Why can I trust this? And how accurate is it really?
These are the two questions we think you should ask before handing fifteen minutes of honest reflection to any exercise on the internet. So rather than dress the method up in jargon, here is a plain-English tour of what happens under the bonnet — the psychology it draws on, the way the questions are structured, and the mathematics that turns your answers into a ranked, considered result.
One method on its own is easy to fool. Three methods, stacked, are not.
Most values quizzes ask you to rate a list of noble-sounding words from one to ten. Almost everyone gives almost everything a nine. That is not dishonesty — it is a well-documented pattern called social desirability bias. We all want to believe we care about integrity, family, kindness and courage in equal measure.
The Splendid Life exercise sidesteps that problem by refusing to let you rate values in isolation. Instead, it puts three different psychological methods in series, so that each stage catches the blind spots of the one before it. By the time you reach the report, your answers have been filtered, forced against each other, and reality-checked against how you actually spend your fortnight.
The Intuitive Filter — thinning sixty into twenty-four.
You begin with sixty carefully curated human values, presented one at a time in a randomised order. Randomisation matters: it prevents you from anchoring on whichever item happened to appear first, and it stops alphabetical or thematic clustering from skewing your intuitions.
This stage is drawn from a well-established clinical technique called the ACT Card Sort, used by therapists to help clients identify what matters without exhausting them. You sort each item into one of three tiers — Core to Me, Important, or Secondary — and the Core tier is strictly capped at twenty-four. Try to add a twenty-fifth and the interface politely refuses, forcing you to swap one out.
A cap creates scarcity, and scarcity forces meaning. If everything can be core, nothing is. The number twenty-four is deliberate: it is large enough to be generous — you will not lose a value you genuinely hold dear — but small enough that the next stage can compare every one of them against every other in a reasonable amount of time.
The Forced Trade-Off — twenty head-to-head decisions.
The twenty-four survivors of stage one are now fed into a pairwise comparison matrix. Instead of asking "how much do you value honesty?", the engine asks "if life forced a choice tomorrow, would you protect honesty or loyalty?". A binary choice. No middle ground. No polite ties.
You are given twenty of these micro-dilemmas, drawn as unique pairs from the twenty-four. Twenty is chosen carefully: enough matches for every value to appear multiple times and for a stable ranking to emerge, but few enough that decision fatigue does not corrode the honesty of your answers.
For every value, the engine keeps two running counts — how many times it has appeared, and how many times it has won. From these it calculates a win ratio:
win ratio = wins ÷ appearances
A value that wins every time it appears has a ratio of 1.0. One that never wins has a ratio of 0. The twelve values with the highest ratios are promoted to your top twelve — the shortlist that goes forward into stage three. Ties are broken gracefully by appearance count, so a value that survived more contests is preferred over one that survived fewer.
This is the same family of mathematics used in Elo-style ranking systems and in multi-criteria decision analysis. It works because it never asks you to be consistent across the whole list — it only asks you to be honest about this pair, right now.
The Behavioural Alignment Audit — ideal self, meet lived self.
The final stage borrows from a strand of humanistic psychology sometimes called the Ideal-Self / Real-Self gap. For each of your top twelve values you answer two sliders, both scored one (low) to ten (high):
- Perceived Importance — how vital this ideal feels to your sense of personal wholeness.
- Recent Real-World Alignment — how actively your last fortnight of actions, calendar entries, and boundaries actually reflected it.
Where the two numbers disagree — a value scored nine or ten for importance but four or less for alignment — the report flags a Fulfilment Gap. Gaps are not judgements. They are the quiet, honest places where friction, resentment, or a low-grade sense of "something is off" tends to accumulate. Naming them is often the first step to closing them.
Why three stages beat any single question set.
Stage one uses intuition — the fast, low-effort part of you that already knows which words feel like home. Stage two uses deliberation — the slower part that has to weigh, compare, and commit. Stage three uses observation — the honest part that looks at your diary rather than your daydreams.
Each stage is designed to catch what the previous one might have flattered. Intuition alone over-selects. Deliberation alone exhausts. Observation alone tells you what you do without telling you what it means. Layered, they triangulate.
The engine is only ever as accurate as the answers you feed it.
No online exercise can climb inside your head. What this one can do is structure the questions in a way that makes it harder to lie to yourself and easier to notice when you are. The randomised order, the strict cap, the forced pairings, and the fortnightly reality check are all quietly doing that work in the background.
Take your time. Answer as the person you actually are on a Tuesday morning, not as the one you would like to be quoted as at a dinner party. If you do that, the report you receive will feel unusually well-fitted — sometimes uncomfortably so. That discomfort, when it appears, is a good sign. It usually means the algorithm has found something true.