Column reference
Fourteen columns, one of them required.
An export writes all fourteen columns in the order below. On import the order does not matter, and unrecognised columns are ignored.
Import reference
The CSV columns LinkJuice accepts.
Only url is required. The other thirteen columns can be left blank or left out entirely, and the same import also accepts JSON. Every rule below describes what the product actually does today.
- url
- The only required column. It must be a complete http:// or https:// address. Also accepted as canonical_url, original_url, link, backlink, 外链, 链接, or 网址.
- type
- Free text describing the kind of target. A blank cell becomes directory. Also accepted as 类型.
- priority
- Accepts p1, 1, high, 高; p2, 2, medium, normal, 中; p3, 3, low, 低. Blank leaves it unset. Also accepted as 优先级.
- status
- Accepts submitted (or submitted_unverified), live (or live_confirmed), and skipped. Blank leaves it unset. Also accepted as 状态.
- keywords
- Free text for keywords or tags. Also accepted as tags, 关键词, or 标签.
- notes
- Free text. Also accepted as note or 备注.
- group
- Your own grouping label, as free text. Also accepted as 分组.
- domain_rating
- A number from 0 to 100, or blank for unknown. A value outside that range, or one that is not a number, skips the whole row. Also accepted as dr or 域名评级.
- authority_score
- A number from 0 to 100, following the same rule as domain_rating. Also accepted as as or 权威分数.
- important
- Accepts 1, true, yes, y, 是 for true and 0, false, no, n, 否 for false. Blank counts as false. Also accepted as 重要.
- sort_order
- Any finite number. Blank becomes 0. Also accepted as order or 排序.
- language
- A language label as free text. Blank uses your site default, otherwise en. Also accepted as lang or 语言.
- fee
- Free text describing the cost. Blank becomes unknown. Also accepted as 费用.
- next_follow_up
- The date you plan to follow up. Also accepted as follow_up or 下次跟进.
How header names are matched
Each header is trimmed, split at camelCase boundaries, lowercased, and then any run of spaces or hyphens becomes a single underscore. Domain Rating, domainRating, domain-rating, and DOMAIN_RATING therefore all resolve to the same column. Columns that match nothing are ignored rather than failing the import.
Blank cells are safe
Apart from url, an empty cell is never an error. Numeric columns are treated as unset, type, language, and fee fall back to their defaults, and important counts as false. You do not need to fill in a complete grid to import.
Why a row gets skipped
There are four reasons: the address cannot be parsed as an http or https URL; a value is not recognised, such as a rating outside 0 to 100 or an important cell that is neither true nor false; the domain is on your blacklist; the same target already appears earlier in the file; or your plan's limit on new targets is reached, in which case targets that already exist are still updated. The import result reports each count separately.
JSON works too
The same import accepts either format and decides by looking at the text: if it starts with [ or { it is read as JSON, otherwise as CSV. JSON must be an array of objects, and the field names use the same aliases listed above.
Re-importing updates instead of duplicating
Targets are matched by site plus canonical URL. Export your list, edit it in a spreadsheet, and import it again to update the existing targets in place. The export is written with a BOM and CRLF line endings so it opens cleanly in Excel.
After the import
Once the list is in, work through the targets.
Notes, priority, status, and follow-up dates stay with each target.