Why people give up in the first week — and how not to
Somebody started, stopped a week later and wrote in his head: "I am not up to this". Below are four reasons this happens, all named by the people themselves. And the main point: not one of them is about ability.
Somebody started, stopped a week later and wrote in his head: "I am not up to this". Below are four reasons this happens, all named by the people themselves. And the main point: not one of them is about ability.
If there is no time to read: five steps
- Rewatching a lesson several times is normal, not a sign that you are not up to it.
- Do not try to keep up with everything you are given. It is physically impossible and not expected.
- One result a week matters more than ten lessons completed.
- Do not compare yourself with the group chat. The people posting there are the ones it worked for — the rest are silent.
- Missed two weeks — carry on from the same place. Do not start again and do not try to catch up.
What the first week looks like from inside
Day one — enthusiasm. Opened it, looked, seems clear enough.
Day three — the volume begins. There turns out to be more material than it seemed. Words appear that have to be memorised. One lesson has to be rewatched twice, because it did not stick first time.
Day five — somebody in the group chat posts a result. A handsome one, made in an evening. You have five days behind you and nothing to show.
Day seven — you do not go in there. Then another day. Two weeks later the tab is closed for good and in your head is the phrase: "this is probably not for me".
And here is what matters most. That phrase is a conclusion not about your ability but about how the first encounter was arranged. The people who get further do not differ in intelligence or age. They handled the four things below differently.
Who I am and why I am writing about the first week
I have been working with neural networks since 2019, and all that time I have watched how people start.
The reasons I list below I did not invent. They are participants' own words — what they wrote themselves explaining why they stopped. Not one added to round the number up.
Reason 1. The volume at the entrance
Word for word from a participant: "at first it was rather hard because of the sheer volume, I had to rewatch the lessons several times".
What happens. Somebody arrives for one task — learning to make pictures, say. And receives a map of a whole field: models, prompts, settings, dozens of tools. And honestly tries to take it in, because it seems that nothing will work without the overall picture.
Three days later the volume becomes more than fits, and a feeling appears that you are falling behind the material.
Why that feeling is false. The full picture is needed by somebody intending to work in the field professionally. You need one task, and one percent of all the material is enough for it.
What to do. Choose one thing you want to be able to do. Everything not about it is postponed rather than missed. It is not going anywhere and will wait until needed.
One participant described the moment it stopped weighing on her: somebody was there who helped break the volume into parts and said where to start. Not learn everything — start with one thing.
Reason 2. Rewatching feels shameful
The second half of the same quotation: "I had to rewatch the lessons several times". And that is what people usually put down as a mark against themselves.
What happens in your head. "Everybody watched once and understood. I am watching for the third time and still getting confused. So it is me."
Why that is untrue. A new field is always taken in through repetition. That is how driving is taught, how a language is taught, how any program is learned. Nobody gets behind a wheel after one explanation.
The only difference is that behind a wheel repetition is considered normal, while in online learning people somehow expect to understand first time.
What to do. Treat rewatching as part of the process rather than a failure. A practical technique: do not rewatch the whole lesson. Rewatch the one piece that was not understood — usually two minutes out of twenty.
Reason 3. "I cannot keep up with trying everything"
Word for word from a participant: "I simply cannot keep up with trying everything that is given".
What happens. Material comes out faster than a person can apply it. A mounting debt appears: three lessons not done, then five, then ten. And the bigger the debt, the less desire there is to go back.
Why it is built that way. The course does not assume you will try every exercise. It gives a set from which you take what you need. It is a library rather than a to-do list.
What to do. Turn the unit of measurement round. Not "how many lessons did I complete this week" but "what one result appeared".
One finished result a week is fifty-two results a year. Ten lessons watched with not one application is nothing.
And a practical consequence: if this week you watched one lesson and made one thing from it, the week went well. Even if nine went unwatched.
Reason 4. Comparison with the chat
The fourth reason is the most galling, because it hits motivation directly.
What happens. People post results in the group chat. A week after the start there are already finished pieces, handsome pictures, somebody's first jobs. You look at that with your two failed attempts and draw a conclusion.
Why the comparison is invalid by construction. The people posting in the chat are the ones it worked for. The ones it did not work for stay silent — and there are always more of them. You are comparing your real situation with somebody else's showcase.
Plus you do not see the context: the person may have had experience in an adjacent field, or spent the whole weekend on it, or this may be his fifth attempt and he did not post the first four.
What to do. Compare yourself not with the chat but with yourself a week ago. A week ago you could do none of what you can do now — even if that is "I can open it and write a request".
And separately on being over 45. If it seems to you that everybody in the chat is younger and therefore faster, that is also a distortion. One of our participants wrote about himself: "I am 71, a military pensioner, and not a very advanced computer user". He did not give up.
There will be no fifth reason
Lists like this usually have five points, because a round number looks better.
I am deliberately not naming a fifth: people did not name one, and inventing a reason for the sake of symmetry means passing off a guess as a fact. Four confirmed are stronger than five with one made up.
What to do instead: three rules
The one-tool rule. Until the first result you use one. New ones are not tried at all — not "carefully", not "at weekends". Every "maybe I should try that one" eats an evening that could have given a result.
The reason is simple: first attempts in any tool come out badly, that is a normal curve. Changing tool at that stage cancels what you accumulated and starts the curve again.
One result a week. Not ten lessons but one finished result you can show. Slower by feel, faster in fact.
One door instead of a catalogue. Do not choose what to solve a task with. Describe the task in words and get a ready chain of steps. That is exactly what the Cerebrum section is for: you write what needs doing in ordinary language, the right models are picked automatically, and the price is visible before launch.
For somebody drowning in volume that is the main thing: nothing has to be chosen.
What to do if you have already given up on studying
The most common mistake on returning is trying to catch up.
Somebody comes back a month later, sees a mountain of missed material and decides to get through it all at once. Does not get through it and gives up for good — now with a sense that the attempt was the second and also failed.
How to do it properly. Do not catch up. Take one task you need right now and close it. Not from where you stopped but from what you need today.
What was missed is not a debt. It is material sitting and waiting until it is needed.
Where to start today
One action, fifteen minutes.
Answer one question for yourself: which of the four reasons was yours. Volume, repetition, "cannot keep up", or comparison.
Usually one is recognisable, sometimes two. And that is already half the solution, because what to fix then becomes clear: narrow it down, allow yourself repetition, change the unit of measurement, or stop looking at the chat.
And then, one task. Not a lesson but a task: something you genuinely need to do. Close it and show the result to somebody close to you.
Against the first reason — the volume — there is a simple way round. Do not choose a tool; take a ready assistant for your task: it is already configured and no settings need working out. The list is long, but exactly one has to be opened — the one named after your task.
Registration is free and opens three days of the Basic plan. Three days is enough to get one result — and its absence is usually what ends in the words "this is not for me".
What to take on the platform for this task
- The “Introduction and theory” course — how language models work, in plain words; Basic plan
- The “Basic prompting” course — Tone of Voice, role prompts and cascading prompts; Basic plan
- Prompt engineering: a foundational course for working with all AI models — $9, Basic plan and above
- Prompting Assistant — an assistant that helps phrase a request (from Basic)
What to read next
Where to start: one evening, one task — how a proper entrance is arranged.
The one-tool rule — why sampling services eats time.
How to write a greeting or a toast — the simplest task for a quick first result.