JIA JINGQIU / METHODS & TOOLS
GEO, in practice.
Google-account pools, channel-specific publishing, and clean ChatGPT / Gemini recommendation observations.
04 / GEO & DISTRIBUTION
GEO: distribution
and recommendation analysis.
My GEO approach has two connected tracks: Google-account pools and Agents organize channel-specific publishing; repeated ChatGPT and Gemini observations show which products, sources and conditions appear. The next change is tied to a concrete content gap or channel task.
WALK THROUGH A DECISION / DEMO
A monitor arm was not recommended.
What should happen next?
DEMO-ARM, its D27 product and the answers below are teaching examples—not actual ChatGPT logs or store results. Specifications, prices and supply still need real evidence. “27-inch” in a name does not prove compatibility.
Fix the question. Keep the failures.
In the US, which desk arms under $150 suit two 27-inch monitors? Explain compatibility requirements.
- SAMPLE SCOPE
- US / English / question v1 / one consumer-product surface and matched session conditions, with six planned runs. In real observations, record the interface, visible model, personalization and search state per run. Keep ChatGPT, Gemini and API results separate.
- ILLUSTRATIVE ANSWER R01
- “Compare two alternative arms. Check weight, mounting interface and desk conditions, not just screen size. DEMO-ARM D27's fit still needs verification.” This illustrative answer cites target compatibility page v1 but does not recommend D27. The page is a simulated source, not a claimed existing URL.
- WHAT IS OBSERVED
- This answer does not recommend the target product; its visible citation concerns compatibility. Preserve the answer, citation location and page version, then inspect the product evidence.
- WHAT REMAINS UNKNOWN
- This does not establish that the model cannot recognize the brand, that one channel has a higher weight, or that more posts would secure inclusion. Visible explanations are not internal ranking logs.
| Run | Status | Target recommendation / citation | Search confirmed |
|---|---|---|---|
| R01 | Valid; target product mentioned | No / Yes | Yes |
| R02 | Valid; target product mentioned | Yes / No | Yes |
| R03 | Valid; target product mentioned | Yes / No | Not confirmed |
| R04 | Valid; target product not mentioned | No / No | Not confirmed |
| R05 | Technical failure; no valid answer | Outside the valid-answer denominator | Not assessed |
| R06 | Planned; no capture record | Retained as unobserved | Not assessed |
Merge duplicate imports of the same answer, but retain genuinely separate runs even if their text matches. This group has 2/4 recommendations and 1/4 target citations. The confirmed-search subset has two valid answers and one citation: report 1/2 separately. A recommendation does not establish that specifications or prices are correct.
One absence, four different possible causes.
| Check first | Condition | Next action |
|---|---|---|
| Does the product fit this purchase? | Actual weight, interface, desk thickness, clearance, total price or US supply fails the requirement. | Do not force a recommendation. Choose an eligible product or study a question it genuinely suits. Do not add unsupported compatibility claims to the material. |
| Does the material explain a genuine fit? | Specifications can be verified, but page v1 lacks installation limits, a compatibility table or an original demonstration. | Build v2 after organizing evidence. Link each claim to specifications, real-product photographs or instructions. Leave unchecked facts pending before expanding purchase claims. |
| Can the material be accessed and retrieved? | The page errors, requires login, blocks retrieval or is not indexed by the relevant search product. | Have the technical owner inspect HTTP access, readable text, robots, indexing directives and internal links. For Google's AI Search features, separately check indexing and snippet eligibility; do not apply those entry requirements wholesale to Gemini App or ChatGPT. |
| What evidence appears in visible sources? | Matched questions repeatedly cite installation experience, comparisons or specialist reviews, while owned pages only list selling points. | Add verifiable installation and comparison material. Where an independent review is appropriate, approach relevant editors with evidence and accept their judgment. Brand accounts must not author supposed third-party recommendations. |
This example prioritizes “compatibility evidence needs improvement” as a working hypothesis: it matters to the purchase and is within the team's control. Whether recommendations change requires verified facts, publication and matched retesting. A cited page does not reveal the algorithmic cause.
One usable source document. Three assigned tasks.
