Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generate UTM-tagged URLs for campaign tracking. Create consistent, organized tracking links for Google Analytics, HubSpot, and other analytics platforms. Supports bulk generation, naming conventions, and campaign documentation. Use when creating tracking links, campaign URLs, or organizing marketing attribution.
.claude/skills/nicepkg-utm-builder/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 71% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 106% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 176% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 194% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 387% | 0% |
Generate consistent, well-organized UTM tracking links for marketing campaigns.
| Parameter | Required | Purpose | Example | |-----------|----------|---------|---------| | utm_source | Yes | Traffic source | google, linkedin, newsletter | | utm_medium | Yes | Marketing medium | cpc, email, social, organic | | utm_campaign | Yes | Campaign name | spring-sale-2025 | | utm_term | No | Paid search keywords | project+management | | utm_content | No | Differentiate variants | cta-button, hero-image |
utm_source: {platform}
utm_medium: {channel-type}
utm_campaign: {yyyy-mm}-{campaign-name}
utm_term: {keyword} (paid search only)
utm_content: {ad-variant}-{placement}Paid:
- google
- linkedin
- meta (facebook + instagram)
- tiktok
- twitter
- bing
Organic Social:
- linkedin-organic
- twitter-organic
- facebook-organic
Email:
- mailchimp
- hubspot
- sendgrid
- newsletter
Referral:
- partner-{name}
- affiliate-{name}
- referralPaid:
- cpc (cost per click)
- cpm (cost per impression)
- display
- video
- retargeting
Organic:
- organic
- social
- referral
Direct:
- email
- sms
- push
Content:
- blog
- pr
- podcast
- webinarFormat: {date}-{name}-{segment}
Examples:
- 2025-03-spring-promo
- 2025-q1-product-launch
- 2025-04-webinar-ai-trends
- 2025-03-ebook-marketing-guide
- evergreen-demo-requestCreate a UTM link for:
URL: https://example.com/pricing
Campaign: Spring promotion on LinkedInOutput:
https://example.com/pricing?utm_source=linkedin&utm_medium=cpc&utm_campaign=2025-03-spring-promoCreate UTM links for a multi-channel campaign:
URL: https://example.com/demo
Campaign: Q2 Product Launch
Channels:
- Google Search Ads
- LinkedIn Sponsored Content
- Email newsletter
- Twitter organic postOutput:
═══════════════════════════════════════════════════════════════
CAMPAIGN: Q2 Product Launch
BASE URL: https://example.com/demo
DATE: 2025-Q2
═══════════════════════════════════════════════════════════════
GOOGLE SEARCH ADS:
──────────────────────────────────────────────────────────────
https://example.com/demo?utm_source=google&utm_medium=cpc&utm_campaign=2025-q2-product-launch&utm_term={keyword}
Use {keyword} placeholder for dynamic keyword insertion.
