Re-ranked batch-capable tools as separate slugs with independent IG; chose bulk vs batch from search wording, not as synonyms on one page. | Dev Logs

Date: 2026-09-03 14:55 Summary: Re-ranked batch-capable tools as separate slugs with independent IG; chose bulk vs batch from search wording, not as synonyms on one page. Visibility: people

[question] 增加批量处理能力的工具，需要单独立项写增益和单独slug，增加bulk或者batch（评估用户搜索习惯进行选择），按照这个逻辑重新整理清单

[try to solve]

Context

The previous review treated batch as an enhancement on existing single-file URLs (multiple + ZIP on image-compress, etc.). The user overrode that: each batch job must be a new tool — its own brief, its own Information Gain, and a distinct slug that contains either bulk or batch, chosen from how people actually search.

That still has to satisfy split-page rules: a new URL is allowed only when the primary control differs (queue, shared settings, skip-on-fail, ZIP/CSV versus one-file preview). Same job with a synonym slug (bulk-compress plus batch-compress) is not allowed. Same pixel filter sold three times (bulk-grayscale / bulk-flip / bulk-border) is not allowed.

Process

1. Re-read funnel + §3.3 H (scene-task slug; split by job type) and IG quota (tool-i18n-seo).
2. Re-cluster the earlier 50 file tools by job type, not by 1:1 pairing with every single-file sibling.
3. Choose bulk vs batch from public tool titles and how-to language (not Planner volume — none was run):
   • bulk: quantity (“in bulk”, bulk compressor/converter/resize) for product photos, WebP folders, SVG sets, Amazon specs, EXIF strip.
   • batch: run-the-same-process-over-a-set (batch watermark, batch compress PDF, batch checksum) for stamps, office PDFs, release hashes.
4. Draft ≥3 IG items per independent URL that are not true of the single-file sibling (queue miss table, ZIP rename, manifest compare, etc.).
5. Land an authority note under docs/seo/keywords/bulk-batch/, pool rows, tracker log, and refresh the analysis canvas. Did not create work-tasks/ (user did not name a slug to implement).

Root cause / analysis

Single-file compress and bulk product-photo compress are different jobs: one is a quality slider on one image; the other is a shared size cap, a per-file fail table, and a ZIP. That is a legitimate split. What is not legitimate is stuffing both bulk and batch into two URLs for the same queue, or minting a bulk clone of every existing image filter.

Head queries such as bulk image compressor are already occupied by cloud hubs. Scene H1 (Bulk compress product photos) is the title-gap fallback until Planner/SERP exist. Without SERP, nothing in this batch is long_gap for the weekly attack slot.

Solution

P0 independent URLs (build, week_attack=no until SERP):

P1 (still new URLs, later): PDF draft watermark, password-protect, owned unlock, SVG icon set, Amazon cover-fill bulk (other social specs as chips — not six bulk-platform URLs), EXIF strip, WASM hero optimize, rotate scans.

One URL + chips, do not split: grayscale + flip + border; Amazon bulk + IG/YT/LinkedIn/OG sizes.

No new slug: existing N→1 / 1→N tools, interactive crop/overlay, compares, extractors, calculators.

Single-file pages stay single-file so they do not pretend the queue job is already shipped.

Notes / boundaries

• Opening work-tasks/{slug}/ still requires an explicit “create this slug” from the user.
• Next coverage step: Keyword Planner + Bing/Google SERP on the P0 five; then 0b briefs.
• zh H1 should follow local wording (批量…), not force the English tokens bulk/batch into Chinese titles.

[actions]

• Rewrote analysis canvas tool-batch-capability-review.canvas.tsx
• Added docs/seo/keywords/bulk-batch/ authority note + README
• Appended pool rows; updated keyword-to-tool-tracker.md snapshot and decision log
• Linked theme from docs/seo/keywords/README.md