Attribution Model Designer

A client needs to understand which marketing touchpoints actually drive conversions, choose the right attribution model for their business, or build a measurement framework that accounts for cross-channel interactions.

---

How it works

1. You provide the channels in use, sales cycle length, current tracking setup, and business model
2. The skill evaluates attribution model options, recommends the best fit with pros/cons specific to the client's context, defines implementation requirements, and designs a reporting structure
3. It returns a complete attribution model recommendation with implementation roadmap and reporting framework Kate can walk the client through

---

Prompt

You are designing a marketing attribution model for a Kate Makrigiannis consulting engagement. Kate uses this to help clients stop guessing which channels are working and start making budget decisions based on evidence. Before writing, read knowledge/voice-tone-guide.md -- use the client-facing voice.

Inputs I will provide:

• Channels in use: {{CHANNELS}} (all marketing channels -- paid search, organic search, social ads, email, content, events, partnerships, referrals, direct sales, etc.)
• Sales cycle length: {{SALES_CYCLE}} (average time from first touch to conversion -- e.g., "same-session for e-commerce" or "90-day B2B enterprise cycle")
• Current tracking setup: {{TRACKING}} (what is in place -- e.g., "Google Analytics 4 with basic UTMs," "HubSpot CRM with form tracking," "nothing formal," "Mixpanel + Salesforce")
• Business model: {{BUSINESS_MODEL}} (e.g., "B2B SaaS, $30K ACV," "DTC e-commerce, $80 AOV," "marketplace with buyer and seller sides")
• Context (optional): {{CONTEXT}} (specific attribution questions, known blind spots, recent channel additions, budget decisions that depend on attribution, team analytics maturity)

Step 1: Current state assessment

Assess the client's attribution maturity and identify gaps:

Attribution Maturity Assessment

Dimension	Current State	Maturity Level	Gap
Tracking coverage	[What is tracked vs. what is not]	[None / Basic / Intermediate / Advanced]	[Key gaps]
Cross-channel visibility	[Can they see multi-touch journeys?]	[Level]	[Gaps]
Cross-device tracking	[Can they connect mobile and desktop?]	[Level]	[Gaps]
Offline touchpoints	[Events, sales calls, direct mail tracked?]	[Level]	[Gaps]
CRM integration	[Marketing data flows to revenue data?]	[Level]	[Gaps]
Reporting cadence	[How often attribution data is reviewed]	[Level]	[Gaps]

Overall maturity: [Level 1-5 with label]

• Level 1: No attribution (last-click by default, no UTMs)
• Level 2: Basic (UTM tracking, single-platform reporting)
• Level 3: Intermediate (multi-channel tracking, one attribution model applied)
• Level 4: Advanced (multi-touch attribution, CRM-connected, regular optimization)
• Level 5: Sophisticated (data-driven/algorithmic attribution, incrementality testing, MMM)

Step 2: Attribution model comparison

Compare all standard models against the client's specific business context:

Model Comparison for {{BUSINESS_MODEL}}

Model	How It Works	Credit Distribution	Best For	Limitations
First-touch	100% credit to first interaction	All credit to awareness channel	Understanding which channels fill the top of funnel	Ignores everything after first touch; overvalues awareness
Last-touch	100% credit to final interaction before conversion	All credit to closing channel	Short sales cycles, single-session conversions	Ignores all earlier touchpoints; overvalues bottom-funnel
Linear	Equal credit to every touchpoint	Even split across all touches	Businesses that genuinely value every interaction equally	Treats a display impression the same as a demo request
Time-decay	More credit to touchpoints closer to conversion	Weighted toward recent touches	Longer sales cycles where recent touches matter more	Still undervalues awareness; decay rate is arbitrary
Position-based (U-shaped)	40% first, 40% last, 20% split across middle	Heavy on first and last	Businesses that value both acquisition and conversion	Middle touches may be more important than 20% suggests
Data-driven	Machine learning assigns credit based on conversion patterns	Varies by actual impact	High-volume businesses with enough data for statistical models	Requires significant conversion volume; black box concerns

Recommendation for This Business

Recommended model: [Model name]

Why this model fits:

1. [Reason tied to sales cycle length]
2. [Reason tied to channel mix complexity]
3. [Reason tied to data/tracking maturity]
4. [Reason tied to business model]

Why not the alternatives:

• [Model X]: Not recommended because [specific reason for this client -- e.g., "sales cycle is 90 days, so last-touch would ignore 3 months of nurturing"]
• [Model Y]: Not recommended because [reason]

Step 3: Channel weighting rationale

Define how each channel should be valued in the recommended model:

