AI Context Explained
Context is the information AI uses to understand your business and provide relevant recommendations. The more context AI has - business goals, target audience, optimization history - the better its insights become. This guide explains what context is, what data AI accesses, and how to improve it.
What You'll Learn
- What context means for AI
- Complete list of data sources AI uses
- How context improves recommendations
- Business Profile importance
- Optimization history tracking
- How to improve AI context
- Context privacy and security
Time needed: 10 minutes to read
What Is Context?
Definition: All information AI can access to understand your business, campaigns, and goals
Why it matters:
- Without context: Generic recommendations
- With context: Tailored to your specific business
Example:
Without context:
AI: "Your CPA is $50. That's higher than average."
With context (AI knows your target CPA is $45):
AI: "Your CPA is $50, which is 11% above your target of $45.
This gap is costing approximately $280/month in efficiency
losses. I recommend improving Quality Score on your top
5 keywords to bring CPA back to target."
[Screenshot: Context data sources diagram] AI synthesizes data from 12 different sources
The 12 Data Sources
1. Campaign Performance Data
What AI sees:
- Campaign names, types, status
- Daily/monthly spend
- Clicks, impressions, conversions
- CTR, CPC, conversion rate
- Impression share data
- Budget settings
- Bidding strategies
Data freshness: Last Smart Sync (shown in chat)
How AI uses it:
- Identify underperforming campaigns
- Compare campaign efficiency
- Recommend budget reallocation
- Detect trend anomalies
2. Ad Group & Keyword Data
What AI sees:
- All ad groups and structure
- All keywords and match types
- Quality Scores
- Keyword-level performance
- Search term data (matched queries)
- Negative keyword lists
How AI uses it:
- Find wasted keyword spend
- Recommend keyword expansion
- Identify negative keyword opportunities
- Analyze Quality Score issues
3. Ad Copy & Creative
What AI sees:
- All ad headlines and descriptions
- Ad status (approved, disapproved, pending)
- Ad performance by variant
- Extension usage
- Image ads (if Shopping/Display)
How AI uses it:
- Suggest ad copy improvements
- Identify policy violations
- Recommend A/B test variations
- Compare ad performance
4. Conversion Tracking
What AI sees:
- Conversion actions defined
- Conversion counts and values
- Attribution windows
- Conversion rates by campaign/keyword
- Time to conversion data
How AI uses it:
- Calculate accurate ROI metrics
- Identify high-converting traffic sources
- Recommend conversion optimization
- Track optimization impact
5. Geographic Data
What AI sees:
- Performance by state/city
- Location bid adjustments
- Geo targeting settings
- Regional conversion rates
How AI uses it:
- Find best-performing locations
- Recommend geo bid adjustments
- Identify expansion opportunities
- Detect regional inefficiencies
6. Device & Time Data
What AI sees:
- Performance by device (mobile, desktop, tablet)
- Performance by day of week
- Performance by hour of day
- Device bid adjustments
- Ad schedule settings
How AI uses it:
- Optimize ad scheduling
- Recommend device bid adjustments
- Identify best performance times
- Suggest budget pacing changes
7. Budget Tracking
What AI sees:
- Monthly budget targets
- Daily spend patterns
- Budget utilization percentage
- Budget pacing trends
- Campaign-level budgets
How AI uses it:
- Alert on budget exhaustion
- Recommend budget increases
- Identify budget waste
- Optimize budget distribution
8. Active AI Alerts
What AI sees:
- All current alerts
- Alert severity levels
- Alert categories
- Past accept/decline history
How AI uses it:
- Reference alerts in conversation
- Explain alert reasoning
- Prioritize recommendations
- Learn from your feedback
9. Business Profile ⭐
What AI sees (if you fill it out):
- Industry vertical
- Business type (B2B, B2C, local, e-commerce)
- Company size
- Target audience description
- Business goals
- Target CPA/ROAS
- Monthly budget
- Website URL
- Main products/services
How AI uses it:
- Calibrate "good" vs "bad" performance
- Understand strategic constraints
- Provide industry-specific advice
- Set appropriate severity levels
Critical: This is the most important context source
10. Optimization History
What AI sees:
- Changes you've made
- When optimizations were implemented
- Performance before/after changes
- Alert accept/decline patterns
- Conversation summaries
How AI uses it:
- Don't repeat failed recommendations
- Build on successful optimizations
- Understand your preferences
- Track optimization impact
11. Custom Context Entries
What AI sees (if you add custom notes):
- Client-specific instructions
- Strategic priorities
- Constraints and limitations
- Campaign goals
- Special considerations
Examples:
"Client is rebranding next month - don't recommend
major campaign restructures until after rebrand"
"Target audience is enterprise, not SMB - filter
recommendations accordingly"
"Client prefers manual bidding - don't suggest
automation"
How AI uses it:
- Tailor recommendations to your specific situation
- Avoid conflicting with known constraints
- Respect strategic priorities
12. Website Analysis (During Onboarding)
What AI sees (one-time during client setup):
- Website content analysis
- Product/service offerings
- Value propositions
- Target audience indicators
- Competitor positioning
- Industry category
How AI uses it:
- Understand business model
- Suggest relevant keywords
- Recommend campaign structure
- Identify unique selling points
How Context Improves Recommendations
Example 1: CPA Evaluation
Without Business Profile:
Your question: "Is my CPA of $42 good?"
