There’s a moment in an entrepreneur’s cycle that gets talked about a lot.
You might not have heard it called this before — but you’ll recognise the feeling.
They call it the valley of disillusionment.
Sometimes the trough of sorrow.
In venture circles, even the valley of death.
Different names. Same experience.
It’s the point where the early excitement wears off and reality starts billing you.
And it comes with a shift that tends to catch people off guard.
People often feel worse at the exact moment they’ve started engaging with reality rather than theory.
It sounds counterintuitive, but it shows up again and again once people move from thinking about something to actually doing it (Bjork).
Early confidence is often high because the task still looks simple from the outside. Then something shifts. The person has to produce, not recognise. They have to continue, not just understand. They have to act under pressure.
And confidence drops.
Not because they’re getting worse, but because they’re now measuring something more demanding.
There’s related work on what’s called the illusion of explanatory depth, where people feel they understand something until they have to explain or use it (Rozenblit & Keil).
That’s when the gap shows up.
Bandura’s work on self-efficacy shows how quickly confidence can drop when effort is interpreted as inability rather than challenge.
Early confidence can also be inflated before people fully understand the complexity of the task (Kruger & Dunning).
In startup language, this is when interest wanes.
Early energy comes from possibility.
Then reality arrives: slower results, hidden constraints, messy execution.
The system stops rewarding enthusiasm and starts demanding capability.
Now bring that back to what you’re doing here.
When you move from:
“I recognise this”
to
“I have to say this, continue with it and hold it in a conversation”
your felt fluency can drop.
The task has changed.
You’re no longer judging familiarity.
You’re testing whether the language is available for use.
And that is a much harder test.
So when you say:
“I don’t remember anything”
“I can’t get the basics right”
“I feel like I’m going backwards”
— your mind racing slightly faster than your mouth can keep up —
there is a very good chance that several things are happening at once.
Your standards have risen, so you notice more than you used to.
The task has become generative rather than recognitional.
And you’re now working under a different kind of pressure.
What matters here is how you interpret this moment.
What feels like backsliding is often the loss of false ease.
You’re no longer asking: does this look familiar?
You’re asking: can I actually use this when I need it?
That shift in question almost always feels worse before it feels better.
Work on “desirable difficulties” suggests that this kind of effort is often a feature of deeper learning rather than a sign of failure (Bjork & Bjork).
And if it still felt easy, you’d be doing something less useful.
And here’s the interesting part that matters for how we’re working in The French Room.
This doesn’t happen once.
With a traditional approach, you might hit this dip at certain stages and then move on.
With 907, you meet it again and again.
Each new word takes you through a similar cycle:
early clarity
contact with complexity
awkward production
then stabilisation
That loop is demanding.
But it also compounds.
Because each time you come through it, you’re building something that holds.
This kind of repeated reconstruction aligns with research on retrieval and spaced learning (Karpicke).
So if you’re doing 907 and feeling that wobble right now — the frustration, the mental effort, the sense that things were somehow easier before —
it adds up.
It matches what we see in other demanding environments.
It matches what behavioural science has been pointing to for years.
And it matches the shift from theory into use.
And this is the part I both do and don’t want for you.
Because it’s uncomfortable.
But it’s also the point where things start to become real.
And where this starts to become something much more interesting.
Bandura, A. (1997). Self-efficacy: The exercise of control
Bjork, R. A., & Bjork, E. L. — Desirable Difficulties in Learning
Karpicke, J. D. — Retrieval Practice and Memory
Rozenblit, L., & Keil, F. — The Illusion of Explanatory Depth (2002)
Kruger, J., & Dunning, D. — Unskilled and Unaware of It (1999)
Gartner — Hype Cycle Model (Trough of Disillusionment)
Startup literature — “Valley of Death” / entrepreneurial emotional cycles