Prioritize using observed topics and citations, owned-content gaps, appropriate publishing access, and production/maintenance cost. Here the compatibility page comes first, followed by a reusable installation demonstration. A forum reply requires an existing relevant question. A cited publication is not necessarily open to submissions.
| Task / purpose | Owner and destination | Delivery and acceptance | If it fails |
|---|---|---|---|
| T01 / Owned compatibility page v2 Explain whether it can be installed | Website owner; authorized brand CMS. The product-information owner verifies specifications; technical staff help check access. | A D27 compatibility table, installation limitations, model identity and v2 version. Include evidenced weight range, interface, desk thickness and clearance. Text, images and specifications agree; the public URL is readable. | Return missing specifications for verification; send page errors to the technical owner. After public access works, inspect retrieval in the target product. Publication is not indexing. |
| T02 / Brand YouTube Provide original installation evidence | Video owner; an authorized brand channel in the Google-account pool. It represents the brand, not an independent buyer. | Interface close-up → desk conditions → continuous installation → adjustment range; add written steps, limitations and the v2 source link. Verify product consistency, captions and public playback. | Without the product or access, retain a script only. Repair the affected shot when footage conflicts with instructions. Pending moderation is not public publication. |
| T03 / Relevant forum reply Resolve an existing compatibility question | Community owner; an existing account and forum that allow brand participation. No new identities for staged questions and answers. | Reply only after finding an existing dual-monitor installation/fit question and checking the rules. Disclose the affiliation, answer conditions and limitations first, then link evidence only as permitted. Verify the actual thread URL, text and visibility. | Do not post without a relevant question or permission. If removed, inspect the reason instead of switching accounts. Return useful questions to the compatibility page. |
The account pool assigns identity, access and responsibility; it does not manufacture endorsements. Five owned accounts reposting one article can produce five URLs, but represent one verified owned publisher and zero independent endorsements. Keep unknown ownership unresolved.
More recommendations can mean the same recommendation rate.
Verify that v2 and the video are actually public, then repeat the question under the same product-surface and session conditions. Retain comparable unchanged questions as controls. “After” below is a separate arithmetic example, not a second batch already collected by the default offline demo.
| Batch | Planned / valid | Explicit recommendation | Target-source citation |
|---|---|---|---|
| Before / demo | 6 / 4; 66.7% valid completion | 2 / 4 = 50% | 1 / 4 = 25% |
| After / separate example | 6 / 6; 100% valid completion | 3 / 6 = 50% | 2 / 6 = 33.3% |
Recommendations rise from two to three, but the rate stays at 50%. Citations move from 25% to 33.3%, with tiny samples and different valid-completion rates. This does not establish improvement; previously failed or unobserved runs could change the conclusion. Run comparable batches under a predefined plan and record model changes and concurrent campaigns, without continuing until the target appears.
- STILL NOT RECOMMENDED
- Recheck fit and page versions; change the question or product if it does not fit. If the product fits, access works and evidence is complete, test the specific gaps that repeatedly appear. Do not start by adding accounts.
- RECOMMENDED, BUT FACTUALLY WRONG
- Separate wrong model/specification claims from an outdated merchant price or stock offer. Correct product facts for the former; verify SKU, market, seller and time for the latter. Request corrections to inaccurate third-party text. Keep correction records and new observations separate.
- CITED, WITH FEW VISITS
- Check actual referrals, destination URLs and visit events first, then whether the page offers a reason to click—such as complete installation steps or a fit decision. Hand-sampled citation counts are not the denominator for website CTR; do not divide visits by them.
Mechanisms and measurement scope used in this example
Supporting links in Google AI Overviews / AI Mode require indexing and snippet eligibility. This applies to Google's Search features, not every answer product. Google AI Search guidance ↗
ChatGPT shopping treats product selection and merchant offers separately, and price or shipping updates may lag. Check referrals in actual analytics, including utm_source=chatgpt.com where present; do not extrapolate site clicks from sampled answers. Shopping results ↗ Publisher and referral guidance ↗
Bing AI Performance's citations, URLs and queries cover supported surfaces, with some data sampled. They are not total web-wide answer volume or channel weights. Bing measurement scope ↗
01 / ACCOUNT POOL & AGENT PUBLISHING
More accounts need clearer roles.
Register Google identities, platform publishing permissions and browser/mobile environments separately. Assign brand, market, language and content purpose, then give the Agent a specific task.
01Google accounts & permissions
Record role ID, represented brand/team, owner, market, language, available services and status. Google sign-in is not permission to publish everywhere; use existing authorized accounts.
