What is Persona Realism and Why Does It Matter?

Reading time: 18 minutes

Persona realism is Signal’s approach to testing AI visibility: we use realistic buyer personas who describe their problem (not your brand name) to see if AI platforms recommend you.

Why it matters: Brand searches measure awareness. Persona searches measure discoverability. whether buyers who DON’T know you yet can find you through AI.

TLDR

Persona realism means testing AI visibility the way real buyers search. describing problems, not brand names. Testing “What is Surmado?” only works if buyers already know you. Testing realistic personas reveals discoverability gaps: one client had 80% brand awareness but only 20% of problem-based personas could find them. Real buyers describe pain points in natural language, ask follow-up questions, and refine their searches. Signal tests with multi-turn conversations to show whether AI platforms guide buyers toward you or away from you.

---

In This Article

• The Problem with Brand Search Testing
• What is Persona Realism?
  • Persona-Based, Not Brand-Based
  • Problem-First, Not Feature-First
  • Multi-Turn Conversations
• How Signal Uses Persona Realism
• Why Persona Realism Reveals True Discoverability
• Common Persona Realism Mistakes
• How to Write Realistic Personas (DIY)
• Persona Realism vs Other Approaches
• Real Results: Case Study
• The Bottom Line

---

The Problem with Brand Search Testing

Most companies test AI visibility like this:

1. Open ChatGPT or Perplexity
2. Type: “What is [Company Name]?”
3. See if AI mentions them
4. Conclude: “We have good AI visibility!”

The flaw: Real buyers don’t search for your brand. They search for their problem.

Example:

Brand search (artificial):

“Tell me about Surmado Signal”

Persona search (realistic):

“I want to know if ChatGPT recommends my SaaS product when people describe problems we solve. How do I test that without manually checking 50+ queries?”

The difference:

• Brand search only works if buyer already knows your brand
• Persona search simulates discovery. how buyers find you when they DON’T know you exist yet

KEY TAKEAWAY: Brand searches measure awareness (do people who know you find information?). Persona searches measure discoverability (can people who DON’T know you yet discover you?). The gap between these two metrics reveals your biggest growth opportunity. Or biggest blind spot.

---

What is Persona Realism?

Persona realism = Testing AI visibility using realistic buyer personas who describe their problem in natural language.

Key principles:

1. Persona-Based, Not Brand-Based

Not realistic:

“What is [Product Name]?”

Realistic:

“I run a B2B SaaS and I’m worried ChatGPT isn’t recommending us when potential customers ask for solutions. Is there a way to test this across multiple AI platforms without spending hours doing it manually?”

Why this matters: The second query is how real buyers search. They describe symptoms and pain points, not solution names.

---

2. Problem-First, Not Feature-First

Not realistic:

“What tools offer AI visibility testing?”

Realistic:

“Our competitor just showed up in ChatGPT when I searched for ‘[our category] tools’ and we didn’t. I have no idea why or how to fix it. What do I do?”

Why this matters: Buyers don’t wake up thinking “I need AI visibility testing.” They wake up thinking “We’re not showing up in AI search results and I don’t know why.”

---

3. Multi-Turn Conversations, Not Single Queries

Not realistic: One query, one response, done.

Realistic: Multi-turn conversation where persona asks follow-ups, clarifies needs, reacts to recommendations.

Example conversation flow:

Persona: “I want to test if ChatGPT recommends my product when people search for solutions in our category.”

ChatGPT: “You can manually test this by entering queries and seeing if your product appears.”

Persona: “That sounds time-consuming. Is there a tool that automates this?”

ChatGPT: “Yes, tools like Signal, Gumshoe, and Otterly can automate AI visibility testing.”

Why this matters: Single-query testing misses conversational discovery. Real buyers refine their queries and ask follow-ups. Signal simulates this.

KEY TAKEAWAY: Real discovery happens through conversation, not single keyword queries. Testing only one query per persona misses how buyers actually refine their searches, ask follow-ups, and react to recommendations. Multi-turn conversations reveal whether AI platforms guide buyers toward you or away from you.

