A client wants to move from one-size-fits-all experiences to targeted, dynamic content and recommendations, whether they are starting from zero or looking to advance from basic rules to sophisticated 1:1 personalization.

---

How it works

1. You provide the product/channels, current personalization (if any), known segments, and available data
2. The skill defines dynamic content blocks, recommendation logic, behavioral triggers, progressive profiling strategy, a maturity roadmap from rules to 1:1, and privacy/consent requirements
3. It returns a complete personalization strategy with implementation priorities Kate can walk the client through

---

Prompt

You are building a personalization strategy for a Kate Makrigiannis consulting engagement. Kate uses this to help clients deliver the right content to the right person at the right time, without over-engineering it or creeping out their users. Before writing, read knowledge/voice-tone-guide.md -- use the client-facing voice.

Inputs I will provide:

• Product/Channels: {{PRODUCT}} (what the product is and which channels are in scope -- e.g., "B2B SaaS app + marketing emails + website," "e-commerce site + email + mobile app," "content platform + newsletter")
• Current personalization: {{CURRENT}} (what personalization exists today, if any -- e.g., "none, everyone sees the same thing," "basic email segmentation by plan tier," "product recommendations on homepage based on purchase history")
• Segments: {{SEGMENTS}} (known customer segments or personas -- formal or informal)
• Data available: {{DATA}} (what data the client collects -- e.g., "email engagement, purchase history, browsing behavior, CRM data, survey responses," or "very little beyond basic signup info")
• Context (optional): {{CONTEXT}} (goals, constraints, tech stack, team capacity, privacy concerns, competitive pressure)

Step 1: Personalization maturity assessment

Evaluate where the client is today and where they should aim:

Current Maturity Level

Level	Description	Status
Level 0: No personalization	Everyone sees the same content, emails, and experience	[Current / Not current]
Level 1: Rules-based	Static segments drive different experiences (e.g., new vs. returning, free vs. paid)	[Current / Not current]
Level 2: Segment-driven	Multiple segments with tailored content blocks, email flows, and product experiences	[Current / Not current]
Level 3: Behavioral	Real-time behavior triggers personalized responses (browse history, engagement patterns)	[Current / Not current]
Level 4: Predictive	ML models predict preferences, churn risk, purchase likelihood and act on predictions	[Current / Not current]
Level 5: 1:1 adaptive	Fully individualized experiences that learn and adapt in real time	[Current / Not current]

Current level: [Level X] Recommended target level (12-month horizon): [Level Y] Why not higher: [Specific constraint -- data limitations, team capacity, diminishing returns for the business model]

Step 2: Dynamic content block definitions

Define what content should change based on who is viewing it. For each channel in scope:

Dynamic Content Blocks: [Channel Name]

Block Location	What Changes	For Whom	Default (Fallback)	Data Source	Priority
[e.g., Homepage hero]	[Headline, image, CTA]	[Segment or trigger]	[What everyone sees by default]	[Behavior / profile / segment]	[P1/P2/P3]
[e.g., Email subject line]	[Subject, preview text]	[Segment or behavior]	[Generic version]	[Data source]	[Priority]
[e.g., Product page]	[Recommended products, social proof]	[Based on browse/purchase history]	[Best sellers]	[Behavioral data]	[Priority]
[e.g., Onboarding flow]	[Steps shown, examples used]	[Role, use case, plan tier]	[Generic onboarding]	[Signup data]	[Priority]
[e.g., Pricing page]	[Plan highlighted, testimonials shown]	[Company size, usage level]	[Most popular plan]	[Firmographic + usage]	[Priority]

Repeat for each channel in scope.

Content variation matrix

For the P1 content blocks, define the actual variants:

Block	Variant A (Segment)	Variant B (Segment)	Variant C (Segment)	Fallback
[Homepage hero]	[For enterprise: "Built for teams of 500+"]	[For SMB: "Get started in minutes"]	[For developers: "API-first platform"]	["The platform that scales with you"]
[CTA]	["Talk to Sales"]	["Start Free Trial"]	["Read the Docs"]	["Learn More"]

Step 3: Recommendation logic

Define the recommendation approach based on available data and business model:

Recommendation Strategy

Approach	How It Works	Best For	Data Required	Client Fit
Rules-based	If [condition], show [content]	Getting started, low data volume	Minimal -- segment membership, basic attributes	[Good / Moderate / Poor]
Collaborative filtering	"Users like you also liked X"	Products with many users and items	Purchase/usage history across users	[Fit assessment]
Content-based	"Based on what you have liked before"	Content platforms, catalogs	Individual user history + item attributes	[Fit assessment]
Hybrid	Combine collaborative + content-based	Mature personalization	Both user history and item metadata	[Fit assessment]
Contextual	Based on current session behavior, time, device	Everyone -- adds relevance layer	Real-time behavioral signals	[Fit assessment]

Recommended approach: [Approach] starting now, progressing to [approach] as data grows.

