Frequently asked questions
ChangeLoop 7 helps you pursue one desired change through small goals, daily actions, and repeating seven-day reviews. These answers explain what the product does, what it stores, and what one week can actually tell you.
No. Seven days are one action and review cycle in an ongoing process. The product supports small steps and clearer next decisions but does not claim a habit, cure, transformation, or permanent result within one week.
No. AI is used only after fixed submissions: initial analysis, method generation, private review, and weekly correction. Intake, daily completion, history viewing, and language changes do not open an unlimited conversation.
Your day-seven reflection is stored in the private dossier. The product shows a next-cycle preview that may suggest continuing, reducing, or changing the action. You decide what to do next. There is no fixed three-week ending.
No. You choose one of exactly three alternatives. The seven-day cycle tests only the selected method. If none fits, the product returns to the cognition rather than generating options forever.
No. Only today's action is available. Future tasks remain locked so attention stays on the current step instead of negotiating with the entire week.
A missed day does not end the cycle. You can use the predefined reduced action, and the result becomes evidence for the day-seven review rather than a broken streak. Keep the original status honest: a smaller fallback does not mean the original task happened. If you forgot to record a day, leave that uncertainty visible instead of filling it as Done from memory. Completing every earlier day is not required before the day-seven review.
No. Daily records are completion states, and the weekly review is one written reflection. The product does not require media evidence.
A counterintuitive cognition should not rely on model confidence alone. ChangeLoop 7 requires at least three traceable sources or cases, relevance explanations, and scope limits before the cognition is released. A real DOI shows that a paper exists; it does not establish that its finding applies to you. Model-generated text and agreement between models are not independent evidence. The claim register distinguishes scientific mechanisms from product rules and outcomes that remain unestablished.
No. AI organizes evidence and proposes bounded methods. You choose the method, provide the real-world observation, and decide whether to continue. The product shows a next-cycle preview but does not silently change your plan.
No. ChangeLoop 7 is a personal behavior-change tool and does not replace professional medical, mental-health, legal, financial, or emergency support.
Your first complete seven-day cycle is free. The launch offer is USD 24.99 per month or USD 239 per year, billed annually. The annual plan is equivalent to $19.92 per month and saves $60.88 (about 20%) compared with 12 monthly payments ($299.88). See the pricing page for current availability, billing, cancellation, and refund information.
Read the privacy policy, evidence and corrections policy, and terms. Questions can be sent to support@changeloop7.com.
It makes the decision concrete enough to test. An illustrative plan such as “after I close my laptop, I place tomorrow's notebook on my desk” names an observable cue and response. It does not prove that the action will help. Gollwitzer and Sheeran's 2006 meta-analysis examined 94 tests of implementation intentions. That research supports studying specified plans, while effects depend on the task and context; it does not validate every AI-generated method. Source accessed October 24, 2026.
A dated observation gives the weekly reflection something to compare with the original intention. The Harkin and colleagues meta-analysis examined 138 studies of progress monitoring and goal attainment. It provides a reason to investigate monitoring, not a success rate for ChangeLoop 7. For example, three recorded completions, two recorded misses, and two unrecorded days describe different evidence than five failures. Neither record independently proves why the week went that way. Source accessed October 24, 2026.
Preserve the record and say that no useful signal appeared. The next-cycle preview can support continuing, reducing, or switching the action; you decide what to do. Distinguish an action you attempted without benefit from an action you never had a reasonable opportunity to try. A fictional user who prepared a notebook on three evenings but still could not start work has evidence about preparation, not evidence that the larger goal was achieved. The review should identify that distinction without blaming the person or automatically repeating the same plan.
The description should name a concrete technique, such as an action plan, a prompt, self-monitoring, or an environmental change. Michie and colleagues' 2013 taxonomy organizes 93 techniques into 16 groups to support precise reporting. This is a vocabulary for describing interventions, not certification that ChangeLoop 7 works. Plain language still matters: you should be able to explain the selected action without knowing the taxonomy. Source accessed October 24, 2026.
The workflow is text-first. You provide an initial description, select one method, record daily completion, and submit one written review on day seven. The app accepts a review of 40 to 1,600 characters and offers the three result choices I noticed a clear change, I noticed a small change, or I did not notice a helpful change. It does not require wearable data, screenshots, recordings, videos, or continuous observation. Use the free weekly review template to prepare a concise account before entering the product.
Daily completion is one part of a larger cycle. The product also asks what assumption is being tested, offers three bounded method alternatives, and uses the weekly record to prepare the next cycle. A tracker may be sufficient when you already know the action and only need a record. The combined tracker, journal, and AI chat comparison explains the differences so you can choose a workflow that fits your task.