Every comms analytics problem eventually traces back to tagging. You cannot report on campaign performance if different team members tag the same campaign three different ways. You cannot track share of voice by topic if your topic tags overlap or are inconsistently applied. You cannot compare sentiment across quarters if half the team uses a 3-point scale and the other half uses 5. This guide establishes a tagging standard that makes your data reliable.
Why Tagging Breaks
Tagging fails in UK comms teams for three predictable reasons:
1. No documented taxonomy. Tags are invented ad hoc. One person tags coverage as "financial results," another tags the same coverage as "Q3 results," a third uses "earnings." All three mean the same thing. Your dashboard shows three separate topics instead of one. 2. No governance. Anyone can create new tags at any time. After 12 months, your monitoring platform has 200+ tags, half of which are used once and then abandoned. 3. No training. New team members are told to "tag the coverage" without being shown the rules. They make their best guess, which is often wrong.
Fixing these three problems solves 90% of tagging issues.
The Standard Tag Set
Define a controlled vocabulary for each tag dimension. Here is a template that works for most UK corporate and agency comms teams.
1. Outlet Tier (Required for Every Clip)
| Tier | Definition | Examples | |---|---|---| | Tier 1 | UK national newspapers, broadcast, and top-tier trade | FT, Guardian, Times, Telegraph, BBC News, Sky News, Independent, PR Week, City AM, Citywire | | Tier 2 | Regional dailies, specialist publications, sector trade | Manchester Evening News, Yorkshire Post, Insurance Journal, The Drum, TechRadar, Campaign | | Tier 3 | Online-only, blogs, aggregators, niche sites | Industry association sites, niche blogs, content syndication sites, local business directories |
Rule: Assign every outlet to a tier and publish the list. When a new outlet appears, the taxonomy owner assigns a tier before it can be tagged. Meltwater, Cision, and Signal AI all allow you to create custom outlet groupings that map to these tiers.
2. Topic Tags (Required — Maximum 20)
Define topic tags that map directly to your business priorities. Not your organisational structure, not your product categories — your priorities.
Example for a UK financial services company:
| Topic Tag | Covers | |---|---| | consumer-duty | FCA Consumer Duty regulation, implementation, compliance | | pricing-transparency | Product pricing, fee disclosure, comparison data | | digital-transformation | Technology investment, platform migration, app launches | | sustainability-esg | ESG reporting, climate commitments, green products | | leadership-profile | CEO/exec mentions, appointments, thought leadership | | financial-results | Half-year and full-year results, trading updates | | m-and-a | Acquisitions, disposals, merger rumours | | product-launch | New products, product updates, feature announcements | | customer-experience | Service quality, complaints, NPS, customer stories | | talent-recruitment | Hiring, employer brand, workplace culture, pay/benefits | | regulatory-action | Enforcement, fines, censure, regulatory investigations | | competitive-landscape | Competitor coverage, market positioning, industry analysis | | data-privacy | GDPR, ICO actions, data breaches, privacy policy | | crisis-incident | Operational failures, security breaches, public incidents | | thought-leadership | Industry opinion, commentary, bylines, conference appearances |
Rule: Every piece of coverage gets exactly one primary topic tag. If coverage spans two topics (e.g., a story about your sustainability product launch), tag the primary topic and add a secondary tag. Never assign more than two topic tags per clip — it creates noise in the data.
3. Campaign Tags (Required for Proactive Coverage)
One tag per active campaign. Use a consistent naming convention:
`YYYY-MM-campaign-name`
Examples:
- `2026-01-salary-data-report`
- `2026-03-consumer-duty-survey`
- `2026-06-q2-results`
Rule: Archive campaign tags after the reporting period ends. Do not delete them — the historical data matters. But remove them from the active tag dropdown so they do not clutter the list.
4. Sentiment (Required for Every Clip)
Use a 3-point scale:
| Tag | Definition | |---|---| | positive | Coverage that would strengthen reputation if read by a stakeholder | | neutral | Factual or balanced coverage with no clear positive or negative lean | | negative | Coverage that would concern a stakeholder if read |
Do not use:
- 5-point scales (very positive / positive / neutral / negative / very negative). The distinction between "positive" and "very positive" is subjective and introduces inconsistency.
