Where the disclosure figures actually come from

A count of people, not of enrollments. Someone who completed four courses is one trainee. It is an easy error to make, and it inflates the figure in a way anyone comparing it with your headcount will notice.

As published — confirm on Qiwa — the count can include more than employees. Co-op trainees placed with you by educational institutions, graduates, and jobseekers who received training may be counted too, and Saudi and non-Saudi trainees reported separately. If you run a co-op or graduate program, those people often belong in the number, and they are usually tracked outside whatever system holds employee training.

The average is total person-hours divided by the number of workers. The real risk is counting session-hours — the length of the session once — instead of hours times attendees. Example: a six-hour workshop for ten people is 60 person-hours. A record that logs it as six session-hours understates the average tenfold.

So the record needs the duration of each activity and who attended it. For training delivered outside the platform, set the duration when you assign it. A workshop assigned as "attend the session" gives you attendance. Assigned as "attend the six-hour session", it gives you the hours as well.

The total is easy to pull from finance. Deciding what belongs in it is harder. As published — confirm on Qiwa — it may count fees paid to training providers and to external and internal trainers, the share of wages of staff who spend part of their time running training, allowances paid to trainees during training (their wages are excluded), and indirect costs such as materials, equipment, and venue rent.

For anything that list does not settle, make the decision once, write it down, and apply it the same way next year. Example: a figure that moves 40% because the accounting treatment changed invites a question.

As published — confirm on Qiwa — the form may ask for the year’s training spend as a proportion of the wages paid to workers in the same year. Once the cost figure is settled and payroll gives you the wage total, it is a division. Check that both halves cover the same population and period: company-wide training spend against the company-wide wage bill, not one division’s spend against everyone’s wages.

The plan for the year ahead asks for the same ratio using the planned training budget. Keep the two apart. Last year’s figure is what you spent, and next year’s is what you intend to spend.

As published — confirm on Qiwa — the form may ask for the types and activities of training delivered, broken down by trainee group: workers, co-op trainees, graduates, and jobseekers. The published list of types is the practical one to use. It names courses, workshops, lectures and seminars, e-learning, study the employer pays for, professional mentoring, on-the-job training, and job rotation.

Sorting a year of training into types at filing time means reading back through all of it and making judgment calls, and two people will sort the same session differently. Record the type when each activity is created. It takes seconds, and the type can matter beyond the breakdown: as published — confirm on Qiwa — each type may convert into training units at a different rate, such as two hours of a course for one unit and four hours of e-learning for one.

All of this comes down to one principle. The disclosure is a reporting problem, and reporting problems are solved when the training happens.

In practice that means one record covering internal and external training, a duration and a type attached to each activity when it is scheduled, and attendance recorded per person. Salalem keeps that completion record, with the evidence behind each external activity, so you can check the figures against it. The filing stays yours. The principle holds whatever you use: a well-kept spreadsheet beats a platform that only sees half the training.