How to Build a Member-Driven Learning Catalog

A member-driven learning catalog is one built from what members actually need and do, not only from what they say in a survey, and the difference between those two things is where most association catalogs go wrong. Nearly every association I work with has run the survey. They asked members what topics they wanted, built courses against the answers, and watched enrollment come in below expectation. The survey was not useless, it was just one of three signals, and on its own it is the weakest. This guide covers the three demand signals worth listening to, how to collect each one, and how to turn them into catalog decisions you can defend to a board.

Key takeaways

• Stated demand is the least reliable signal. What members say they want in January is not what they enroll in come renewal.
• Revealed demand is already in your data. Search terms, abandoned enrollments, and completion rates tell you more than any survey.
• Mandated demand outranks both. What a licence or credential requires is not a preference, it is a purchase members must make somewhere.
• Continuing education is a top reason people join. MGI's benchmarking puts CE at 42% of joining reasons, up from 39% two years earlier.
• Members should shape the catalog, not own it. Accreditation standards and clinical accuracy are not up for a vote.
• A catalog needs a spine and a map. The three signals tell you what to build. A competency map tells members what it all adds up to.

What does member-driven actually mean?

It means your catalog is built from evidence about members rather than from the education committee's assumptions or last year's schedule copied forward. It does not mean members decide what you publish. That distinction matters, because the version where members vote on topics produces a catalog that chases enthusiasm and neglects obligation.

The useful framing is that members generate three different kinds of demand signal, and a good catalog reads all three. Most associations read one.

Why does asking members what they want mislead you?

Because stated preference and actual behaviour diverge in predictable ways, and surveys capture only the first.

People overstate interest in aspirational topics. Leadership, strategy, and innovation reliably score well on surveys because members want to be the kind of professional who studies those things. Enrollment lands elsewhere, on the practical and the required.

Surveys also suffer from who answers them. Your most engaged members respond, which means you are sampling the segment least representative of the ones you need to retain. And a survey asks about topics in the abstract, stripped of the two factors that actually drive enrollment: whether it carries credit, and whether it fits in an evening.

There is a timing problem as well. Surveys capture interest at the moment of asking, but enrollment decisions happen months later against a different calendar, a different workload, and a renewal deadline that was not front of mind in the spring.

None of this makes surveys worthless. It makes them the input you weight lowest and validate against behaviour.

What are the three demand signals?

Read them together, and when they conflict, trust them in the order below.

The ordering surprises people, because stated demand is the one that feels most like listening. It is also the one most easily distorted by whoever attended the meeting.

How do you collect each signal?

Mandated demand comes from mapping your members' licences and credentials to their renewal requirements. How many credits, in what categories, on what cycle, and which of those categories you currently cover. The gaps in that map are the highest-confidence catalog opportunities you will ever find, because members are obligated to buy them from someone.

Revealed demand is sitting in systems you already run. The queries members type into your catalog search that return nothing. Courses with high enrollment and poor completion, which signal a real topic served badly. Waitlists. Repeat viewing of specific recorded sessions. Which conference sessions filled first. This is behavioural evidence and it costs nothing to collect, which is why it is strange that so few associations look at it. We go deeper on this instrumentation in the metrics you should be tracking in your association's LMS.

Stated demand is worth gathering, with better questions. Instead of asking which topics interest members, ask what task they could not complete last month, what they had to look up outside your resources, and where they went instead. Concrete recall beats abstract preference.

How do you turn signals into catalog decisions?

Run it as a standing cycle rather than an annual planning offsite.

1. Build the requirement map first. Every credential your members hold, its renewal cycle, and your current coverage against it. This is the spine of the catalog.
2. Pull the behavioural report quarterly. Failed searches, low-completion courses, waitlists, and drop-off points.
3. Layer stated demand on top. Use it to break ties between opportunities the first two signals rank similarly, and to spot emerging topics that have not shown up in behaviour yet.
4. Score candidates on obligation, reach, and effort. A topic that meets a requirement, serves a large segment, and reuses existing faculty beats a novel topic every time.
5. Commit to a small number and instrument them. Three well-built courses tied to real requirements outperform twelve built on enthusiasm.
6. Close the loop publicly. Tell members which of their inputs produced which course. Participation collapses when feedback disappears into a committee.

How do you turn demand signals into a competency map?

The three signals tell you what to build. They do not tell a member what your catalog adds up to. Those are different questions, and most catalogs only ever answer the first.

Think about a member four years into their career who wants to move into a management role. They are not asking what courses you offer this autumn. They are asking what they need next, and a list sorted by topic, format, or credit type cannot answer that. Neither can a search box. The member either works it out themselves or gives up and asks a colleague.

The fix is not rebuilding the catalog. It is tagging the catalog you already have against a competency framework, so every course sits somewhere on a path rather than floating in a list.

Do not invent the framework

This is where associations lose a year. A taxonomy your education committee designs in a workshop feels productive and drifts within twelve months, because nothing outside the room depends on it. Borrow a framework your members are already measured against instead.