AI: "A CPA of $42 is moderate. Industry average varies
by vertical. You should compare to your specific goals."
(Generic answer - not helpful)
With Business Profile (target CPA: $35):
Your question: "Is my CPA of $42 good?"
AI: "Your CPA of $42 is 20% above your target of $35.
This represents approximately $420/month in lost efficiency.
To bring CPA down to target, I recommend:
1. Improve Quality Score (currently 5.2, should be 7+)
2. Add 12 negative keywords I identified
3. Pause 3 underperforming ad groups
These changes should reduce CPA by ~18%, bringing you
to $34.40."
(Specific, actionable, quantified)
Example 2: Budget Recommendations
Without Monthly Budget Context:
AI: "You could increase budget to capture more traffic."
(Vague, may not be feasible)
With Monthly Budget ($5,000/month specified):
AI: "You're currently spending $3,800 of your $5,000
monthly budget (76% utilization). This leaves $1,200
on the table.
I recommend reallocating this $1,200 to:
- Brand Campaign: +$800 (budget-limited, high ROI)
- Non-Brand Campaign: +$400 (room for expansion)
Expected result: +35 conversions/month without exceeding
your total budget."
(Specific, respects constraints, quantified impact)
Example 3: Strategic Alignment
Without Industry Context:
AI: "Consider expanding to Shopping campaigns."
(May not apply to your business)
With Industry (Professional Services / B2B):
AI: "Given your professional services B2B business model,
Shopping campaigns are not applicable (you don't sell
physical products).
Instead, I recommend:
1. LinkedIn Ads for B2B targeting
2. Remarketing to nurture leads through long sales cycle
3. Thought leadership content campaigns
These align better with B2B service sales."
(Industry-appropriate, strategic)
Business Profile: The Most Important Context
Why It Matters
Business Profile unlocks:
- Accurate performance benchmarks
- Industry-specific recommendations
- Goal-aligned optimization
- Appropriate severity ratings
ROI of 10 minutes to fill out:
- 3-5x more relevant recommendations
- Fewer generic suggestions
- Better alert accuracy
- Faster decision-making
Key Business Profile Fields
Target CPA (critical):
Without: AI can't judge if $50 CPA is good or bad
With: AI knows if you're above/below target
Monthly Budget (critical):
Without: AI might suggest $10K/month strategies
With: AI respects your $2K/month reality
Business Goals (important):
Without: AI optimizes for clicks
With: AI optimizes for your actual goals (leads, sales, calls)
Target Audience (important):
Without: AI suggests broad targeting
With: AI recommends audience-aligned strategies
Industry (helpful):
Without: AI gives generic advice
With: AI provides industry best practices
Optimization History: AI Learning
How It Works
Every change you make is tracked:
Oct 1: Increased Brand Campaign budget $100→$150
Result: +22 conversions, CPA stayed same ($42→$41)
AI learns: Budget increases work for this client
Oct 8: Tried Display Campaign
Result: 5 conversions, $95 CPA (vs $42 average)
AI learns: Display doesn't work for this client
Oct 15: Added 15 negative keywords
Result: CTR +12%, wasted spend -$280/month
AI learns: Negative keywords are high-impact here
Future recommendations improve:
AI won't suggest: Display campaigns (learned it failed)
AI will suggest: More negative keyword reviews (learned it works)
AI will suggest: Budget increases (learned they're effective)
Accept/Decline Feedback Loop
When you accept alerts:
- AI notes what was valuable
- Similar alerts get higher priority
- Recommendation style reinforced
When you decline alerts:
- AI notes what wasn't useful
- Similar alerts get lower priority
- Recommendation style adjusted
Over time: AI becomes personalized to your preferences
Custom Context Entries
What They Are
Free-text notes you can add about client-specific considerations
Examples:
Strategic constraints:
"Client is switching from products to services next quarter.
