Progress, Motivation and Remembered Experience
Transcript
Made automatically from the video's sound, so it may contain mistakes. Click a line to play from there.
Welcome back to The Explainer. Now, having successfully matched instant feedback, layout stable skeletons, optimistic rollbacks and polite ARIA live announcements to the 0.1, 0.4, 1, and 10 second human time tiers in our previous session, we're ready to shift our perspective.
Today we're going to ground multi-step motivation and memory in empirical human factor science. We are strictly separating real proven perceptual psychology from those folk UX cargo cults we see absolutely everywhere.
So now that we've got momentary latency sorted, we are zooming right out. We're moving from the micro moments of a single button click to the macro level of a user's entire journey. It's absolutely crucial we understand how people actually experience multi-step motivation, and perhaps even more importantly, how they remember an experience after it ends.
We really need to pull apart the experiencing self from the remembering self. To do that, we'll follow a user chronologically through a typical software as a service flow, looking at how to repair broken pathways at every stage, from onboarding and daily engagement right through to offboarding and legal guardrails, finishing up with a really practical timeline audit.
Right, let's jump straight into onboarding and the progress illusion, looking at how we actually welcome people and the behavioural science behind goal motivation. The absolute foundation here is the goal gradient effect. Now, this concept dates all the way back to Clark Hall's 1932 behaviorism studies.
Originally, he found that rats in a maze actually run faster as they get closer to the food box. But, you know, this isn't just about animals. It fundamentally dictates how we should be designing multi-step onboarding flows today.
Simply put, as we humans approach a reward, our effort accelerates. To put this into a human context, there's a brilliant 2006 cafe loyalty study by Kivitz, Erminsky, and Zeng. They empirically prove that the time between buying coffees dramatically decreases as the percentage of effort remaining shrinks.
As customers moved from their very first stamp to their last, the days between purchases literally dropped by 20%. They accelerate towards the goal and then, fascinatingly, experience a massive post-reward reset slump the moment they achieve it.
This ties perfectly into the endowed progress effect. Take a look at this similar car wash study by Nunez and Dreys. Users who received a 10-slot loyalty card with a plausible two-stamp head start hit a 34% completion rate.
But those who got a completely blank eight-slot card, only 19% completed it. And here's the kicker. However, we really need to bust a massive myth right now. Arbitrary fake progress bars do not work.
We've all seen it, right? You open an app, haven't typed a single character, and it cheerfully declares your profile is 40% complete. That's a broken pathway.
It breeds instant cynicism, because users need a genuine plausible reason for any head start. To repair this, we should use honest, screen reader-friendly, semantic HTML task lists, very much like the gov.uk design system does.
Clear, honest textual tags, like completed, in progress, or not yet started, build real trust. And please, when users do finish a task, respect their prefers reduced motion settings with simple static celebration badges, instead of blasting them with unthrottled full-screen confetti animations.
Moving on to daily engagement, let's talk about the streak trap. Daily streaks are incredibly common, but they are a serious double-edged sword. To truly understand user engagement, we lean on Desi and Ryan's self-determination theory.
The key takeaway here is that any well-designed streak system has to foster genuine intrinsic motivation. And for humans to feel intrinsically motivated, they absolutely need three things: autonomy over their choices, a feeling of competence in their abilities, and a sense of relatedness to others.
That means we have to clearly separate extrinsic coercion from intrinsic motivation. If you're relying on coercive points, punishing loss aversion tactics, and those awful focus hijacking nagging models, you are triggering the overjustification effect. You're literally going to hollow out a user's intrinsic curiosity and burn them out completely.
A repaired streak system feels like genuine encouragement. It uses humane streak freezes and celebrates real mastery milestones to build that deep sense of competence. And this naturally brings us to Goodhart's Law.
