The first objection anyone raises to measuring behavioural intent is this: intention is not the same as behaviour. Someone can say they intend to change a specification decision after attending a CPD session and then not change it. So what exactly is being measured?
It is a fair objection. It deserves a direct answer before anything else is said.
Intention is not behaviour. No serious researcher has ever claimed otherwise. But intention predicts behaviour — measurably, consistently, substantially better than the alternatives currently used to evaluate CPD. That distinction is the whole argument. And it is worth being precise about what it means, because the alternative — measuring satisfaction, recording attendance hours, or not measuring anything at all — is not a neutral choice. It is a choice to measure things that tell you almost nothing useful about whether professional practice changed.
What the research actually says
In 1991, Icek Ajzen published what became one of the most replicated models in social psychology: the Theory of Planned Behaviour. The central finding was straightforward. For volitional behaviours — actions a person can choose to take — stated intention is the most proximal predictor of whether they actually do it. Everything else (attitude, social norms, perceived capability) operates through intention. Remove intention from the model and predictive power collapses.
A decade later, Armitage and Conner (2001) ran a meta-analysis across 185 independent studies testing the model empirically. Across those studies, intention accounted for 27% of the variance in subsequent behaviour. That figure is worth pausing on. Not because it is large — it is not — but because the comparison class matters. Satisfaction ratings after CPD sessions have been shown in multiple analyses to have near-zero correlation with whether attendees changed professional behaviour (Alliger et al., 1997). Hours of attendance have no established correlation with behavioural outcomes at all. Against those comparators, an instrument that accounts for 27% of the variance in subsequent behaviour is not a weak proxy. It is the strongest near-term predictor that can be measured at the point of a learning event.
There is also an honest caveat that needs naming plainly: not all stated intentions become actions. Sheeran and Webb (2016), synthesising the research on what they called the “intention–behaviour gap,” found that a medium-to-large shift in intention produces only a small-to-medium shift in behaviour. There is a documented population of what they termed “inclined abstainers” — people who genuinely intend to act and do not. This is not a fatal objection to measuring intent. It is, however, the reason why intent measurement alone is insufficient, and why the Realisation Rate — tracking whether stated intention was acted upon at 30 and 90 days — is not a feature of this platform but a structural necessity of its methodology.
The precision of the intention also matters. Gollwitzer and Sheeran (2006) demonstrated in a meta-analysis of 94 studies that forming a specific implementation intention — if situation X arises, I will do Y — produces substantially better goal attainment (effect size d+ = 0.65) than forming a vague general intention. A CPD submission that asks “what specifically will you do differently?” is not a survey. It is an intervention. The act of articulating a specific commitment makes enactment measurably more likely than not articulating it.
What CPD Impact measures, and why those dimensions
Four dimensions of intent are recorded from every submission:
| Dimension | What the learner is signalling | Behavioural signal |
|---|---|---|
| Practice Change Intent | A specific, named change to how they’ll work — not a general sense of value | Strong |
| Assumption Challenge | An existing belief or default has been unsettled, even without a plan yet | Mixed |
| Exploration Intent | Curiosity to investigate further — no commitment to act | Mixed |
| No Change | Content was received but nothing about practice or belief moved | Weak |
Practice Change Intent captures whether the session prompted an intention to change a concrete professional decision — a specification, a design approach, a calculation method, a procurement recommendation. This is the dimension most directly linked to the outcomes CPD providers say they want their sessions to produce: changed professional behaviour. It is weighted most heavily in the composite score precisely because it is the most commercially and professionally consequential signal.
Assumption Challenge captures whether the session prompted a professional to reconsider something they previously held as settled. Not to act immediately — but to question a prior belief. This dimension is weighted second because it is the dimension most directly linked to what professional bodies mean when they describe CPD as “continuing professional development” rather than “continuing professional information.” A session that generates only Practice Change Intent — do more of this, less of that — is operationally useful. A session that also generates Assumption Challenge is epistemically useful. The distinction matters for practice directors evaluating programme quality.
Exploration Intent captures genuine engagement without immediate commitment. The professional will investigate further, read the cited paper, raise the question at a project meeting. This is the intermediate state: not yet Practice Change Intent, not indifference. It is the leading indicator whose follow-through the 30-day check-in is designed to catch.
No Change is recorded without weighting and without suppression. It is the honest answer — present in the data, available to any provider willing to look at it. A session with a high No Change rate is not necessarily a poor session. It may be the wrong session for that audience. The signal is diagnostic, not punitive. But it must be there, unsmoothed, for the dataset to be credible.
The composite score these dimensions produce is not a judgment. It is a distribution. And the distribution tells you something that a satisfaction rating never can: whether the session moved the people who attended it.
The Kirkpatrick question
A useful framework for situating this methodology is Kirkpatrick’s four-level model of training evaluation (1994). Level 1 is reaction: did attendees find it useful? Level 2 is learning: did they absorb the material? Level 3 is behaviour: are they applying it? Level 4 is results: has it changed organisational outcomes?