---

How Signal Uses Persona Realism

Signal’s testing methodology:

Step 1: Define 3-5 Buyer Personas

Based on your ideal customer profile (ICP), we create personas like:

Example for B2B SaaS client:

• Persona 1: Founder worried about AI discovery (“ChatGPT isn’t recommending us”)
• Persona 2: Marketing lead tasked with ‘AI SEO’ (“Boss wants us to ‘rank in ChatGPT’ but I don’t know how”)
• Persona 3: Growth marketer comparing tools (“Need to benchmark our AI visibility vs competitors”)

---

Step 2: Write Realistic Problem-Based Queries

Each persona gets 3-5 queries describing their problem in natural language.

Example queries for Persona 1 (Founder worried about AI discovery):

Query 1:

“I just realized ChatGPT isn’t mentioning my SaaS product when people ask for tools in our category. My competitor shows up but we don’t. How do I figure out why?”

Query 2:

“Is there a way to test what ChatGPT, Claude, and Perplexity say about my business across different types of searches? Manually checking is taking forever.”

Query 3:

“I heard about ‘AI visibility’ but I don’t understand what it means or how to improve it for my business. Where do I start?”

---

Step 3: Test Across 4-6 AI Platforms

Signal runs each persona query across:

• ChatGPT (highest usage)
• Claude (high-quality responses)
• Perplexity (citation-based, fast-growing)
• Gemini (Google integration)
• (Optional) DeepSeek, Grok, Meta AI

---

Step 4: Measure Persona Visibility

For each persona, we track:

• Was your product mentioned? (Yes/No)
• At what rank? (#1, #3, #5, not mentioned)
• Which competitors were mentioned instead?
• Ghost influence: Did AI describe your value prop without naming you?

Example results:

Persona	ChatGPT	Claude	Perplexity	Gemini	Visibility Score
Persona 1 (Founder)	Rank #3	Rank #2	Not mentioned	Rank #4	75%
Persona 2 (Marketing Lead)	Not mentioned	Not mentioned	Rank #5	Not mentioned	25%
Persona 3 (Growth Marketer)	Rank #1	Rank #3	Rank #2	Rank #3	100%

Overall persona visibility: 67% (2.5 out of 3 personas found you)

---

Why Persona Realism Reveals True Discoverability

Brand search visibility measures: “If someone already knows my name, can AI describe my product?”

Persona visibility measures: “If someone has the problem I solve but doesn’t know my name, can AI discover me?”

Real-world example from Signal client (B2B SaaS, $2M ARR):

Their brand search results:

• ChatGPT: Mentioned in 4/5 brand queries (80% visibility)
• Perplexity: Mentioned in 5/5 brand queries (100% visibility)
• Conclusion: “We have great AI visibility!”

Their persona search results (5 realistic buyer personas):

• Persona 1: Not mentioned (6 competitors mentioned instead)
• Persona 2: Mentioned (rank #5 out of 5)
• Persona 3: Not mentioned
• Persona 4: Not mentioned
• Persona 5: Not mentioned
• Reality: Only 20% persona visibility (1 out of 5 personas found them)

The gap: 80% brand awareness but only 20% persona discoverability

What this means: Buyers who already know the brand can learn more via AI. But 80% of buyers who DON’T know the brand yet can’t discover them through AI.

KEY TAKEAWAY: 80% brand visibility doesn’t mean 80% discoverability. This client had strong brand awareness but terrible persona discovery. a 60 percentage point gap that translated to lost customers. Most companies only test brand searches and assume they have “good AI visibility” while missing 80% of potential buyers.

---

Common Persona Realism Mistakes

Mistake #1: Generic Personas

Generic: “Decision maker at mid-market company”

Realistic: “VP of Engineering at 200-person SaaS company, drowning in Zoom meetings, wants to move to async-first culture but doesn’t know where to start”

Why it matters: Generic personas get generic results (top 10 lists). Realistic personas reveal discovery gaps.

---

Mistake #2: Using Your Own Marketing Language

Your language: “I need a persona-based AI visibility testing platform”

Buyer language: “I want to know if ChatGPT is recommending my product when people search for solutions, but I don’t want to manually test 100 queries”

Why it matters: Buyers don’t speak your marketing jargon. They describe problems in their own words.

