Business judgment on AI products
Habits are more than a checkbox: log numbers and times, and a few days off does not wipe the streak.
01
Start inside the user's day · Public facts + workflow reasoning · 2026-08-28
Habits are more than a checkbox: log numbers and times, and a few days off does not wipe the streak.
Public materials do not yet show how users complete this job today or what they replace.
The product targets friction in this job, but public user evidence does not yet show the cost, frequency, or consequence of leaving it unsolved.
Checkbox apps treat “not quite” as failure, erase real effort, and people quit. Don't ship another streak calendar—sell a ledger that records the actual numbers and does not punish a pause, to the people willing to log the truth.
It promises a simpler way to complete this job: Habits are more than a checkbox: log numbers and times, and a few days off does not wipe the streak. The exact adoption motive and repeat use are not yet verified.
① Whether it actually open-sources as planned — open source + self-hosting is the only; clear growth story it has; ② Whether Android ships — no Android locks out half the market; ③ Whether any paying user shows up publicly — for an indie product, the standard is; survival, not growth
Keep watching. It promises a simpler way to complete this job: Habits are more than a checkbox: log numbers and times, and a few days off does not wipe the streak. The exact adoption motive and repeat use are not yet verified.
the counterintuitive design for tracking products — "skip" is a deliberately kept escape hatch. Real effort is not binary; giving users a "pause, not fail" option sustains long-term use better than punitive streaks. Numeric and time-of-day record types also carry more analytical value than a plain check. Free / $3.99 monthly / $49 lifetime, AI behind the paid tier, data export always free — a replicable structure for consumer tools.
Freemium. Free: 5 habits, yes/no and numeric types, one auto chart per habit, 60 days; of stats, multi-device sync. Pro at $3.99/month: unlimited habits, time-of-day and; select types, AI access (MCP), Apple Health, candle chart and custom … ① Whether it actually open-sources as planned — open source + self-hosting is the only; clear growth story it has; ② Whether Android ships — no Android locks out half the market; ③ Whether any paying user shows up publicly — for an indie product, the standard is; survival, not growth
Habits are more than a checkbox: log numbers and times, and a few days off does not wipe the streak.
It promises a simpler way to complete this job: Habits are more than a checkbox: log numbers and times, and a few days off does not wipe the streak. The exact adoption motive and repeat use are not yet verified.
An English validation note will follow from the public evidence.
The product claims to help users complete: “Habits are more than a checkbox: log numbers and times, and a few days off does not wipe the streak.”. User evidence has not yet verified pain intensity or the cost of doing without it.
The assessment is recorded; an English explanation is pending.
The assessment is recorded; an English explanation is pending.
The assessment is recorded; an English explanation is pending.
02
Market comparison
The Chinese–English market comparison is not complete yet. A conclusion follows only after its coverage and verifiable evidence are recorded.
03
The call and next move come first; the full read retains the evidence and counterevidence.
A habit tracker that refuses to stop at checkmarks — check, number, time-of-day, and pick-from-a-list sit side by side, so you see "the numbers and the time" rather than just "done or not."
Individually built by bohdanstefaniuk (Habit Pocket, habitpocket.io). The author's own pain, as stated: most trackers only know done/not-done, but he wants to log actual effort — walking 9k steps today when the goal is 10k is still progress; three days of camping without running should not destroy a streak, hence a "skip" feature. iOS and web, offline-first, not open-sourced yet, with a plan to open-source and support self-hosting.
Read: the classic personal tool built for the maker's own pain. Four record types and a forgiving streak are the selling points — small and honest, but no team and no growth engine.
Two old behaviors. First, checkmark habit apps (Streaks and the like): binary state only, so a near-miss like "9k steps" gets recorded as failure, real effort is erased, and motivation erodes. Second, keeping metrics in Excel — the author's own words: "I wanted something friendlier than an Excel spreadsheet": records anything but has zero experience.
It replaces "either a fake win or a full failure" binary logging, and the "records-everything-but-painful" generic spreadsheet. The core difference is the data model: "time of day" and "numbers" are first-class citizens instead of everything being squashed into a boolean.
Freemium. Free: 5 habits, yes/no and numeric types, one auto chart per habit, 60 days of stats, multi-device sync. Pro at $3.99/month: unlimited habits, time-of-day and select types, AI access (MCP), Apple Health, candle chart and custom chart builder. Lifetime at $49. No Android yet (planned). Data exports as CSV/JSON at any time; deleting the account wipes everything immediately.
Read: restrained and honest pricing; $49 lifetime is the common indie-developer choice. But no reminder notifications and no Android means the growth engine is weak.
| Dimension | Call |
|---|---|
| Founder-product fit | The author uses it himself; the pain is real, and the motives behind number-logging and skip design all line up |
| Product insight | "Record real effort, not a binary state" has insight, but the category is crowded |
| Execution quality | Offline operations-log and MCP integration — not rough engineering |
| Timing | Habit tracking is a red ocean; four record types alone will not differentiate a market |
The most copyable thing is the "skip, not fail" mechanism. Real-world effort is not binary. Three days camping without running is not abandoning running. Giving "pause" an expression that does not break a streak sustains long-term use better than punitive checking-in — a general design lesson for any retention-oriented product.
"Record numbers and time" rebels against binary check-ins, but it cuts both ways. Numeric logging is truer and more analyzable, but it costs more to enter. A check is a second; a number is three. Three seconds of friction will shed most users after ninety days. It bets on the small-but-real niche of people willing to log real data.
The MCP interface is the smartest differentiation. Letting a user's AI read habit data directly over MCP upgrades the tracker from "logging tool" to "personal data base." The headroom of that interface is much larger than the four record types — once the data layer opens, whether the tool is a habit tracker barely matters.
But 6 points plus a red-ocean category caps its ceiling. No reminders, no Android, no social layer, and the head of the habit-tracking market is already taken by checkmark apps. A differentiated data model only carves out a small market.
① Whether it actually open-sources as planned — open source + self-hosting is the only clear growth story it has ② Whether Android ships — no Android locks out half the market ③ Whether any paying user shows up publicly — for an indie product, the standard is survival, not growth
Product logic: the counterintuitive design for tracking products — "skip" is a deliberately kept escape hatch. Real effort is not binary; giving users a "pause, not fail" option sustains long-term use better than punitive streaks. Numeric and time-of-day record types also carry more analytical value than a plain check.
Pricing structure: Free / $3.99 monthly / $49 lifetime, AI behind the paid tier, data export always free — a replicable structure for consumer tools.
Unproven. An honest little product with a useful mechanism, but a red ocean, no Android, no reminders, and 6 points of traction. Treat it as a reference, not a target.
04
Evidence trail
05
Use these searches when the official site is missing or the current link is only a lead.