How to Choose Training Software When Nobody on Staff Is a Trainer

Software does not remove training work; it moves it. List the jobs first: writing the first version of each lesson, updating it when a procedure changes, assigning it to the right people, checking that they understood and answering questions on the floor. Next to each job, write the name of the person who does it today. In a company without a trainer, most lines name the owner or a busy manager. The software you choose should take at least one of those lines off a person's desk, or it will sit unused. Circle the line that costs the most hours, because that is the job to test hardest.

The U.S. Office of Personnel Management gives agencies the same advice for choosing an LMS: define your requirements before you talk to vendors. Its guide warns that without them, the choice drifts toward vendor features instead of your needs. A small company needs less paperwork than a federal agency, but the order still holds. One page is enough.

• The roles you hire for most often, and the tasks each must do alone by the end of week one.
• Where your procedures live today: binders, files, recordings or one person's memory.
• Where staff will open training: a shared computer, their own phones or a tablet at the counter.
• The records you must keep, such as who finished which version of a procedure.
• Who will approve training content before staff see it.

Polished sample courses tell you little. Pick one real procedure that new hires get wrong, and give the same file to every vendor on your shortlist. Ask each one to produce a short lesson, a job aid and a few check questions from it. Then time your own effort: how long setup took, how much you had to correct and who had to be involved. A tool that looks fast in a demo but needs hours of your edits does not take work off your desk. Keep the drafts side by side, so you can compare them later.

The OPM guide says to test an LMS from the learner's side, with plain questions: how people know what is assigned, where they see finished courses and how they take a test. Ask a new or junior employee to try each option on their own phone, with no help from you. Watch where they get stuck. Also ask each vendor whether its learner screens are tested against WCAG 2.2, the web accessibility guideline the W3C published in October 2023. Staff with a visual or motor impairment need to finish the same training as everyone else.

Do not watch demos alone. Bring the person who will approve training content, because they will judge the drafts most honestly. Bring one frontline employee, who will show you whether the learner side works in real conditions. If someone else will add people and groups, bring them too, since admin work is easy to overlook in a demo. Keep the same people for every vendor, so the comparison stays fair.

Use one sheet for every vendor, so each option faces the same questions. Before the first demo, mark each row as must-have or nice-to-have, based on the jobs you listed in step one. Score every row 0, 1 or 2 after testing it with your own material.

• 0 means the vendor could not show it with your material.
• 1 means it worked, but only with a workaround, extra admin or the vendor's help.
• 2 means it worked in your own test, without help.
• Decision rule: drop any vendor with a 0 on a must-have row. For the rest, count must-have scores double and add up the column.
• If two vendors tie, choose the one with the higher scores on first version and updates.

Read the exit terms before the feature list. Ask which courses you can take with you if you cancel, and in what format. Ask whether library courses stay with the vendor. Ask what happens to completion records, and how long you can download them after the contract ends. Ask who inside the vendor's company can see your procedures. Ask how the price changes when you add people or locations, so the second year holds no surprise. None of this is legal advice, so have someone who reads contracts for you check the final wording.

Elvina fits when your training must come from your own procedures and nobody has time to write it. Sources go into Company Knowledge with an owner, and Creation Studio drafts lessons, job aids, quizzes and avatar video lessons from them, available now. Your manager assigns training to groups on the Learning Platform and sees progress, also available now. Staff open their lessons in the Web workspace, which is available now.

Reminders for people who have not started are planned, not available today. At this stage our team does the drafting with you on the platform, while Elvina picks up each step as it is confirmed. Elvina is a poor fit if you mainly need a large library of general courses, or training that must meet a regulator's certification rules. It does not certify anyone or replace required compliance training.

A fictional aquarium maintenance company services fish tanks in offices and restaurants with a small team of technicians in vans, and the owner has never had a trainer. New technicians learn by riding along, and each one picks up a slightly different way to do a water change.

• The owner writes the one-page needs list and names the senior technician as the approver.
• Each shortlisted vendor gets the same tank service procedure and a short video the senior technician recorded.
• A newer technician tries each option on a phone in a parked van, and the owner notes where he gets stuck.
• First version and updates are marked must-have, because nobody has time to write, and one vendor drops out with a 0 on updates.
• Before signing, the owner exports one course from the chosen tool to confirm it can leave in a standard format.

Decide the measures before you sign, and record where you start. Otherwise the choice will be judged on how the demo felt.

• Hours the owner and managers spend on training work each month, by job from your list.
• Days from a procedure change to the updated lesson reaching staff.
• Assigned lessons opened within the first week, from the tool's own records.
• Tasks a manager signs off on during a new hire's first week.
• Corrections the approver makes per lesson, which should fall as sources get cleaner.

How do I choose training software? List the training jobs nobody has time for, write a one-page needs list, then give every vendor the same real procedure. Score each option on the same sheet, with the most weight on who writes and updates lessons. Check the exit terms before you sign.

• How long should a software trial take? Long enough to run one real procedure from source to approved lesson to a new hire finishing it. That shows setup effort, correction work and what learners see. A trial that only shows sample courses tells you about the vendor's content, not about yours.
• What should we ask about leaving a vendor? Ask which courses you keep, in which format, and whether records can be downloaded after you cancel. Standard formats such as SCORM or xAPI let courses move to another system. Library courses may stay with the vendor, so confirm that point in writing.

Public sources, checked on the date shown.

• OPM's guide to choosing an LMS says to define requirements first, so the selection follows the organization's needs rather than vendor features. Source: https://www.opm.gov/policy-data-oversight/training-and-development/reference-materials/procuring_lms.pdf (checked 2026-10-02).
• The same OPM guide advises testing an LMS from the learner's side, including how people find assigned courses and take tests. Source: https://www.opm.gov/policy-data-oversight/training-and-development/reference-materials/procuring_lms.pdf (checked 2026-10-02).
• W3C published WCAG 2.2, its current web accessibility guideline, on October 5, 2023, and says WCAG applies to the web on mobile. Source: https://www.w3.org/WAI/standards-guidelines/wcag/ (checked 2026-10-02).