Training and Development Tools: What Each Category Does, and the Gap Most of Them Leave

Most organisations run a training needs analysis. It is a serious piece of work: interviews, surveys, performance data, a picture of where the capability gaps actually are. It usually concludes, correctly, that different people need different things.

Then everyone goes on the same course.

That gap between what the analysis found and what the calendar delivers is the most expensive thing in corporate learning, and it is worth holding in mind while reading any list of tools, including this one. Every category below is genuinely useful. Almost all of them are useful at the same part of the problem.

Learning Management Systems

The LMS is the oldest category and still the largest. Moodle, Cornerstone, Docebo, TalentLMS, SAP SuccessFactors Learning and dozens of others. An LMS is a system of record: it holds courses, assigns them, tracks completion, and produces the compliance report.

What it is genuinely good at is administration at scale. If you need to prove that four thousand people completed anti-bribery training before an audit, an LMS is exactly the right tool and nothing else comes close.

What it is not built for is whether any of it worked. Completion is the metric an LMS was designed to capture, and completion is not learning. A person can finish a module, pass its quiz, and change nothing about how they work on Monday. The system will record that as a success because, on its own terms, it is one.

Learning Experience Platforms

The LXP category, including tools like Degreed, EdCast and 360Learning, emerged as a response to that. Instead of assigning courses, an LXP curates and recommends them, pulls in content from many sources, and lets people follow their own interests.

It solves a real problem. Mandatory catalogues produce compliance, not curiosity, and an LXP moves the centre of gravity toward the learner.

The trade is that self-direction assumes people know what they need. Sometimes they do. Often the gap someone most needs to close is precisely the one they cannot see, which is why the recommendation engine tends to surface more of what a person is already interested in rather than what their last performance review implied.

Assessment and diagnostic tools

Personality and behavioural instruments: DISC, MBTI, the Big Five or OCEAN model, Hogan, StrengthsFinder, 360 feedback platforms like Culture Amp or Qualtrics.

These are the tools that actually address the personalisation problem, because they produce a picture of an individual rather than a population. A good assessment tells someone something about themselves that they could not have got from a course.

The recurring failure here is not the instrument, it is what happens next. An assessment is usually a one-off event: a report is generated, a workshop discusses it, and then it goes in a drawer. The report describes a person accurately and then has no mechanism to affect how that person works. Ask someone what their DISC profile was and many will remember. Ask what they did differently because of it and far fewer will have an answer.

Coaching and mentoring platforms

BetterUp, CoachHub, Torch, Sounding Board and the wider coaching marketplace. These connect people to human coaches, usually on a subscription, with matching, scheduling and progress reporting built in.

This is the category that most directly addresses behaviour change, because a coach works on the specific situation a person is actually in. When a human coach is good, nothing on this list beats it. Someone who has run your kind of team for twenty years brings judgement that no tool reproduces.

The constraint is arithmetic. Executive coaching typically runs from the low hundreds to over a thousand US dollars per session, which is why organisations that buy it buy it for a small band of senior people. The manager three levels down, who is making decisions about other people's careers every week and was never trained to, does not get a coach. Not because nobody thinks they need one, but because the budget does not reach that far.

Microlearning and habit tools

Axonify, Arist, and the broad category of short-form, spaced-repetition delivery. These take the research on forgetting seriously and space content out over weeks instead of concentrating it into a day.

Spacing genuinely helps retention, and this category is right about something most of the others ignore. Where it tends to stop short is that it spaces out delivery rather than practice. Being reminded of a concept on day nine is better than not being reminded. It is still not the same as being helped through the difficult conversation you actually have on day nine.

Content libraries

LinkedIn Learning, Coursera for Business, Udemy Business, MasterClass at Work. Enormous catalogues, low cost per seat, professional production.

For breadth and for self-motivated learners these are excellent value, and the economics are hard to argue with. The well-documented weakness is completion: enterprise licence utilisation is often reported in the low double digits, and completion rates lower still. A library that nobody opens is cheap per seat and expensive per outcome.

The pattern across all six

Read those categories together and something becomes visible. They are overwhelmingly optimised for the middle of the process: getting content to people, tracking that it arrived, and making the arrival more pleasant or better spaced.

The middle is not where learning usually fails. It fails at the two ends.

The gap before: the plan is not personal

The needs analysis identifies fifty different gaps and the calendar delivers one course. Assessment tools can close this gap, and mostly do not, because the assessment and the learning plan live in different systems and nothing connects the result of one to the content of the other. The instrument tells you where someone is. The catalogue is the same for everybody regardless.

The gap after: nothing follows the room

This is the larger one. Hermann Ebbinghaus described the forgetting curve in 1885 and it has held up: without reinforcement, most of what is learned decays within days. Every practitioner knows this. Very little of the tooling is built around it.

What would close it is not more content. It is something present at the moment the learning is needed, which is usually weeks later, usually unscheduled, and usually at a point when the person feels least able to go looking for a module. The situation arrives on a Tuesday afternoon in a conversation that is going badly. That is when the training was supposed to pay off, and that is precisely when no tool is in the room.

What to ask when you are choosing

If you are evaluating tools, these questions separate the categories faster than a feature comparison will.

1. Does what this person is shown depend on anything about this person? If the answer is the same catalogue for everyone, the needs analysis is decorative.
2. What happens on day thirty? Not what content exists, but what actively reaches the learner after the session has ended.
3. Can it be there when the situation is, rather than when the calendar is? Scheduled support is support that arrives at a convenient time rather than a necessary one.
4. What is the metric? If the answer is completion, you are measuring attendance. Ask what behaviour the tool claims to change and how it would know.
5. Who does it reach? A tool bought for the top fifty leaders leaves the people making the most daily decisions about other people entirely uncovered.

Where our own tool sits

We build one of these, so treat this section with the scepticism it deserves and check it against the questions above.

Cultiv8Hub is built specifically for those two gaps rather than for the middle. Before: nine assessments, free, whose results generate the learning plan rather than sitting beside it, so what a person is shown depends on what the assessment found. After: an AI mentor called the Cultiv8tor that works from that same profile, remembers every previous session, and is available at the hour the situation actually occurs rather than the hour a coach has free.

Two honest limits. The mentor is AI. It has never been passed over for promotion or had to let someone go and then drive home, and if you can give someone a good human coach, do that. What AI changes is the arithmetic: it reaches the manager three levels down who was never going to get a coach at any price. And this does not replace classroom training. Some things need a room and other people in it. It is built to make what happened in that room survive contact with a real week.

You can take all nine assessments and have five conversations with the mentor without paying anything, which is a more useful way to judge the claim than reading us describe it.

The short version

Every category on this list does something well. An LMS runs compliance, an LXP serves the curious, assessments produce genuine self-knowledge, human coaching changes behaviour when you can afford it, spacing beats cramming, and libraries are cheap breadth.

What almost none of them do is connect the diagnosis to the plan, or put anything in the room on the Tuesday afternoon when it matters. When you are comparing tools, that is the question worth asking first, because it is the one the feature list will not answer.