• Medical societies and CME teams: the six ACGME and ABMS core competencies, which are patient care, medical knowledge, practice-based learning and improvement, interpersonal and communication skills, professionalism, and systems-based practice. Every physician member was trained and assessed against these, so the labels need no explanation.
• CPA societies and CPE providers: NASBA's Fields of Study, which state boards already accept for credit. Your reporting has to speak this language anyway.
• Anyone running a certification: your own exam blueprint. If you have a credential, the blueprint is already a competency framework that a psychometrician was paid to build and defend. Reuse it, and our guide to psychometric test development covers how those blueprints get built.

The point of borrowing is that the framework survives a change of staff, a change of committee, and a change of strategic plan, because it belongs to someone else.

The tagging worksheet

Tag each existing course on five columns. It is dull work and it is usually a day, not a quarter.

Two rules keep the worksheet honest. One competency per course, because a course tagged to four competencies is a course nobody can find. And role must be a decision, not a hedge: if you cannot name the stage a course is for, that is a finding about the course, not about the framework.

What tagging unlocks

Once the catalog is tagged you can show a member the gap between what they have completed and what their role or credential requires. That view is the difference between a catalog and a reason to renew, and it is the thing a course list has never been able to do. It also makes the requirement coverage number easier to compute, because coverage becomes a query rather than a spreadsheet someone maintains by hand.

OasisLMS supports this directly. Courses carry competency tags, learning paths are built from those tags rather than hand-assembled, and a member sees their remaining gap on their own dashboard instead of in an email from staff.

What should members not decide?

Three things, and being clear about this protects the whole exercise.

Members do not decide clinical or technical accuracy. Popularity is not evidence, and an accredited activity has to meet a standard regardless of how well it polls.

Members do not decide accreditation compliance. Disclosure, independence from commercial influence, and content review cycles are fixed obligations.

Members do not decide what gets retired. Retirement is driven by content currency and accreditation review dates, and a course can be simultaneously well liked and out of date.

What members do shape is topic priority, format, depth, and timing. That is a large and genuinely valuable remit, and framing it honestly avoids raising expectations you then have to override.

How do you know it worked?

Compare each new course against the signal that produced it. Courses built from mandated demand should show high completion and credit claim rates, because the audience had to be there. Courses built from revealed demand should show strong enrollment relative to promotion effort. Courses built from stated demand alone are the ones to watch closely, since that is where the survey-to-behaviour gap shows up.

Track requirement coverage as a standing number too: the share of your members' mandated credits that they could earn entirely within your catalog. Moving that figure up is the clearest expression of a catalog built around members rather than around habit, and it feeds directly into the engagement work we cover in our guide to member engagement.

When should a course come off the map?

A competency map creates a problem a list never had. It shows you your overlaps, and most catalogs have more than anyone expects.

Three ageing courses covering one competency is worse than one current course covering it. The member has to work out which one is authoritative, and there is no way for them to tell from the outside. Meanwhile all three are consuming review cycles, accreditation attention, and support questions.

The rule worth adopting is simple: a competency should have one canonical course and, at most, one supporting piece. When tagging turns up a third, that is a decision point rather than a filing problem. Retire, merge, or promote one to canonical and mark the others as supplementary, and do it in the same session where you found the overlap, because the alternative is a backlog nobody returns to.

The retirement criteria themselves have not changed and are covered in our guide to sunsetting online courses. What the map changes is that it makes the case obvious. A course can be well liked, moderately enrolled, and still the third-best answer to a question you have already answered twice.

Frequently asked questions

How often should we review the catalog against member demand?
Behavioural data quarterly, the credential requirement map annually or whenever a board changes its rules, and surveys no more than once a year. Surveying more often lowers response quality without adding signal.

What if members ask for topics we cannot accredit?
Offer them as non-credit content if demand is real, and be explicit that they do not carry credit. Trying to force an unsuitable topic through accreditation wastes more than it returns.

Do member advisory committees still have a role?
Yes, as interpreters rather than deciders. A committee reviewing behavioural data alongside the requirement map produces far better judgement than one generating topic ideas from personal experience.

How do we handle conflicting demand between member segments?
Segment the catalog rather than averaging it. Early-career and late-career members want different things, and a blended catalog serves neither well.

Is it worth building courses for a small segment?
If the segment has a mandated requirement you can meet, usually yes. Obligated demand converts at rates that discretionary content never reaches, even at low volume.

Do we need a skills taxonomy before we can do any of this?
No, and building one first is the usual way this stalls. Start with the framework your members are already measured against, whether that is the ACGME competencies, NASBA's Fields of Study, or your own exam blueprint. A borrowed framework you can apply this month beats a bespoke one you are still debating next year.

What if our courses map to competencies unevenly?
That is the map doing its job. Uneven coverage is a finding, not a flaw in the method. Six courses against one competency and none against another tells you exactly where the next build goes, and it is usually a more defensible answer than anything a survey produced.

The bottom line

Letting members shape your learning catalog is right, but only if you are honest about which member input is evidence. Surveys tell you what people aspire to. Your own platform data tells you what they actually do. Their credential requirements tell you what they have no choice about. Build the spine from the third, prioritise with the second, and use the first to break ties.

Most associations cannot do this today for a mundane reason: the requirement map lives in someone's head, and the behavioural data is spread across a website, an events tool, and a learning platform that do not talk to each other. Getting those into one place is the practical starting point, and it is the first thing we look at in our guide to choosing a member education platform. If you want to see what your own catalog looks like read this way, book a demo and we will map it with you.