Don't recommend Shopping campaigns."
Special targeting:
"Focus on enterprise buyers ($1M+ company size).
Avoid small business targeting."
Temporary situations:
"Website is being redesigned Oct 15-25.
Pause conversion tracking alerts during this period."
Budget context:
"Budget will 3x in Q4 for holiday season.
Plan expansion strategies now."
Preference notes:
"Client prefers manual bidding for control.
Don't suggest automated strategies."
When to Use Custom Context
Use when:
- AI doesn't have a structured field for your situation
- Temporary constraints exist
- Strategic priorities are unique
- Client has strong preferences
- Special circumstances apply
Don't use for:
- Information that belongs in Business Profile
- Data AI already has (campaign performance)
- Things that change frequently
Context Privacy & Security
What Is Private
Your data stays in your account:
- Other users cannot see your client data
- Other users cannot see your conversations
- Other users cannot see your custom context
Your accept/decline feedback:
- Only affects your account's AI behavior
- Contributes to aggregate model improvement (anonymized)
- Never identifies you or your clients
What AI Cannot See
Completely private:
- Google Ads login credentials
- Payment information
- User passwords
- Other clients (in per-client chat)
- Unrelated user data
External data:
- Google Analytics (not integrated)
- CRM data (not integrated)
- Email marketing data
- Social media data
Improving AI Context
Fill Out Business Profile (10 minutes)
Highest impact action:
- Go to Client Details → Settings
- Fill out business profile fields
- Save
Immediate benefit: Next AI conversation 3-5x better
Add Custom Context (5 minutes)
For unique situations:
- Go to Client Details → Context tab
- Add custom context entry
- Describe special considerations
Example:
Title: "Seasonal Budget Strategy"
Note: "Client doubles budget Nov-Dec for holiday season.
Any expansion recommendations should account for this."
Keep Optimizations Documented
Track what you do:
- Accept alerts when you implement recommendations
- Note major changes in custom context
- AI learns faster with clear feedback
Regular Data Syncs
Fresh data = better context:
- Set up auto-sync (daily recommended)
- Manual sync before important decisions
- Context is only as good as data recency
Common Questions
Q: Can AI see my other clients in per-client chat? A: No. Per-client chat is sandboxed to that client only. Global chat can discuss all clients.
Q: Does filling out Business Profile cost anything? A: No. It's included and highly recommended for better AI.
Q: Can I delete or edit custom context? A: Yes. Go to Context tab, edit or delete any entry anytime.
Q: How often should I update Business Profile? A: When goals change (new target CPA, budget increase, strategy shift).
Q: Does AI remember conversations forever? A: Conversations auto-summarize after 10 messages. Full history kept for 90 days.
Q: Can I export my context data? A: Business Profile and custom context can be exported as part of client data export.
Troubleshooting
Problem: AI gives generic recommendations
Cause: Missing Business Profile
Solution:
- Fill out Business Profile (especially target CPA, budget, goals)
- Wait for next conversation
- AI will immediately use new context
Problem: AI suggests things that don't apply
Cause: Missing industry or business type context
Solution:
- Update Business Profile with industry
- Add custom context note explaining business model
- Example: "B2B SaaS, not e-commerce - don't suggest Shopping"
Problem: AI recommends things you already tried
Cause: Optimization history not captured
Solution:
- Accept alerts when you implement them
- Add custom context note about past attempts
- Example: "Tried Display campaigns Q1 2025 - poor ROI, don't suggest again"
Next Steps
Maximize AI effectiveness with these guides:
- Business Profile Setup - Complete Business Profile guide
- Custom Context Entries - Advanced context management
- AI Chat Getting Started - Use AI Chat effectively
- Smart Alerts - How context improves alerts
Fill out Business Profile now for immediate improvement
Add custom context for unique situations
Keep context updated as business evolves
Last Updated: October 4, 2025