- Automated sentiment from Meltwater, Cision, or Signal AI as the final tag. Automated accuracy is 60-75%. Use it for initial sorting, then have a human review and correct.
5. Spokesperson Tags (Optional but Recommended)
Tag every piece of coverage that includes a quote or attribution from a named spokesperson. Use their full name in a consistent format:
`firstname-lastname`
Examples: `jane-smith`, `david-chen`, `sarah-wilson`
This allows you to report on spokesperson visibility: who is being quoted most, in which outlets, and on which topics.
6. Geography Tags (Optional — Useful for Multi-Region Operations)
If your organisation operates across UK regions:
| Tag | Covers | |---|---| | uk-national | UK-wide coverage, no regional focus | | london | London and South East | | north-west | Manchester, Liverpool, Lancashire | | north-east | Newcastle, Sunderland, Durham | | yorkshire | Leeds, Sheffield, York, Hull | | midlands | Birmingham, Nottingham, Leicester | | scotland | All Scottish coverage | | wales | All Welsh coverage |
Rule: Only use geography tags if you actually report by region. Adding tags "just in case" creates maintenance burden with no benefit.
Governance
Taxonomy Owner
Assign one person as the taxonomy owner. This is typically the analytics lead or the most senior person on the monitoring team. The taxonomy owner:
- Approves all new tags
- Reviews and merges duplicates monthly
- Trains new team members on tagging rules
- Publishes the current taxonomy document
Change Control
When someone wants a new tag:
1. They request it from the taxonomy owner with a justification: "I need a tag for [X] because [we cannot report on Y without it]" 2. The taxonomy owner checks whether an existing tag covers the need 3. If a new tag is genuinely needed, the owner adds it to the taxonomy document and communicates the change to the team 4. If the need is temporary (a one-off campaign), the tag is added with an expiry date
Rule: Never create a new tag without going through this process. Undisciplined tag creation is the single biggest cause of analytics failure.
Common Mistake: The Tag Explosion
A UK agency managing monitoring for a FTSE 100 client allowed all six team members to create tags freely in Meltwater. After 18 months, the account had 340 tags. A significant number were duplicates ("ESG", "esg", "ESG-reporting", "sustainability/ESG"), several were used once and abandoned, and 15% were misspelled. The quarterly report was unreliable because coverage was scattered across overlapping tags. It took two weeks of full-time work to clean and consolidate the taxonomy. The fix was simple: lock tag creation to one person and publish a controlled vocabulary.
Monthly Audit Process
On the first working day of each month, the taxonomy owner runs a 30-minute audit:
1. Export the tag list from your monitoring platform 2. Check for new tags created without approval (lock these or merge them) 3. Review tag usage counts — tags used fewer than 3 times in the past quarter may be candidates for removal 4. Spot-check 10 clips for tagging accuracy. Pull 10 random clips from the past month and verify: correct tier, correct topic, correct sentiment, correct campaign (if applicable) 5. Update the taxonomy document with any changes
If the spot-check reveals an error rate above 15%, schedule a retraining session. If it is below 5%, your tagging discipline is good.
Platform Configuration
Meltwater
- Use "Tags" for topic and campaign tagging
- Use "Custom Scores" for sentiment (override automated sentiment)
- Use "Media Lists" for outlet tier grouping
- Export via API or scheduled CSV for dashboard integration
Cision
- Use "Tags" for all custom classifications
- Create "Outlet Lists" for tier grouping
- Use "Prominence" and "Sentiment" fields for quality scoring
- Export via CisionOne Analytics or API
Signal AI
- Use "Labels" for topic tagging
- Create "Narratives" for theme tracking
- Use custom "Portfolios" for outlet tier grouping
- Signal AI's AI-assisted tagging is more accurate than most platforms but still requires human spot-checking
Whichever platform you use, document the mapping between your taxonomy and the platform's tagging fields. When you switch platforms (and you will eventually), this mapping document saves weeks of migration work.