02Browser & mobile
Bind browser tasks to the correct AdsPower environment and native-app tasks to an authorized device. Verify the actual identity and destination; keep credentials and device/proxy details private.
03Queue & submission
Each task has a topic, channel, destination, content version, approved scope and owner. Prefer existing APIs/connectors; use permitted browser/device flows when a suitable API is unavailable.
04Readback & exceptions
Avoid concurrent writes to one destination. Reopen and check text, media, links and visibility. Inspect uncertain submissions before retrying; hold permission failures, challenges and rate limits.
Account count is execution capacity, not independent users or a ranking weight. Keep publishing accounts separate from measurement sessions. Do not rotate identities to evade restrictions or manufacture customer reviews.
02 / CHANNEL MAP
Name the channel. Define its job.
Combine observed citations, owned-content gaps, publishing conditions and production cost when assigning tasks. A cited source is not necessarily open to submissions. This is a candidate list, not an algorithmic weight table or a requirement to use every channel.
01 / Owned material
Brand website · Product / comparison / FAQ pages · Blog · Help / developer docs
| Work to do | Completion boundary |
|---|---|
| Specifications, compatibility, prices, limitations, original demonstrations and instructions. Maintain one factual source for channel versions. | Authorized owned publishing; access, indexing and citations need separate checks. |
02 / Google properties
YouTube · Blogger · Google Sites · Google Business Profile
| Work to do | Completion boundary |
|---|---|
| Demonstrations with text, focused pages, brand facts and local business information. | A Google account is account context, not universal publishing permission. Business Profiles require an eligible real business. |
03 / Public articles
LinkedIn Articles / Newsletter · Medium · Substack
| Work to do | Completion boundary |
|---|---|
| Use-case tutorials, comparisons and buying decisions with facts, examples and limitations. | Separate public articles from paywall previews and email; verify identity and editor permissions. |
04 / Communities & Q&A
Reddit · Quora · Hacker News / Show HN · Indie Hackers · Stack Overflow · Stack Exchange · Industry Discourse forums
| Work to do | Completion boundary |
|---|---|
| Find relevant real questions, then provide useful answers, demonstrations or a relevant project introduction. | Check each community separately. Technical Q&A is not an ad inventory; disclose affiliation, without fabricated users or endorsements. |
05 / Developer material
GitHub · Hugging Face · DEV Community · Hashnode
| Work to do | Completion boundary |
|---|---|
| Real code, documentation, model information, technical tutorials and reproducible demonstrations. | Use actual projects and relevant contributions, not empty repositories or unrelated promotional comments. |
06 / Directories & reviews
Product Hunt · BetaList · AlternativeTo · G2 · Capterra · Trustpilot · Clutch
| Work to do | Completion boundary |
|---|---|
| Maintain accurate product profiles, apply to relevant categories, and keep reviews authored by genuine users. | Distinguish submission, paid application, review and publication. Account pools may manage profiles, not write fake customer reviews. |
07 / Social discovery
X · Threads · Instagram · Facebook Pages · TikTok · Pinterest
| Work to do | Completion boundary |
|---|---|
| Short demos, diagrams, updates, discussions and platform-appropriate product content. | Check visibility per post; being published does not prove retrieval or citation by an answer engine. |
08 / News & industry material
Owned newsroom · PR Newswire · Business Wire · Trade media / specialist blogs · Partner / customer sites
| Work to do | Completion boundary |
|---|---|
| Genuine news, cases and primary material; offer a relevant story and evidence to editors. | Paid distribution is not independent reporting. Editors control acceptance; not every outlet accepts contributed articles. |
09 / Product listings
Shopify · Amazon · TikTok Shop · AliExpress · Mercado Libre · Shopee
| Work to do | Completion boundary |
|---|---|
| Keep product names, variants, specifications, prices, availability, markets and media accurate. | Manage authorized stores only. Listing, product recommendation, seller ranking and orders are distinct results. |
10 / Private channels
Discord · Slack · Email
| Work to do | Completion boundary |
|---|---|
| Collect real questions, feedback and cases; publish only authorized, redacted material. | Member visibility is not public web access; do not assume private content is a citation asset. |
03 / RECOMMENDATION ANALYSIS
Trace a recommendation back to a content gap.
Begin with genuine unbranded needs; test branded, comparison and purchase questions separately. Fix market, budget, use case, prompt version and time batches. Do not keep asking until the target appears.
