Adding to a routine without breaking it
The routine has been running for two months and it’s stopped being an effort. Naturally you want to add to it — there are things you left out to keep it short, and now there’s room.
This is the most dangerous moment in a routine’s life after the first month, and it’s dangerous precisely because everything is going well. Almost every routine that dies after establishing itself dies of growth, and the growth always happened during a period when adding felt obviously safe.
Why success is the risk factor
Adding is most attractive when the routine is cheapest, and it changes exactly that. The reason you feel able to add is that the current version costs you almost nothing to execute. That cheapness came from familiarity, and familiarity is specific to the version you’ve been doing. A new component is unfamiliar, so it costs attention, so the whole session costs more than it did — including on the bad days you’d stopped worrying about.
The floor moves without anyone deciding to move it. This is the mechanism that does most of the damage. You add something, and within a fortnight the routine’s definition of “done” has quietly expanded to include it, so the smallest acceptable version is now bigger than it was. Nobody made that decision. But now the bad-day version doesn’t fit into a bad day, and all-or-nothing is back.
Additions arrive in clusters. Nobody adds one thing. Once the routine is under revision, three or four things go in, because they were all on the list of things you’d left out and the list was written in one sitting.
And the reason for adding is often boredom rather than a genuine need for more. Boredom is a real problem and expanding the routine is one of the worse solutions to it, because the novelty wears off and the length stays.
The timing rule that matters
The intuitive time to add is when it’s going well. The better time is when the routine has become genuinely uninteresting and has stayed that way for a while.
The reason: a routine you’re still slightly enthusiastic about is a routine still running partly on enthusiasm, which means you can’t yet tell how much structure it has. Novelty decays before results arrive, and a routine that hasn’t been through a flat stretch and out the other side is not established, it’s early. Adding to it loads weight onto something that hasn’t set.
The signal to look for isn’t a duration — it’s a property. Has the routine survived a bad week without you thinking about it? Did it happen while you were annoyed, rushed, and unenthused? If it’s never been tested, don’t add; wait for a test to occur naturally, and one will.
THE TEST — is it safe to add
· Has the routine survived at least one week you'd call
bad, without conscious effort? → if no, don't add
· Is your floor written down, and is it the same as it
was a month ago? → if no, fix that first
· Are you adding because something is missing, or because
you're bored? → bored → don't add,
change something
instead
· How many things do you want to add? → more than one
→ add one
All clear → one addition, then four weeks of nothing.How to add
One thing. Not one session’s worth of things. One.
Then wait, and don’t count the waiting as slack. Four weeks of the new version before anything else changes, and the point of the wait is to see the new version meet a bad day. If it survives one, you know something. If nothing difficult has happened yet, you know nothing and the clock hasn’t really started.
Keep the floor where it was. Explicitly, in writing: the floor is the old version, unchanged. The addition is part of the normal version and not part of the minimum. This is the single most effective protection available and it costs nothing but a note. If the floor is allowed to drift up with each addition, you’re not growing the routine, you’re raising the bar for what counts as a day.
Don’t touch the cue. Additions frequently come with a plan to move the routine to a better time, because the longer version needs more room. Now you’ve changed two variables and if it falls over you won’t know which one did it — and the cue was the part that was working.
Add at the end, not the beginning. A new component at the front sits between you and the routine you already do reliably, so every session now begins with the unfamiliar part. At the end, a bad day can simply stop before it.
Growing by density instead of length
Length is not the only axis, and it’s the one with the worst adherence properties because it’s the one that competes with a bad day. Two alternatives worth knowing:
More often, same size. If you’re at five days and the sessions are short, going to seven adds volume without making any single session harder to fit. Requires solving the weekend if that’s where the gaps are.
Harder, same duration. Progressing what you already do rather than appending to it. Zero adherence cost, because the session’s footprint in your day is unchanged. What “harder” means for your particular routine is a question for whoever designed it — and if that was a clinician, it’s their call rather than yours.
When it does fall over
If you add something and within a few weeks you’re missing days you weren’t missing before, the diagnosis is simple and the fix is unpopular: remove the addition entirely and go back. Not reduce it, not do it every other day — remove it, and confirm the old version re-stabilises.
People resist this because it reads as going backwards. It isn’t; it’s the only way to establish that the addition caused the problem, and it protects the version you know works. A routine you have to rebuild from scratch costs vastly more than the four weeks you lose by reverting.
What this costs
Adding one thing every four or five weeks is slow. Over a year that’s not many additions, and if you have a specific target you’re working toward it may be too slow to meet it. That’s a genuine trade and if you’ve got a deadline it deserves consideration — though a deadline is its own risk to adherence.
Keeping the floor fixed also means the gap between your normal session and your minimum widens over time, and a bad day eventually produces something that feels trivially small relative to what you usually do. That gap is uncomfortable and it’s the price of the routine surviving bad days at all.
What this doesn’t fix
None of this tells you what to add, which is outside what this site addresses.
And if the routine hasn’t yet survived a month, this post isn’t for you yet. The problem at that stage isn’t growth, it’s whether the thing fires at all.