LINKEDIN SPONSORED:
──────────────────────────────────────────────────────────────
Variant A (Image):
https://example.com/demo?utm_source=linkedin&utm_medium=cpc&utm_campaign=2025-q2-product-launch&utm_content=image-a
Variant B (Video):
https://example.com/demo?utm_source=linkedin&utm_medium=cpc&utm_campaign=2025-q2-product-launch&utm_content=video-b
EMAIL NEWSLETTER:
──────────────────────────────────────────────────────────────
Header CTA:
https://example.com/demo?utm_source=newsletter&utm_medium=email&utm_campaign=2025-q2-product-launch&utm_content=header-cta
Footer CTA:
https://example.com/demo?utm_source=newsletter&utm_medium=email&utm_campaign=2025-q2-product-launch&utm_content=footer-cta
TWITTER ORGANIC:
──────────────────────────────────────────────────────────────
https://example.com/demo?utm_source=twitter-organic&utm_medium=social&utm_campaign=2025-q2-product-launch
═══════════════════════════════════════════════════════════════
TRACKING SPREADSHEET
═══════════════════════════════════════════════════════════════
| Channel | Source | Medium | Campaign | Content | Full URL |
|---------|--------|--------|----------|---------|----------|
| Google | google | cpc | 2025-q2-product-launch | - | [url] |
| LinkedIn | linkedin | cpc | 2025-q2-product-launch | image-a | [url] |
| LinkedIn | linkedin | cpc | 2025-q2-product-launch | video-b | [url] |
| Email | newsletter | email | 2025-q2-product-launch | header-cta | [url] |
| Email | newsletter | email | 2025-q2-product-launch | footer-cta | [url] |
| Twitter | twitter-organic | social | 2025-q2-product-launch | - | [url] |Create UTM links for A/B testing two landing pages:
URL A: https://example.com/landing-v1
URL B: https://example.com/landing-v2
Campaign: Homepage redesign test
Channel: Google AdsOutput:
VARIANT A (Control):
https://example.com/landing-v1?utm_source=google&utm_medium=cpc&utm_campaign=2025-03-homepage-test&utm_content=control-v1
VARIANT B (Treatment):
https://example.com/landing-v2?utm_source=google&utm_medium=cpc&utm_campaign=2025-03-homepage-test&utm_content=treatment-v2
TRACKING NOTE:
Compare conversion rates in Google Analytics:
Acquisition → Traffic Acquisition → Filter by utm_contentUse ValueTrack parameters for dynamic insertion:
- {keyword} - Search keyword
- {matchtype} - Match type (e, p, b)
- {device} - Device (m, t, c)
- {adposition} - Ad position
Example:
?utm_source=google&utm_medium=cpc&utm_campaign=brand&utm_term={keyword}&utm_content={adposition}Use LinkedIn macros:
- {{CAMPAIGN_NAME}}
- {{CREATIVE_NAME}}
- {{CAMPAIGN_GROUP_NAME}}
Note: LinkedIn auto-appends some tracking; combine carefully.Use URL parameters in ad setup:
- {{ad.name}}
- {{adset.name}}
- {{campaign.name}}
Example:
?utm_source=meta&utm_medium=cpc&utm_campaign={{campaign.name}}&utm_content={{ad.name}}HubSpot auto-recognizes UTM parameters.
Ensure consistency with HubSpot campaign naming.
Best practice:
- Create HubSpot campaign first
- Use exact campaign name in utm_campaign
- Tag all assets (emails, pages, ads) with same campaignFor social media, use URL shorteners that preserve UTM:
Example workflow:
| Mistake | Problem | Solution | |---------|---------|----------| | Inconsistent capitalization | Creates duplicate campaigns in GA | Always use lowercase | | Spaces in parameters | Breaks URL | Use hyphens or underscores | | Missing utm_medium | Poor channel attribution | Always include medium | | Generic campaign names | Hard to analyze | Use date + descriptive name | | Not documenting | Lost tracking context | Maintain spreadsheet | | Duplicate utm_source | Confusion in reporting | Standardize naming |