Example: In the US, which desk arms under $150 suit two 27-inch monitors? Explain compatibility requirements.
01Recommendation
Keep the complete answer, products, ordered positions only when ordered, visible reasons and original capture/export. A mention is not necessarily a recommendation.
02Visible sources
Open actual cited pages. Separate citations, related links, verification links and retrieved candidates. Unshown sources stay unknown.
03Buying conditions
Map price, dimensions, compatibility, market, delivery and limitations to evidence. Compare what the cited page provides with what your product material lacks.
04Specific change
Turn the gap into an exact task: a specification table, product-page correction, demonstration, factual update or tailored channel content.
This is evidence inspection from visible results, not recovery of hidden reasoning. A follow-up explanation or requested score is a new answer—not the original ranking log or internal weights.
Sample ChatGPT web/shopping, OpenAI API, Gemini App/API and Google AI Overviews / AI Mode separately. Record visible model, locale, login, Memory, search and session conditions for every run; unknown stays unknown.
04 / CLEAN DATA
Clean observations before comparing them.
Preserve raw answers. Models propose classifications; rules check duplicates and fields; people review ambiguous products, source relationships and ownership.
| Object | Treatment |
|---|---|
| Questions & runs | Group equivalent intent; separate branded and unbranded prompts. Remove duplicate imports of one response, not independent real runs with identical answers. |
| Brands / products / sellers | Separate brand, family, SKU/variant and seller. Merge aliases only with evidence; review ambiguity. |
| URLs | Retain raw, tracking-cleaned and verified canonical URLs. Remove only confirmed tracking parameters; preserve variant, color and model identity. |
| Publishers | Record domain, author, original work, syndication and ownership. Reprints and several owned accounts are not independent endorsements. |
| Link roles | Separate citations, related/verification links and retrieved candidates. Count one URL once for within-answer coverage; store citation positions separately. |
| Failures & missing sources | Keep timeouts, limits and challenges separate from non-recommendations. Retain valid answers with no visible sources in the overall denominator. |
Start with four tables
runs stores conditions and raw answers; answer_items stores products and visible reasons; sources stores links and publisher relationships; actions stores changes, owners, readback and experiment IDs.
runs: run_id, topic_id, surface, locale, session, search_observed, captured_at, raw_answer
answer_items: run_id, brand, product, variant, seller, recommendation, visible_reason
sources: run_id, raw_url, clean_url, canonical_verified, publisher, ownership, link_role
actions: topic_id, gap, destination, owner, version, readback, experiment_id05 / CONTENT CHECKS & RETESTS
Score content gaps, not an imaginary algorithm.
First check hard conditions: budget, market, specifications, factual support and publishing permission. Then check fit, specifications, limitations, current price/availability, comparisons, demonstrations, identity and public access.
Use 0 = confirmed missing, 1 = partial, 2 = sufficient and unknown = unchecked, each with evidence. This is an operator content checklist, not ChatGPT scores or Gemini weights. Prioritize by business relevance, recurring gaps, feasibility and cost.
| Result | Denominator |
|---|---|
| Mention / recommendation | Count answers mentioning and recommending the predefined target product separately, divided by valid answers under matching conditions. Repeated mentions within one answer count once. Track brand mentions separately, outside this product-level example. |
| Visible citation rate | Answers citing the pre-defined target pages/publishers divided by all valid answers in the group. Also report a confirmed-search subset; do not mix denominators. |
| Source coverage | Report unique URLs, domains and independent publishers separately. Related links and candidates are not citations. |
| Visits & business | Use actual visits, qualified leads, orders/subscriptions. Keep branded search, direct traffic and self-reported sources separate. |
Predefine prompts, repeats and time batches. Change a small topic/page group, retain comparable unchanged topics and retest under matched conditions. Report sample size, variation, model updates and parallel campaigns; a single appearance or before/after change is not causal proof.
Technical roles and what currently runs
Account tools handle authorized actions. Models may assist question grouping, writing and extraction; operators verify original text, entities and citations. Visits and business outcomes come from actual analytics. The current code starts with prepared local input, keeping API observations separate from consumer-browser results.
python3 scripts/review.py examples.json --example geo_sample
python3 scripts/review.py examples.json --example distributiongeo_sample checks a matching surface, market, language, question/version, conditions group and target product/source. It merges identical imports of one run and rejects conflicting records. Planned, observed and valid counts remain separate from failures, refusals, non-triggered and unobserved runs. It calculates mentions, recommendations, citations and a confirmed-search subset; valid uncited answers remain in the denominator.
