Reputation and media inventory (sample case)

1. Purpose and scope

• Subject: [CLIENT] (person/organisation) and known spelling variants
• Period: [START DATE] – [END DATE]
• Source types: news sites, X, Facebook, Instagram, YouTube, TikTok, forums/blogs
• Criterion: define in writing from the start which content counts as "negative" (e.g. an accusation of a crime, a statement presenting a final judgment, a reputation-damaging characterisation)

2. Use three separate numbers

Instead of one total, report three measures separately; this keeps the numbers from being confused and makes them resistant to objection.

• Total publications detected: the number of all unique URLs/posts in the inventory
• Negative content: the number of publications classified per the definition (favourable or neutral news is kept separate)
• Republished content: copies of the same story/agency text on other outlets (not presented as an independent claim, but reported as a spread effect)

3. Inventory table by year

Keep a separate table for each year. Recommended columns:

• Date and time
• Source / account
• Full URL (or post link)
• Subject and the accusatory statement in the content (short quote)
• Negativity class: negative / neutral / favourable
• Source of the claim: who said it, which statement or document it is based on
• Republished? (if yes, the row number of the first publication)
• Archive link and screenshot file name
• Engagement (views, likes, reshares) and measurement date

4. News search

• Search terms: [FULL NAME], spelling variants, title and event keywords
• Clearly eliminate results about other people with the same or similar names and note the reason for exclusion
• Mark agency stories (AA, İHA, ANKA, etc.) as a single source; link copies to the republished column
• Search year by year; record the date, query and number of results of each search

5. X (Twitter) search

The platform does not give a total result count and results may be personalised; therefore the number consists of the unique posts recorded.

• Advanced search: all of these words [FULL NAME]; any of these words [KEYWORD1] OR [KEYWORD2]; none of these words [LOOKALIKE NAMES]; date range period by period
• Split the search by years or even months; record the query and its date for each part
• For each post, enter the link, date, account name, text and engagement count in the table
• Take a screenshot and an archive record of the post; assume it can be deleted
• If an exact count is needed, use a professional social media monitoring service or the platform's official data access

6. Evidence

• For each record: URL, publication date, access date and time
• Screenshot and archive (e.g. web archive) record; keep the hash value of the files
• For important content, electronic evidence capture or notarised capture
• IP/log information can only be requested from the platform and service provider by order of a prosecutor or court

7. Note on the legal framework

• Not every negative news story is unlawful; the right to report, the public interest and freedom of expression are assessed separately
• The outcome of proceedings (e.g. acquittal) alone does not show that every statement was untrue; the reasoned decision and its finality should be added to the file
• Claims such as false statements, defamation or violation of personality rights are examined individually; this must be done by a qualified lawyer

8. Safe wording in a petition

"In the search covering the period [START] – [END], [N] unique news URLs and [M] social media posts were identified, of which [K] were classified as negative content. The dates, links and screenshots of the identified content are presented in the evidence list. Due to platform-related limitations, the exact total may be higher than these figures."