Recommendation rules (starting set)

Define 5-10 initial recommendation rules the client can implement immediately:

Rule #	Trigger	Action	Channel	Expected Impact
R1	[e.g., User viewed product category X 3+ times]	[Show top items in category X on homepage]	[Web]	[Higher engagement with homepage]
R2	[e.g., Cart abandoned with items > $100]	[Email with abandoned items + 10% offer]	[Email]	[Cart recovery]
R3	[e.g., New user, first session]	[Show getting-started guide instead of advanced features]	[In-app]	[Faster activation]
R4	[Trigger]	[Action]	[Channel]	[Impact]
R5	[Trigger]	[Action]	[Channel]	[Impact]

Step 4: Behavioral triggers and response actions

Trigger-Response Framework

Trigger Category	Specific Trigger	Response Action	Timing	Channel
Engagement signals	[e.g., Visited pricing page 2+ times]	[Show chat widget with "Questions about pricing?"]	[Real-time]	[In-app]
Engagement signals	[e.g., Opened 5+ emails without clicking]	[Switch to shorter, CTA-focused email format]	[Next email send]	[Email]
Lifecycle events	[e.g., 7 days since signup, no core action]	[Send activation nudge with quick-start guide]	[Day 7]	[Email + push]
Lifecycle events	[e.g., Subscription renewal in 14 days]	[Show renewal reminder with usage summary]	[14 days before]	[In-app + email]
Risk signals	[e.g., Usage dropped 50% week-over-week]	[Trigger customer success check-in]	[Within 48 hours]	[Email / CS outreach]
Risk signals	[e.g., Support ticket with negative sentiment]	[Escalate to account manager, pause upsell emails]	[Immediate]	[Internal routing]
Growth signals	[e.g., Approaching usage limit on current plan]	[Show upgrade prompt with value comparison]	[At 80% of limit]	[In-app]
Growth signals	[e.g., Added 3+ team members]	[Send team collaboration tips, suggest team plan]	[After 3rd invite]	[Email]

Step 5: Progressive profiling strategy

Define what to learn about each user and when, without asking for everything upfront:

Progressive Profiling Roadmap

Stage	Timing	What to Learn	How to Learn It	Why Now
Signup	Registration	[Name, email, company size (if B2B)]	[Signup form -- 2-3 fields max]	[Minimum viable profile to route the experience]
Onboarding	First session	[Role, primary use case, goals]	[In-app survey or onboarding flow choices]	[Personalize onboarding immediately]
Week 1	After 2-3 sessions	[Feature preferences, integration needs]	[Track behavior + optional 1-question survey]	[Enough context to tailor recommendations]
Month 1	After value delivered	[Satisfaction, additional needs, team size]	[NPS survey + usage pattern analysis]	[Relationship established, willing to share more]
Ongoing	Continuous	[Evolving preferences, new needs]	[Behavioral tracking, periodic check-ins]	[Profile stays current as needs change]

Data collection principles

• Ask only what you will act on. If you will not personalize based on "job title," do not ask for it.
• Infer before asking. If behavior reveals the answer (e.g., which features they use reveals their use case), do not interrupt with a survey.
• Trade value for information. Every question should come with a visible benefit: "Tell us your role to see a customized dashboard."
• Respect the interrupt budget. Maximum one profiling question per session. Batch profiling kills engagement.

Step 6: Personalization maturity roadmap

12-Month Roadmap

Phase	Timeline	Focus	Key Deliverables	Prerequisites
Phase 1: Foundation	Months 1-3	Rules-based personalization	[5-10 rules implemented, tracking in place, segment definitions live]	[Analytics setup, segment definitions, content variants created]
Phase 2: Segments	Months 4-6	Segment-driven personalization	[Dynamic content blocks live on key pages/emails, A/B testing framework]	[Phase 1 complete, enough data for segment validation]
Phase 3: Behavioral	Months 7-9	Behavioral triggers and responses	[Trigger-response system live, real-time personalization on 2-3 channels]	[Event tracking, behavioral data pipeline, trigger logic platform]
Phase 4: Intelligence	Months 10-12	Predictive and adaptive	[Recommendation engine v1, predictive churn scoring, automated optimization]	[6+ months of behavioral data, ML capability or vendor]

Phase 1 implementation plan (detailed)

Week	Action	Owner	Dependency
Week 1-2	[Define segments and content variants for top 3 pages]	[Marketing + Product]	[Segment definitions from audience-segmentation]
Week 3-4	[Implement tracking for personalization triggers]	[Engineering]	[Analytics platform access]
Week 5-8	[Build and launch first 5 rules]	[Marketing + Engineering]	[Content created, tracking live]
Week 9-12	[Measure impact, iterate, plan Phase 2]	[All]	[4+ weeks of data]

Step 7: Privacy and consent considerations

Privacy Framework

Consideration	Approach	Implementation
Consent collection	[How and when to get permission for personalization]	[Cookie banner, preference center, in-app consent]
Data minimization	[Collect only what is needed for personalization]	[Audit data collection against personalization use cases]
Transparency	[Users should understand why they see what they see]	["Why am I seeing this?" link, personalization settings page]
Opt-out	[Users can disable personalization]	[Preference center with granular controls]
Data retention	[How long behavioral data is stored]	[Define retention periods by data type -- e.g., 90-day behavioral, 2-year transactional]
Regulatory compliance	[GDPR, CCPA, other applicable regulations]	[Legal review of personalization data flows, DPIA if needed]

Kate's Talking Points

• "You are at Level [X] personalization maturity today. Within 12 months, you can reach Level [Y], which means [concrete example of what changes for the user]."
• "The highest-impact starting point is [specific content block or trigger] because [it affects X% of users and the current one-size-fits-all version is leaving value on the table]."
• "You do not need AI or a recommendation engine to start. Rules-based personalization using the segments you already have can deliver [X%] improvement while you build toward more sophisticated approaches."

Related skills: Uses /audience-segmentation for segment definitions that drive personalization rules. Feeds into /email-campaign-builder for personalized email flows. Pairs with /activation-optimization for personalizing the onboarding and activation experience.

---