The synthetic before input matches the example above: six planned runs, five observed, four valid; recommendations 2/4, citations 1/4, confirmed-search citations 1/2, and one duplicate import merged. After is a separate page calculation. distribution checks an existing publication ledger. Neither connects to models, collects observations, posts, normalizes URLs/product entities or verifies citations automatically. Callers supply answer/evidence references and boolean judgments. The counts are not model weights or proof of optimization impact.
18 specific channels + a general article template.
All channel names are shown below. Select a name or use arrow keys to view its method. The original dual-monitor examples remain; choose channels for a real project, not a publish-everywhere plan. Developer communities need genuine technical work, and software directories need eligible products. ARTICLE is a general template, outside the 18-channel count. These are operating examples, not publication or citation results.
Let the image make the problem clear first.
- IMAGE
- A full-desk scene, an installation close-up, and horizontal/vertical screen arrangements; match every image to the facts.
- COPY
- Open with “check before you buy,” then list ports, weight, and desk requirements without inventing first-hand experience.
- HANDOFF
- Topic, image files, copy version, brand-relationship disclosure, and target account role.
- CHECK
- The actual images, copy, account identity, and public state; clicking Submit is not completion.
The Agent prepares the carousel and copy from verified facts. The authorized brand-account owner approves publication; reopen the post to check media, identity and public visibility. Track saves, enquiries, referrals and actual citations separately—not engagement as a citation weight.
Write from the work scenario, not a stack of product keywords.
- ANGLE
- Dual-screen writing, remote collaboration, and desk setup—explain why an adjustment is needed.
- PROOF
- Use product information to support compatibility claims; keep the team's identity clear and do not frame brand content as an independent review.
- FORMAT
- Question → checklist → detail image → further reading.
- OUTPUT
- Review platform copy and images separately, and retain a public URL that can be checked.
Distinguish a brand Page from a real member identity and verify the available article or post permissions. The Agent prepares evidence and platform copy; after approval, check the byline, published text and public URL. Track professional enquiries, referrals and actual citations; brand statements are not independent support.
One short question, one point.
- ANGLE
- “A compatible screen size does not guarantee a compatible installation.” Follow with clear compatibility checks.
- FORMAT
- Short sentences, one detail image, and one useful conclusion—not a compressed long-form post.
- ACCOUNT
- Use a genuinely authorized brand or team identity; do not amplify the same content through multiple accounts.
- OUTPUT
- A distinct platform version, source record, and publish-verification record.
The Agent prepares a short post or a relevant reply around one verified fact. Publish from the brand or team account within the approved scope, then read back thread order, media, links and visibility. Observe useful discussion and referrals; impressions and engagement are not model recommendation weights.
Long-form content should solve one problem on its own.
- OUTLINE
- Use case → compatibility conditions → installation constraints → common misconceptions → sources.
- VALUE
- Even without clicking a product link, the reader should be able to make one useful decision.
- CHANNEL FIT
- Start with the community and platform context; do not spread thin ads or lightly reworded templates everywhere.
- OUTPUT
- Complete copy, sources, images, and public URL, with the brand relationship made clear.
Make installation evidence easy to revisit.
- CONTENT
- Continuous installation footage, interface and desk close-ups, and limitations; include readable steps and source links.
- IDENTITY & ACCESS
- An authorized brand channel. Google sign-in establishes identity, not every permission; brand demonstrations are not buyer reviews.
- AGENT WORK & READBACK
- The Agent drafts scripts, chapters and descriptions from verified facts. An owner checks the real product and approves upload. Recheck channel identity, public playback, captions and links; moderation remains pending.
- OBSERVATION
- Track views and visits separately. If matched answer samples cite the video, retain the actual link and position; views are not citations.
One article, one installation or selection question.
- CONTENT
- Write a focused illustrated tutorial with verified specifications, original demonstrations and version references—not a copied brand profile.
- IDENTITY & ACCESS
- A genuine brand or team blog. Check the Google account's author/editor access to that blog and disclose ownership.
- AGENT WORK & READBACK
- The Agent turns evidence into a draft and checks image rights and links before approval. Reopen the published text, images and exact URL; repair only the affected article.
- OBSERVATION
- Check public access and actual citations for the target question. Several posts by one owner are not independent recommendations.