After launching, verify in GA4:
Newsletter:
?utm_source=newsletter&utm_medium=email&utm_campaign={date}-{topic}&utm_content={position}
Drip Sequence:
?utm_source=hubspot&utm_medium=email&utm_campaign={sequence-name}&utm_content=email-{number}
Transactional:
?utm_source=transactional&utm_medium=email&utm_campaign=order-confirmationOrganic Post:
?utm_source={platform}-organic&utm_medium=social&utm_campaign={date}-{topic}
Paid Social:
?utm_source={platform}&utm_medium=cpc&utm_campaign={date}-{campaign}&utm_content={creative-name}Blog Post:
?utm_source=blog&utm_medium=organic&utm_campaign={post-slug}
Guest Post:
?utm_source={publication}&utm_medium=referral&utm_campaign=guest-post-{date}
Podcast:
?utm_source={podcast-name}&utm_medium=podcast&utm_campaign={episode}Co-Marketing:
?utm_source=partner-{name}&utm_medium=referral&utm_campaign={joint-campaign}
Affiliate:
?utm_source=affiliate-{name}&utm_medium=referral&utm_campaign=affiliate-programWorks well with:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 14,186 | 13,019 | -8% | 1 | 1 | 0% | 3,137 | 5,361 | +71% | 0 | 0 | — |
case-02 | fail→pass | 11,097 | 7,686 | -31% | 1 | 1 | 0% | 1,931 | 3,969 | +106% | 0 | 0 | — |
case-03 | fail→pass | 8,053 | 9,555 | +19% | 1 | 1 | 0% | 1,671 | 4,614 | +176% | 0 | 0 | — |
case-04 | fail→fail | 6,583 | 6,521 | -1% | 1 | 1 | 0% | 1,305 | 3,762 | +188% | 0 | 0 | — |
case-16 | fail→pass | 7,287 | 6,655 | -9% | 1 | 1 | 0% | 1,225 | 3,605 | +194% | 0 | 0 | — |
case-05 | fail→pass | 4,499 | 4,926 | +9% | 1 | 1 | 0% | 706 | 3,441 | +387% | 0 | 0 | — |
case-06 | pass→pass | 9,838 | 10,244 | +4% | 1 | 1 | 0% | 1,792 | 4,343 | +142% | 0 | 0 | — |
case-07 | fail→pass | 9,045 | 9,852 | +9% | 1 | 1 | 0% | 1,551 | 4,350 | +180% | 0 | 0 | — |
case-08 | fail→fail | 9,278 | 7,547 | -19% | 1 | 1 | 0% | 1,549 | 3,878 | +150% | 0 | 0 | — |
case-09 | fail→pass | 9,620 | 8,040 | -16% | 1 | 1 | 0% | 1,615 | 3,829 | +137% | 0 | 0 | — |
case-10 | fail→pass | 11,206 | 8,557 | -24% | 1 | 1 | 0% | 1,904 | 4,028 | +112% | 0 | 0 | — |
case-11 | fail→pass | 11,563 | 9,349 | -19% | 1 | 1 | 0% | 1,951 | 4,152 | +113% | 0 | 0 | — |
case-12 | fail→fail | 9,521 | 8,349 | -12% | 1 | 1 | 0% | 1,611 | 3,973 | +147% | 0 | 0 | — |
case-13 | fail→fail | 7,743 | 5,480 | -29% | 1 | 1 | 0% | 1,255 | 3,563 | +184% | 0 | 0 | — |
case-14 | fail→pass | 8,164 | 9,644 | +18% | 1 | 1 | 0% | 1,368 | 4,406 | +222% | 0 | 0 | — |
case-15 | fail→pass | 8,486 | 8,701 | +3% | 1 | 1 | 0% | 1,422 | 3,990 | +181% | 0 | 0 | — |
case-17 | fail→fail | 8,486 | 8,873 | +5% | 1 | 1 | 0% | 1,565 | 3,937 | +152% | 0 | 0 | — |
case-18 | pass→pass | 10,956 | 8,282 | -24% | 1 | 1 | 0% | 1,929 | 3,964 | +105% | 0 | 0 | — |
case-19 | pass→fail | 11,506 | 9,773 | -15% | 1 | 1 | 0% | 2,085 | 4,425 | +112% | 0 | 0 | — |
case-20 | pass→fail | 10,887 | 11,370 | +4% | 1 | 1 | 0% | 1,599 | 4,376 | +174% | 0 | 0 | — |
case-21 | pass→pass | 12,496 | 11,430 | -9% | 1 | 1 | 0% | 2,283 | 4,783 | +110% | 0 | 0 | — |
case-22 | pass→pass | 9,452 | 9,539 | +1% | 1 | 1 | 0% | 1,579 | 4,050 | +156% | 0 | 0 | — |
case-23 | pass→pass | 13,178 | 13,136 | -0% | 1 | 1 | 0% | 2,198 | 4,861 | +121% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted. The headline lift of +39 percentage points is the difference between those two pass rates over the 23 comparable cases. 2 cases got worse with the skill loaded, and they are included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.