Channel Weighting

Channel	Role in Journey	Typical Position	Weight Rationale	Measurement Notes
[Paid Search - Brand]	Conversion/capture	Last touch	[High conversion intent but often captures existing demand -- weight carefully]	[Easy to over-credit; consider brand lift studies]
[Paid Search - Non-brand]	Discovery/consideration	First or mid	[Introduces new prospects to the brand]	[Track assisted conversions separately]
[Organic Search/SEO]	Discovery/education	First or mid	[Long-tail value, compounds over time]	[Content assists are often invisible in last-touch]
[Social Ads]	Awareness/retargeting	First or last	[Dual role: cold audience and warm retargeting behave differently]	[Separate prospecting from retargeting in reporting]
[Email]	Nurture/conversion	Mid or last	[High conversion but only reaches known contacts]	[Often gets last-touch credit for work other channels started]
[Content Marketing]	Education/trust	Mid	[Rarely gets conversion credit but influences heavily]	[Track content-assisted conversions as a separate metric]
[Events/Webinars]	Relationship/trust	Mid	[High-value touches, hard to track digitally]	[Manual CRM tagging usually required]
[Referrals/Word-of-mouth]	Trust/conversion	First or last	[Often invisible in digital tracking]	[Post-purchase survey: "How did you hear about us?"]

Step 4: Implementation requirements

Tracking Infrastructure

Requirement	Current State	Action Needed	Priority	Effort
UTM parameter framework	[In place / Partial / None]	[Define UTM taxonomy, implement across all channels]	[P1/P2/P3]	[Hours/days estimate]
Conversion tracking pixels	[State]	[Action]	[Priority]	[Effort]
Cross-domain tracking	[State]	[Action]	[Priority]	[Effort]
CRM integration	[State]	[Action]	[Priority]	[Effort]
Offline conversion import	[State]	[Action]	[Priority]	[Effort]
Consent/cookie management	[State]	[Action]	[Priority]	[Effort]

UTM Taxonomy

Define a consistent UTM structure the client should use across all channels:

Parameter	Convention	Example
utm_source	Platform name, lowercase	google, facebook, linkedin, newsletter
utm_medium	Channel type	cpc, social, email, organic, referral
utm_campaign	Campaign name, kebab-case	q1-brand-awareness, product-launch-2026
utm_content	Creative or ad variant	headline-a, video-30s, banner-300x250
utm_term	Keyword (search only)	{keyword} dynamic parameter

Cross-Device Considerations

Scenario	Frequency	Solution	Complexity
Mobile browse, desktop convert	[High for B2B, moderate for DTC]	[Logged-in user ID matching, Google Signals, probabilistic matching]	[Medium-High]
Multi-device research	[Common for high-consideration purchases]	[User ID implementation across platforms]	[High]
App-to-web transitions	[Relevant if client has mobile app]	[Deep linking, deferred deep links, Firebase]	[High]
Offline-to-online	[Events, print, TV driving web visits]	[Vanity URLs, QR codes, post-visit surveys]	[Low-Medium]

Step 5: Reporting structure

Attribution Dashboard Design

Report	Frequency	Audience	Key Metrics	Purpose
Channel performance summary	Weekly	Marketing team	Attributed conversions, CPA, ROAS by channel	Tactical optimization
Multi-touch journey analysis	Monthly	Marketing leadership	Top conversion paths, average touches to conversion, path length	Strategic channel planning
Model comparison report	Quarterly	CMO/VP Marketing	Same data under multiple models side-by-side	Validate model choice, catch blind spots
Incrementality snapshot	Quarterly	Marketing + Finance	Holdout test results, lift measurements	True causal impact vs. correlation

Key Metrics to Track

Metric	Definition	Why It Matters
Attributed conversions	Conversions assigned to each channel under the selected model	Core budget allocation input
Assisted conversions	Conversions where a channel appeared in the path but did not get primary credit	Reveals hidden channel value
Assist-to-last ratio	Assisted conversions / Last-touch conversions per channel	Channels with high ratio are undervalued in last-touch
Average path length	Number of touchpoints before conversion	Indicates journey complexity
Time to conversion	Days from first touch to conversion	Validates or challenges sales cycle assumptions
Cross-device conversion rate	Conversions that span multiple devices	Quantifies cross-device measurement gap

Kate's Talking Points

• "Your current setup is at Level [X] attribution maturity. The biggest gap is [specific gap]. Closing that gap would change how you allocate roughly $[X] in marketing spend."
• "I recommend [model] because your [sales cycle / channel mix / data maturity] means [specific reasoning]. Under this model, [channel] is likely getting [more/less] credit than it deserves today."
• "The first implementation step is [specific action]. That alone will give you visibility into [specific blind spot] within [timeframe]."

Related skills: Feeds into /campaign-post-mortem for measuring campaign performance under the selected model. Pairs with /marketing-roi-analyzer for channel ROI calculations. Uses /analytics-dashboard-design for building the attribution reporting dashboard.

---