Turn scattered material into a usable reference page.
- CONTENT
- Compatibility tables, installation steps, downloads and update dates. Match the original brand identity rather than creating several supposed review sites.
- IDENTITY & ACCESS
- A brand documentation account with editor and publishing access; verify the intended public audience.
- AGENT WORK & READBACK
- The Agent organizes sections and attachments. An owner checks facts and visibility, then inspects the published version—not the editor preview—including text, downloads and navigation.
- OBSERVATION
- Record access conditions and cited pages. Google hosting does not guarantee retrieval, citation or extra weight.
Maintain an eligible real business, not a placeholder.
- CONTENT
- Accurate address/service area, hours, business description, photographs and contact details. Do not invent locations for an online-only product.
- IDENTITY & ACCESS
- The verified owner of a business meeting current eligibility rules. A Google login does not grant access to that profile.
- AGENT WORK & READBACK
- The Agent compares existing fields with approved business facts. After authorization, update specific fields and recheck moderation and public display; never author customer reviews.
- OBSERVATION
- Measure profile accuracy and available real visits or contacts. Keep local-profile performance separate from general product recommendations.
Publish a comparison that stands on its own.
- CONTENT
- Explain selection, installation mistakes or trade-offs around a user's task, with evidence and exclusions—not many retitled copies.
- IDENTITY & ACCESS
- A genuine author or brand publication. Check author access, editorial approval requirements and any paywall.
- AGENT WORK & READBACK
- The Agent drafts structure and copy; an owner verifies facts, attribution and sources. Separate submission from publication and inspect the visible article/preview, media and links.
- OBSERVATION
- Separate reads, real visits and citations. Retain rejection reasons instead of resubmitting under another identity.
A public article and an email are different deliveries.
- CONTENT
- Source-backed topics, updates or cases. Public archives explain the issue; email goes only to eligible subscribers.
- IDENTITY & ACCESS
- A genuine brand newsletter's administrator/editor. Check subscription status, sending scope and paid/public settings.
- AGENT WORK & READBACK
- The Agent prepares article and email versions. Preview and confirm whether the action is save, publish or send. Inspect the archive and sending record separately; a draft is not delivered mail.
- OBSERVATION
- Assess retrieval only for public pages. Do not merge email opens, visits and model citations.
Answer the thread before deciding whether it needs a link.
- CONTENT
- Respond to an existing installation, use or selection question with specific conditions, limitations and checkable evidence—not a generic promotion.
- IDENTITY & ACCESS
- A genuinely affiliated account with disclosed brand ties. Check each subreddit's participation and self-promotion rules.
- AGENT WORK & READBACK
- The Agent finds relevant threads and drafts the reply and rule notes. After approval, reply to the chosen thread. Reopen text, identity and removal/moderation state; inspect uncertain submissions before retrying.
- OBSERVATION
- Check whether the answer helps, remains visible, brings real visits or is actually cited. No vote manipulation or cross-account endorsements.
An answer should work without an advertising click.
- CONTENT
- Explain a specific question with selection conditions and sources. Separate opinions from verified facts.
- IDENTITY & ACCESS
- A genuine author with clear affiliation. Choose an existing relevant question and check current publishing/promotion rules.
- AGENT WORK & READBACK
- The Agent organizes context and evidence, then drafts a self-contained answer. After approval, submit and recheck text, links, visibility and moderation; do not spread identical replies.
- OBSERVATION
- Track retention of the public answer, interaction, visits and citations by question. Exposure is not product endorsement.
Create a repository only for something real and runnable.
- CONTENT
- Actual code, reproducible examples, a README, inputs/outputs and limits. A real compatibility tool can fit; an empty repository for an ordinary product does not.
- IDENTITY & ACCESS
- A repository the person or team is authorized to maintain. Identify sources, licenses and maintainers; no promotion in unrelated issues.
- AGENT WORK & READBACK
- The Agent documents and tests existing code. An owner reviews the release, then verifies access, links, runnable examples and version consistency.
- OBSERVATION
- Track useful feedback, issues, documentation visits and actual citations separately. Stars are not model recommendation weights.
Explain a real implementation, not a product profile.
- CONTENT
- Show inputs, processing, failures and results from actual code. Skip this channel for products without relevant technical work.
- IDENTITY & ACCESS
- A genuine developer or brand engineering member, with affiliation, code ownership and sources disclosed.
