I had no access to your analytics, Search Console or Partner Dashboard, so every number here was measured from outside. Each section opens with a short answer, with the detail below it.
Prepared by Ihor · data collected 16 Oct 2026 · links checked 17 Oct 2026
Each row opens the section with the full reasoning and evidence.
| You asked | Short answer | Section |
|---|---|---|
| Go through our apps | Clone.Store's AI is real. But its default output is a legal risk for the merchant, and nothing public says that Clone.Store installs Blocky. | 1 · Apps → |
| …and our funnels | The measurable leak is at the top: AI assistants and App Store search almost never show you. Below that, nothing is tracked, so conversion can't be judged yet. | 2 · Funnels → |
| …and our positioning | Two names tell two stories, and neither matches what AI assistants say. Use one name and one specific promise: a store in the style of one you like, built around your own products. | 3 · Positioning → |
| A plan to reach the goals, with reasoning | 8 actions in dependency order: measure, fix what's true about you, get indexed, then earn reviews and citations. | 4 · Plan → |
| What can be automated, and what can't or shouldn't | Code for measurement and product guardrails. AI only drafts, and a person approves. People handle anything public, legal or relationship-based. | 5 · Automation → |
| Transition Blocky to an AI Store Builder | No new AI needed. Make the existing AI reachable from Blocky, pass 3 gates, then rename the listing "Clone.Store: AI Store Builder". | 6 · Transition → |
408 = 34 buyer questions without a brand name × ChatGPT and Claude × search on and off × 3 runs.
Assistants repeat "best of" lists, Reddit, Shopify docs and App Store listings, and none of these mention Blocky or Clone.Store. Claude's search provider (Brave) holds just 2 clone.store pages.
Clone.Store's "Connect a store" installs Blocky, but nothing public says so. The only public explanation is a competitor's page (Dropmagic), which calls Clone.Store a "no AI" front for Blocky.
My test build reused another brand's photos (Shrek included), showed "1,842 reviews" for a new product and invented clinical claims. 47% of Claude's and 21% of ChatGPT's answers about Clone.Store add a legal warning.
It has no AI, 4 screenshots, no free plan, no Built for Shopify badge and 4 reviews in 90 days. A search for "blocky" shows a different app first, and ChatGPT reviews that app instead of yours (3 of 3).
The listing has no GA4 and your links carry no UTM tags. Visits from Claude arrive with no referrer, so they show up as "direct" traffic.
| # | Action | When | Effort |
|---|---|---|---|
| 0 | Turn on tracking — GA4 on the listing, Search Console, Bing Webmaster, UTMs, "Where did you hear about us?" | Week 1 | ~3 h |
| 4 | Get indexed — sitemap and robots.txt for blockyapps.com, submit both sites, IndexNow | Week 1 | ~2 h |
| 1 | Decide on one name — the listing itself goes live only after the 3 gates | Decide week 1 | Decision |
| 2–3 | Safe build defaults and your own facts (Pricing, About, Terms, one price) | Weeks 1–4 | S–M |
These are the only blockers to starting week 1.
Unblocks action 1 (one name)
If any answer is no, action 1 needs product work first
Unblocks action 2 (safe output)
Needed to size the impact of every action
Every claim is one of four kinds, and the wording says which.
| Checked on a date | I opened the public page myself on 16 or 17 Oct 2026. Pages can change after that. |
| Measured once | My own run on 16 Oct 2026, e.g. the AI answers or the one test build. I give counts ("8 of 15") so you can see the sample size. |
| Reported by a merchant | A public App Store review, quoted as published with the developer's reply. I haven't reproduced these problems; some may already be fixed. |
| My hypothesis | My reasoning, marked as such, with how to test it. |
How I got the data. I asked ChatGPT (OpenAI API) and Claude 43 questions buyers ask, with web search off and on, 3 times each: 531 answers in total. I also checked App Store rank for 10 searches, compared Blocky's listing and reviews with 15 competitors, crawled both websites, reviewed 25 "best of" lists plus Reddit and YouTube, and ran one live Clone.Store build. Every answer and source is in the evidence/ folder.