A single measurement tells you where you are. Two measurements — separated by meaningful time and targeted intervention — tell you whether anything actually changed, by how much, and why. This distinction is not semantic. It is the difference between evidence and assumption, between proof and belief. It is the scientific foundation of PDA's two-phase design.
This statement sounds obvious. Yet the vast majority of learning and development programmes, coaching engagements, and organisational interventions are designed, delivered, and evaluated without one. A workshop is run. Participants fill out a satisfaction survey. The facilitator receives positive feedback. The organisation concludes the programme was successful. None of this constitutes measurement of change — because without a baseline, there is no reference point against which to compare the post-intervention state.
The problem is not that organisations do not care about outcomes. Most do. The problem is that establishing a valid baseline requires investment — in time, in methodology, and in the discipline to measure before the intervention rather than only after. Historically, the tools required to do this rigorously were expensive, slow, and specialist. The result: the L&D industry built an entire culture of programme evaluation that measures satisfaction (easy) rather than behavioural change (hard).
In 1963, Donald Campbell and Julian Stanley published Experimental and Quasi-Experimental Designs for Research — a paper that became the foundational text of evaluation methodology. Campbell and Stanley established a hierarchy of research designs based on their ability to produce valid causal inferences: to determine not just whether something changed, but whether the intervention caused it to change.
At the top of their hierarchy sat the true experimental design: random assignment, control group, pre-measurement, post-measurement. Below that sat a series of quasi-experimental designs for real-world settings where true randomisation is impossible. At the bottom — and explicitly identified as incapable of supporting causal inference — sat the post-test-only design: measuring outcomes after an intervention with no pre-measurement baseline.
The post-test-only design is, structurally, what most L&D evaluation looks like today. Participants are assessed after the programme. Scores are interpreted as evidence of programme impact. Campbell and Stanley demonstrated, more than 60 years ago, why this inference is invalid: without a baseline, the post-programme score could reflect pre-existing capability, natural maturation, historical events, or any number of factors unrelated to the programme itself.
"The fundamental problem of programme evaluation is not measurement — it is the absence of a counterfactual. Without knowing where participants started, we cannot know where the programme took them."Campbell, D.T. & Stanley, J.C. (1963), Experimental and Quasi-Experimental Designs for Research
Donald Kirkpatrick's four-level evaluation model, first published in 1959, remains the most widely used framework for L&D evaluation in the world. The four levels measure reaction (participant satisfaction), learning (knowledge acquisition), behaviour (transfer to the workplace), and results (organisational impact).
Levels 1 and 2 can be measured at a single point in time, immediately after a programme. Levels 3 and 4 cannot — by definition. Behaviour change requires time to manifest and requires a comparison point. Results require a reference against which to measure improvement. Jack Phillips' later ROI methodology, which adds a fifth level (return on investment), requires the same pre-post architecture as Kirkpatrick's Levels 3 and 4.
The PDA two-phase design is the structural implementation of this 60-year-old evidence base. Phase 1 (IMPACT) establishes the baseline — the quantified starting point across all diagnostic dimensions. Phase 2 (DELTA) measures the same dimensions after targeted interventions and calculates the statistical difference. The result is what Kirkpatrick Levels 3 and 4 require: a valid, time-separated, comparable measurement of change.
There is a second, less widely understood reason why single-point measurement misleads: regression to the mean. First identified by Francis Galton in 1886 and formalised in statistical theory, regression to the mean describes the tendency for extreme measurements to move toward the average on subsequent measurement, regardless of any intervention.
In practical terms: if a team is in a period of acute stress when measured, their scores on stress indicators will be high. If they are measured again six months later — with or without any intervention — their scores will typically be lower, simply because acute stress states are not sustained indefinitely. A coach who intervenes after a high-stress measurement and reports improvement six months later may be observing natural regression to the mean rather than the effect of their coaching.
The only way to distinguish real intervention effects from natural regression to the mean is to compare the rate of change against a validated norm, or to use a within-subject design that tracks the pattern of change over multiple time points. PDA's two-phase methodology addresses this by establishing a population-adjusted baseline in Phase 1 and measuring DELTA change against that adjusted reference in Phase 2 — separating the signal of genuine improvement from the noise of natural fluctuation.
The 4–6 month interval between Phase 1 and Phase 2 is not arbitrary. It reflects the convergence of several independent lines of research on the timescales of meaningful behavioural and organisational change.
The 4–6 month window optimises for intervention maturation while minimising confounding. It is long enough for real change to manifest and short enough to maintain causal attribution to the programme.
The consequences of the absence of Phase 2 measurement are not theoretical. They are experienced daily by coaches and HR professionals who deliver excellent programmes but cannot defend their value when the renewal conversation comes. Without Phase 2 data, the following questions are unanswerable with evidence:
These are not supplementary questions. They are the questions that CFOs ask, that boards ask, and that procurement departments ask when a programme is up for renewal. Without Phase 2 data, the answer to all of them is: "We believe the programme was effective." With Phase 2 data, the answer is: "Stress indicators fell by 41%. Engagement rose by 28%. Leadership perception scores improved by 0.8 standard deviations. Here is the board-ready evidence."
For organisations and for the coaches and consultants who serve them, the two-phase methodology is ultimately about competitive advantage. Organisations that can measure the ROI of their people investment make better decisions about where to invest next. Coaches and consultants who can demonstrate measurable impact win renewals, referrals, and the ability to command premium fees.
The irony is that most barriers to measurement are not technical — they are structural. The tools required to establish a rigorous pre-post measurement framework at the team level have historically been expensive, complex, and specialist. PDA Platform removes these barriers, making board-ready two-phase measurement accessible to any team, at any scale, delivered by any qualified coach or HR professional.
The science behind why this matters has been established for 60 years. The only thing that was missing was the means to act on it.