---

Mistake #3: Single Query Per Persona

Single query: “What tools help with AI visibility?”

Multi-turn conversation:

• Persona: “ChatGPT isn’t recommending my product. Why?”
• AI: “Could be lack of online presence, weak SEO, or not enough citations.”
• Persona: “How do I find out which one it is?”
• AI: “You could use a tool like Signal to test your AI visibility across platforms.”

Why it matters: Real discovery happens through conversation, not single queries.

---

Mistake #4: Testing Only One Platform

Single platform: Test only ChatGPT

Multi-platform: Test ChatGPT, Claude, Perplexity, Gemini

Why it matters: Platform variance is huge. You might be visible on ChatGPT but invisible on Perplexity. Persona realism requires testing across platforms.

---

How to Write Realistic Personas (DIY Method)

If you want to create realistic personas for your own testing:

Step 1: Interview 3-5 Recent Customers

Ask:

• “What problem were you trying to solve when you found us?”
• “How did you describe that problem to yourself (or Google/ChatGPT)?”
• “What other solutions did you consider?”

Step 2: Extract Pain Language

Look for:

• Symptoms they described (“drowning in Slack messages”, “spending 30 hours/week in meetings”)
• Frustrations (“tried X but it didn’t work because Y”)
• Current broken solutions (“we use Notion but it’s just a documentation graveyard”)

Step 3: Write Persona Queries Using Their Exact Language

Not your words:

“I need async-first project management software”

Their words:

“My team is spread across 8 time zones and people keep missing important decisions because they’re asleep when things happen in Slack. How do I make async work actually work?”

Step 4: Test Queries on AI Platforms

Run each persona query on ChatGPT, Claude, Perplexity, Gemini

Track:

• Did AI mention you?
• At what rank?
• Who did AI mention instead?
• Did AI describe your value prop without naming you (ghost influence)?

---

Persona Realism vs Other Testing Approaches

Approach 1: Brand Search Testing

Method: “What is [Company Name]?”

What it measures: Brand awareness (do AI platforms know you exist?)

What it misses: Discoverability (can buyers who DON’T know you find you?)

Use case: Checking if AI has basic info about your company

---

Approach 2: Generic Keyword Testing

Method: “Best [category] tools”

What it measures: Category presence (do you show up in top 10 lists?)

What it misses: Persona-specific discovery (do buyers with YOUR specific problem find you?)

Use case: Benchmarking against obvious competitors

---

Approach 3: Persona Realism (Signal’s Approach)

Method: Realistic buyer personas describing problems in natural language

What it measures: True discoverability (do buyers who have your problem but don’t know your name discover you through AI?)

What it captures: Conversational discovery, platform variance, ghost influence, competitive positioning

Use case: Understanding how real buyers discover (or fail to discover) you through AI

---

Real Results: Persona Realism in Action

Case study: B2B SaaS client (project management for remote teams)

Brand search results:

• 80% visibility (AI mentioned them in 4/5 brand queries)
• Client conclusion: “Great AI visibility!”

Persona search results (5 buyer personas):

• Overwhelmed Remote Manager persona: Not mentioned (Perplexity recommended Notion, Twist, Loom, Asana, Linear, Basecamp instead)
• Ex-Agency PM persona: Mentioned rank #5 (Perplexity recommended Linear, Asana, Monday, ClickUp, then client)
• Documentation-Obsessed Lead persona: Not mentioned (Perplexity recommended Notion, Confluence, Coda, Slite)
• Zoom Fatigue VP persona: Not mentioned (Perplexity recommended Twist, Loom, Notion, GitLab, Basecamp)
• Compliance-Worried Startup persona: Not mentioned (Perplexity recommended Jira, Monday, ClickUp, Smartsheet)

Visibility gap: 80% brand awareness → 20% persona discoverability = 60 percentage point gap

What they fixed (based on Signal report):