The industry standard for AEC CPD evaluation, to the extent that evaluation happens at all, is Level 1. Satisfaction surveys and post-session feedback forms. The profession has been running at Level 1 for decades and calling it measurement.
The intent signal operates at Level 3. Not because the session is followed by observation of behaviour — it is not — but because the signal is the most valid near-term predictor of Level 3 outcomes that is feasible to collect at the point of a learning event. The Realisation Rate is the nearest available signal toward Level 4: whether stated Level 3 intentions were enacted in actual professional decisions.
There is a boundary worth naming explicitly. What FRAKT measures is not professional development in the macro, career-arc sense. The kind of development that transforms a competent practitioner into a trusted authority — accumulated over years through complex projects, difficult judgements, peer recognition, regulatory scrutiny — is not visible in a post-CPD submission. That trajectory operates at a level of resolution and time scale no post-session instrument can reach, and CPD Impact does not claim otherwise. What this platform measures is whether a specific session, on a specific day, moved the professionals who attended it toward a specific change in professional behaviour. That is a narrower claim than “professional development.” It is also a more honest one.
What the Realisation Rate adds
Intent is the leading indicator. The Realisation Rate is the lagging indicator.
At 30 days, attendees who submitted a Practice Change or Exploration Intent response receive a short check-in: did they act on the intention they recorded? At 90 days, a second check-in captures whether any action taken has become durable practice, or has faded.
This matters for the platform’s credibility in a way that intent data alone cannot achieve. A CPD with high Practice Change Intent and a high Realisation Rate is demonstrably changing what professionals do in practice. A CPD with high Practice Change Intent and a low Realisation Rate is generating intention without action — and that gap is a question worth asking, both of the session design and of the conditions in the professional’s practice environment that may be preventing transfer.
Hughes, Zajac, Woods and Salas (2020) demonstrated in a meta-analysis of training sustainment research that peer support, supervisor support, and organisational context collectively explain 32% of variance in whether trained behaviours persist over time. The Realisation Rate does not isolate these factors — it observes the outcome. But it creates the dataset in which those patterns can eventually be examined: whether certain disciplines show systematically lower follow-through, whether certain session formats produce more durable intent, whether certain provider types generate higher Realisation Rates for the same level of stated intent.
Limitations
Three things this methodology does not measure, stated directly:
It does not assess whether a specification change was a good specification change. A professional can intend to change a decision, change it, and change it wrongly. The signal records intent and enactment. It does not evaluate quality.
It does not capture passive learning. A professional might retain something genuinely valuable from a CPD session without forming an immediately expressible intent. The signal catches conscious intent at the point of attendance; what happens in the background is not visible.
It does not adjust for context. The work environment — access to decision authority, project timing, organisational culture — determines whether an intention can be enacted regardless of how genuine it is. The Realisation Rate captures this in aggregate, but cannot decompose it.
These are real limitations. Naming them is not a weakness in the methodology. It is what makes the methodology credible. The alternative — attendance hours — has none of these limitations because it has no ambition. It does not try to measure anything about professional development at all.
The standard that already existed
The most striking thing about CPD outcome measurement is not that it is hard. It is that it has been formally required for twenty-six years without anyone building the infrastructure to make it possible.
ISO 10015 has mandated training effectiveness evaluation since 1999. ISO 21001 requires educational organisations to demonstrate that learning activities produce the outcomes they claim. Any CPD provider delivering sessions to AEC professionals operates, in principle, under both standards. Almost none have the measurement infrastructure to comply with either.
Behavioural intent measurement at the point of attendance, tracked through a Realisation Rate at 30 and 90 days, is not a novel invention. It is the practical implementation of an obligation that has existed for a generation. The obstacle has never been the idea. It has been the absence of independent infrastructure to collect, hold, and report the signal without the provider touching it.
FAQs
Behavioural intent is what a person reports they will do differently as a result of an experience — measured immediately after attending a session. Research across 185 studies on the Theory of Planned Behaviour (Armitage & Conner, 2001) shows that stated intention accounts for 27% of the variance in subsequent behaviour — substantially more informative than satisfaction ratings or hours logged. Measuring intent at the point of CPD attendance is the most reliable near-term indicator available of whether a session will change professional practice.
CPD Impact collects four dimensions of behavioural intent from AEC professionals immediately after attending CPD sessions: Practice Change Intent, Assumption Challenge, Exploration Intent, and No Change. These are combined into a composite signal score. Additionally, the Realisation Rate tracks whether stated intentions were acted upon at 30 and 90 days, converting intent data into longitudinal evidence of practice change — the nearest available lagging indicator of whether the session produced a measurable professional outcome.
No. Stated intention does not guarantee behaviour change — and this platform does not claim otherwise. What the research shows (Sheeran & Webb, 2016) is that intention predicts behaviour better than any other measure that can be collected at the point of a learning event. A CPD that generates high Practice Change Intent across a substantial number of attendees is substantially more likely to influence professional decisions than one that does not — regardless of how much the attendees enjoyed the session. The 30 and 90-day Realisation Rate is the mechanism that converts the prediction into evidence.
