Product walkthrough / checked October 24, 2026
How ChangeLoop 7 turns one concern into a seven-day cycle.
The product takes you through a fixed intake, an evidence-linked explanation, three possible methods, one selected action, and a weekly review. This walkthrough explains what you enter, what you receive, and which decisions remain yours. The first complete cycle is free for an eligible account.
What is ChangeLoop 7?
ChangeLoop 7 is self-directed software for testing a small behavior method over a week. It is designed for a recurring obstacle that can be described and observed. The cycle produces a record and a next-step decision; it does not guarantee a lasting habit, diagnose a condition, or promise that the generated interpretation is correct.
1. Prepare one concrete problem for the intake
You answer seven fixed questions. The intake collects your answers without interrupting them with AI coaching or an open conversation. Prepare the desired change, the repeated pattern, the situation in which it appears, previous attempts, an observable sign of progress, and real constraints. Use ordinary language and include only details needed to understand the behavior.
For a fictional example, Maya wants to begin a design brief before opening team messages. A useful starting answer is: "When I sit down at 9 a.m., I open messages and postpone the brief until lunch." Another answer could explain that Maya has ten minutes before a meeting and cannot silence urgent work calls. These invented details show the level of specificity, not the answers the product expects everyone to give.
2. Read the explanation and check its evidence
The analysis returns one plain-language idea about the pattern, sourced cases or research context, and three possible methods. The evidence requirement includes at least three traceable sources. Treat the interpretation as a hypothesis to inspect. Check whether its explanation resembles what you wrote and whether a cited study actually addresses the claim beside it.
A traceable citation is not proof of a personal prediction. A study about planning may support a planning technique without showing that it will fix Maya's mornings. If an output depends on a false assumption, do not treat that assumption as an established fact about yourself. The research method explains the boundary between published findings, generated interpretations, and product decisions.
3. Choose one of three methods
The three methods are alternatives. You select one; the product then assembles its execution framework. Compare what each option requires in your actual situation, including its cue, minimum action, and reduced fallback. Avoid choosing a method solely because it sounds more ambitious. An action you cannot attempt during an ordinary day will provide little information about its usefulness.
In a fictional illustration, Maya could select a five-minute start on the brief. The minimum might be writing three rough points before opening messages. This is an example of a bounded action, not a promised generated result. Maya would judge the displayed proposal against the ten-minute constraint and the need to remain reachable. Choosing one method does not mean agreeing to use it forever.
4. Record today's action during the week
Days one through six expose only the current task; future tasks remain locked. You record Done or Not done for that task. A missed task reveals a fallback that was prepared with the plan. Daily check-ins do not call AI and do not open a chat. There is no requirement to upload a photograph, video, or daily written explanation.
Keep the completion decision tied to the actual minimum. If Maya only opens the document but the minimum requires three points, the original action is Not done. A smaller fallback can still be useful without changing that status into a successful completion. If the action becomes unsuitable, there is no obligation to continue it merely to protect a completion count.
5. On day seven, record status and write one review
The weekly review asks for a paragraph of 40 to 1,600 characters. Characters are not words: a brief, concrete paragraph is enough. You also choose whether you noticed a clear change, a small change, or no helpful change. Completing every previous action is not a prerequisite for reviewing the week. Missed and unrecorded days are relevant information.
A fictional review could say: "I wrote the three points on four mornings. Two early calls interrupted the start, and one day I did not record it. Starting was easier when the brief was already open. Next time I want to test preparation before I finish work." This describes observations and a proposed adjustment without claiming the week established a permanent habit. The review is saved and used for the next-cycle analysis.
6. Inspect the next-cycle preview before continuing
The product keeps the written review and updates the long-term dossier used for continuity. It shows a preview of a possible next step, such as keeping the method, reducing its size, or changing direction. You decide whether to continue. A successful first week does not require increasing the workload, and a difficult week does not require repeating the same method unchanged.
There is no fixed three-week endpoint. Continuing after the introductory cycle requires a subscription. The official pricing page, accessed October 24, 2026, lists USD 24.99 per month or USD 239 per year. Check the displayed terms before subscribing. These prices describe access to the workflow; they are not a promise of a particular personal outcome.
Know what happens to the information you enter
The account-based workflow stores your dossier and uses model providers at analysis checkpoints. It is different from the free printable worksheet, whose field values stay in the browser page. Read the privacy policy for processing and retention details. Do not enter payment credentials, identification numbers, medical records, or someone else's confidential information.
This walkthrough was checked against the current product flow on October 24, 2026. The seven-day cadence is a product choice. For wider context, Harkin et al.'s original progress-monitoring meta-analysis, accessed the same day, supports monitoring as a general ingredient but does not validate ChangeLoop 7. For tool selection rather than operating instructions, read the tracker, journal, and AI chat comparison.