- AGENT WORK & READBACK
- The Agent turns a verified implementation into a tutorial and snippets. After checking community rules, publish and inspect code formatting, demonstrations, sources and public text.
- OBSERVATION
- Review solved technical questions, useful discussion and actual citations. Reads do not establish software adoption or product sales.
Maintain technical explanations readers can actually use.
- CONTENT
- Document architecture choices, API usage, troubleshooting and examples for a real version—not a marketing article dressed as engineering.
- IDENTITY & ACCESS
- An authorized engineering or documentation member with clear authorship and brand affiliation.
- AGENT WORK & READBACK
- The Agent checks the current implementation and drafts the tutorial, diagrams and examples. Publish after approval; inspect URL, code, images and version, retaining original/reprint relationships.
- OBSERVATION
- Review gaps preventing task completion, feedback, visits and citations. Several owned blogs remain one publisher.
Check project fit before preparing a launch.
- CONTENT
- A genuinely usable project meeting current listing rules: purpose, audience, real demonstration, pricing and limits. Do not disguise an ordinary product as new software.
- IDENTITY & ACCESS
- The genuine maker or authorized submitter, with affiliation disclosed. Check eligibility and the current launch process first.
- AGENT WORK & READBACK
- The Agent prepares assets, demonstrations, copy and reply drafts. After approval, submit through the available process. Recheck review/live status, links and facts; no manufactured votes.
- OBSERVATION
- Measure qualified visits, first use and feedback. Keep leaderboard position, votes and model citations separate.
Maintain real software facts; users write their own reviews.
- CONTENT
- Prepare functions, team fit, pricing and limits for eligible software/services. The physical D27 arm is not a software listing.
- IDENTITY & ACCESS
- An authorized vendor-profile administrator. Applying for or claiming a profile is separate from authentic customer reviews.
- AGENT WORK & READBACK
- The Agent prepares field changes, category candidates and evidence. An owner approves applications/updates and rechecks moderation and public fields. Do not write or post reviews for customers.
- OBSERVATION
- Track profile accuracy, qualified referrals and verifiable independent citations separately. Paid services or a public profile do not constitute an independent recommendation.
Help buyers judge where the software fits.
- CONTENT
- Real functions, organization fit, deployment, pricing conditions and support scope. Do not claim missing capabilities to enter popular categories.
- IDENTITY & ACCESS
- An authorized software-vendor contact. Do not create software profiles for ordinary hardware simply to add another channel.
- AGENT WORK & READBACK
- The Agent verifies facts and prepares listing/update material for the current application and review process. Recheck category, fields, links and status; customers independently author their reviews.
- OBSERVATION
- Separate listing acceptance, paid referrals, qualified enquiries and actual citations. One directory page is not broad third-party endorsement.
Give editors evidence. Leave the judgment to them.
- CONTENT
- Pitch a compatibility problem, original test or real industry development with methods, images and interview material. Do not disguise brand claims as independent reviews.
- IDENTITY & ACCESS
- An identified brand PR contact approaching editors or specialist authors with a relevant readership and remit.
- AGENT WORK & READBACK
- The Agent matches a pitch to published editorial requirements and drafts the fact pack and email. Contact only after approval; distinguish sent, accepted and published. Recheck byline, disclosures, quotes and the public page.
- OBSERVATION
- Review relevant audience reach, qualified referrals and material actually cited. Editors may reject the pitch or disagree; label paid distribution separately.
Open the task ledger and measurement method
| Field | Why it is retained |
|---|---|
| task_id / topic / source | Trace the task back to the original product and topic evidence |
| account_role / platform / content_version | Know whom the account represents, where it was posted, and which version was used; never expose account credentials publicly |
| state / submitted_at / checked_at | Keep PREPARED, BLOCKED, SUBMITTED_UNVERIFIED, and PUBLIC_VERIFIED separate |
| canonical_url / visible_copy / visible_media | Move to confirmed only after checking the visible copy, images, and exact public URL |
| failure_reason / owner / next_action | Let the next operator continue; unverified work must not be blindly reposted |
Published, crawlable, indexed, cited by AI, referred visits, and purchases are distinct stages. Hold the question, market, language, and review time constant, save the page actually cited, then connect visits and orders. This page currently shows the account execution and verification process; AI citations, referred visits, and orders for this content still need to be tracked separately.