block or over spaced intervals, a decision whose effect is examined in comparing spaced practice with massed practice.
Development, implementation and evaluation in the ADDIE model
Development is making what was planned: the material, the exercises, the measurement tools, and whatever stays with the trainee after the programme ends, such as job aids. It closes with a trial on a small group, which is the cheapest step in the model, but also a step that can be skipped. Two hours with five people can reveal what a month of review on paper does not.
Implementation is the delivery and everything around it: who delivers, where, whether the trainees have actually been released from their work, and whether participation is recorded in a learning management system. One thing that can undo the design at this stage is not a matter of design at all: a trainee who attends while their phone keeps ringing about their work.
Evaluation asks four different questions, measured with different tools. What did the trainees think? Did their knowledge change? Did their work change? Did the indicator that prompted the programme move? The first is the easiest of the four and tells the least; the third is the hardest and tells the most, and it can only be measured once time has passed since the return to work, by reference to the relevant performance standards.
These four are set out as questions measured by different tools, and the definition of the ADDIE model does not attribute them to any numbered model. Dividing evaluation into numbered levels is a separate model in its own right, not part of the ADDIE model, even where the two are presented together.
How the ADDIE model loses its fifth stage
Evaluation is the last stage, and for that reason it is the stage that can be dropped when time runs short or the budget runs out. A programme that was delivered and never evaluated is repeated the following year unchanged, because there is nothing on which to base a change.
The effect accumulates. An organisation that ran six programmes over two years and evaluated none of them knows nothing about its training beyond the number of hours and the number of attendees. When it is asked what its spending achieved, it has no answer; its budget is cut, it delivers less, and the original cause remains in place.
The arrow that returns from evaluation to analysis is what makes the ADDIE model a cycle rather than a line. The result of evaluation revises the gap described in the first stage: either the gap has closed and attention moves to another, or it has not, and the analysis behind it is reconsidered from the beginning. Going from evaluation straight back to delivery repeats what did not work.
The ADDIE model and the decision to build at all
The ADDIE model describes how a programme is built; it does not say whether one should be built in the first place. That is a question of arithmetic, asked after analysis and before design. Take the following figures, which are assumed in order to show the structure:
- Designing and developing the programme takes 120 hours of work at a cost of SAR 150 an hour, so the cost is SAR 18,000, paid once.
- Delivering it with a trainer from within the organisation costs SAR 1,200 for each group of 12 trainees, or SAR 100 per trainee.
- The alternative is an external course at SAR 900 per trainee.
The saving per trainee is 900 minus 100, or SAR 800. The break even point is 18,000 divided by 800, which is 22.5, so 23 trainees. An organisation that will train fifteen employees buys the course; one that will train fifty builds the programme. The figures are assumed to show where the break even point lies and are not a standard taken from any source, and the definition of the ADDIE model sets no duration for any stage and no ratio between the stages.
The calculation carries two conditions. First, the external course can be compared with the built programme only if it actually covers the gap described in the analysis; otherwise the two being compared do not do the same job. Second, a built programme has a working life. If the procedure it was built on changes after a year, a single year’s intake carries the whole SAR 18,000, and the break even point then has to be reached by the people who can be trained within that year alone.
The ADDIE model as an order that allows a return
One criticism made of the ADDIE model is that it runs in a single line, so a flaw is discovered only after money has been spent on every stage. The criticism holds against anyone who applied the model as a line, but it does not follow from the structure of the model.
The cost of late discovery can be calculated. A flaw in a written objective is corrected at the design stage in one hour. The same flaw, found after development, requires the material built on it to be made again, which in the 120 hour example could reach 30 hours, or SAR 4,500. Found after implementation, it adds the time of everyone who attended and did not benefit.
Two reviews are therefore added to the order without removing it: a review of the objectives and measures with the person who owns the work, before any material is made, and the trial on a small group before general delivery. Both are deliberate returns, not departures from the ADDIE model.
When the ADDIE model does not fit
- An urgent need. A change to a procedure that takes effect next week cannot wait for five stages. A written notice and a job aid at the point of work can suit it better, and once the procedure has settled, the question is whether a gap remains that justifies a programme.
- A small number of trainees. The break even calculation above applies: a programme is not built for five people, whose need can be met through mentoring, job shadowing or an external course.
- A subject that keeps changing. Material whose content changes every few months can consume more in updating than it did in building, and it is better kept as a short reference that is updated than as a programme that is redesigned.
- A gap that is not one of knowledge. This is the limit set by the analysis itself: a gap caused by a missing tool, an absent authority or conflicting priorities is not closed by a programme, however well that programme is designed.
What the ADDIE model is not
- A learning management system. It is a tool in which participation is recorded and from which content is made available, whereas the ADDIE model is an order of work that produces the content. An organisation that bought the system without going through the stages has published, on an organised tool, content that was not derived from a gap.
- Train the trainer. That approach prepares the person who delivers; the ADDIE model prepares what is delivered. An excellent trainer delivering a programme designed on the wrong gap delivers it very well to the wrong purpose.
- The annual training plan. It lists what will be delivered, when and on what budget, and the ADDIE model is the order in which each item on it is built. The plan comes first, and the model works inside each of its items.
The ADDIE model and the Saudi Labor Law
In what we read of the Saudi Labor Law (نظام العمل), its Implementing Regulation (اللائحة التنفيذية) and the model work regulation (النموذج الموحّد), we found no provision that imposes a method for designing training.
Training and qualification contracts are a separate matter with their own sources, and nothing about them is to be inferred from the definition of the ADDIE model.
Before the ADDIE model is applied
The ADDIE model is useful when the gap has been described, the measure has been written and the party who will evaluate is known. It is not useful when the request is for a programme on a topic named before anyone asked why, because its first stage is then carried out as a formality to justify a decision already taken.
The test before starting is a single question: what will change in these people’s work, and how will we know three months from now that it has changed? Whoever has an answer can proceed through the five stages; whoever does not has a problem that is not one of design.
This is an explanation of the concept, not legal advice.
A standalone Saudi HR system
One employee file holding the contract, the documents and their expiry dates, the attendance record, leave, salary and end-of-service entitlements. End-of-service, overtime and leave-balance calculations are built into the system.
A standalone system on its own subscription. The connection to Qoyod Accounting is now available.