It's a systems heuristic that perfectly explains why streaks often fail. When you make a daily engagement metric the primary target, it instantly ceases to be a good measure. Under pressure for control purposes, users will inevitably just game the system to preserve the streak, completely abandoning any meaningful learning or interaction.
So how do we actually build a repaired, multimodal engagement pathway? Well, to protect neurodivergent learners from anxiety-inducing all or nothing burnout, we build in two-day streak freezes. On mobile, we ensure the thumb zone has primary and cancelled tap parity, meaning at least 44 by 44 CSS pixels for both.
For visual accessibility at 400% zoom, we stack task list elements vertically so nothing clips. For keyboard users, we ensure clean tab scanning straight to pause with streak freeze without burying it in retention detours.
And for screen readers, we use explicit W3COGA compliant literal copy, abandoning those completely unexplained, silent visual progress bars. Right, let's navigate to offboarding and the peak end rule. Cancellation is a hypercritical moment because it defines exactly how a user will remember your entire product.
This is where we encounter the peak end rule. Karneman, Fredrickson, Schreiber and Redlmeyer made a massive discovery. Human memory practically ignores the actual duration of an experience. When we look back on an event, our remembering self doesn't average out the whole thing.
Instead, it takes a quick emotional snapshot of two things. The most intense moment, which is the peak, and the final finish, the end. Their 1993 cold presser trial proved this in a completely counterintuitive way. Subjects put a hand in painfully cold 14 degrees Celsius water for 60 seconds.
Then in a second trial, they did the exact same thing for 60 seconds, but kept their hand in for an extra 30 seconds while the water warmed up just slightly to 15 degrees. Against all logic of duration, subjects actually preferred to repeat the longer 90-second trial.
Because it ended on a slightly warmer, less painful note, their remembering self genuinely preferred the longer exposure to discomfort. Which leads us into dark patterns and guardrails. Because we really have to ask: what happens when that final peak is a severely negative one?
We need to clear up a major misunderstanding here. Duration neglect applies to bounded retrospective episodes, not daily operational tools. In software, if you have slow steps, you get real-time drop-off.
Furthermore, negativity dominance means that one terrible final peak, like a surprise renewal fee or a lost draft error, will absolutely ruin the memory of your product, completely overriding any cute end screen animation. Alarmingly, AI prompt-generated growth templates routinely ship these traps, perfectly matching Harry Brignall's deceptive design taxonomy.
You know the sort. Confirm shaming, where the opt-out button says, no thanks, I want my career to stagnate. Or the classic roach motel.
One click to subscribe, but a six-screen maze with 10-pixel low-contrast text links just to cancel. Deploying these deceptive traps is legally perilous. They directly violate both the US FTC's click-to-cancel mandates and the EU Digital Services Act Article 25.
Worse, they disproportionately harm vulnerable W3C Koga users who are dealing with executive function fatigue. Regulatory bodies are cracking down on this hard. The only repaired compliance alternative is symmetric one-click cancellation, proactive seven-day renewal reminders, and a dignified offboarding receipt that keeps their exported certificates intact.
Now, for our final section, the valence timeline audit. Let's make this practical and ensure we're building honest, legally compliant experiences. I highly encourage you to pause the explainer right here.
Grab this reference sheet and actively plot out your own product's journey. Map the experiencing self on a minus five to plus five valence timeline from the very start to the absolute finish. Find out where your peaks are and spot those frustrating dips.
Once you've done that, plot the final snapshot for the remembering self. And finally, run a rigorous DSA Article 25 compliance check to guarantee you are genuinely offering symmetric one-click cancellations and eliminating all deceptive design.
So now that we've successfully repaired the progress journey, stripping out those persistent folk UX myths and grounding our multimodal designs in genuine perceptual psychology and honest engineering, how will you apply these principles to balance complex flexibility and strict trade offs in your very next interface?
Keep that provocative question firmly in mind because it is exactly what we'll be tackling in our season 11 finale, episode 7, applying the principles together. I'll see you then.