• Added “async decision documentation” blog content (Persona 3 was asking for this)
• Added SOC 2 compliance FAQ (Persona 5 was asking for this)
• Changed homepage hero to emphasize “async-first” (Persona 4 was searching for this)

Results after 3 months:

• Persona visibility: 20% → 80% (4 out of 5 personas now find them)
• AI-driven trial signups: 3.2x increase
• Cost per trial: $45 → $18

KEY TAKEAWAY: Persona realism isn’t just theory. It drives measurable results. By closing the gap between brand visibility (80%) and persona discoverability (20% → 80%), this client tripled AI-driven signups and cut customer acquisition costs by 60%. The personas that couldn’t find you represent revenue you’re leaving on the table.

---

The Bottom Line

Persona realism = Testing AI visibility the way real buyers discover solutions

Why it matters:

• Brand searches measure awareness (people who already know you)
• Persona searches measure discoverability (people who DON’T know you yet)

The gap between brand visibility and persona visibility is where you’re losing opportunities.

One Signal report reveals this gap and shows you how to close it.

---

Related Reading

• Understanding Your Signal Report
• We Tested 5 Buyer Personas on Perplexity: Only 1 Found Us
• How AI Discovers Businesses
• Signal vs AI Visibility Tools Comparison

Want to test your persona visibility? Run a Signal report ($50) with realistic buyer personas and see which ones can actually discover you through AI. Brand searches lie. Personas tell the truth.

---

Quick Answers

What’s the difference between brand search and persona search?

Brand search: “What is [Company Name]?” - Tests if AI knows you exist Persona search: “I have [problem description]…” - Tests if buyers can discover you

Example:

• Brand: “What is Surmado Signal?” → AI describes it (brand awareness)
• Persona: “I need to test if ChatGPT recommends my product” → AI may or may not mention Signal (discoverability)

Why persona matters: 80% brand visibility doesn’t mean 80% discoverability. Real buyers don’t search for your brand. They describe their problem.

---

How many personas should I test?

Minimum: 3-5 personas (covers 70-80% of your ICP) Ideal: 5-8 personas (covers edge cases + core segments) Enterprise: 10+ personas (multiple buyer roles, use cases, industries)

Signal Essentials ($25): Tests 1 core persona Signal Pro ($50): Tests 3-5 personas + competitive landscape

---

Can I write my own personas?

Yes! Best personas come from customer interviews.

How to write realistic personas:

1. Interview 3-5 recent customers
2. Extract their exact pain language (“drowning in Slack”, “missing decisions”)
3. Write queries using THEIR words, not your marketing language
4. Test across ChatGPT, Claude, Perplexity, Gemini

Bad persona: “I need AI visibility software” Good persona: “I want to know if ChatGPT recommends my SaaS when people ask for solutions in my category, but manually testing is taking forever”

See DIY persona guide →

---

Why is persona realism better than keyword testing?

Keyword testing: “Best project management software” → Generic top 10 lists Persona testing: “My remote team keeps missing decisions in Slack across 8 timezones” → Specific solutions for specific problems

Why personas win:

• Real buyers describe problems, not categories
• AI recommendations change based on problem context
• Personas reveal ghost influence (AI describes you without naming you)
• Multi-turn conversations (not single queries)

Example: Keyword “best CRM” → Salesforce, HubSpot dominate. Persona “affordable CRM for 5-person startup” → Different recommendations.

---

What makes a persona “realistic”?

3 criteria:

1. Problem-first, not feature-first

   • “I need a tool with SSO and SCIM”
   • “I need to prove to IT that our project management tool is secure enough for enterprise”

2. Natural language, not jargon

   • “Seeking async-first collaboration platform”
   • “My team is spread across 8 time zones and people miss important updates when they’re asleep”

3. Multi-turn conversation

   • Not: Single query → done
   • Instead: Query → AI response → Follow-up → Clarification → Discovery

Signal tests realistic multi-turn conversations, not just single keyword queries.

---

How is persona testing different from traditional SEO keyword research?

Traditional SEO: Optimize for keywords Google shows in Search Console Persona testing: Test if AI discovers you when buyers describe problems conversationally

Key differences:

Traditional SEO	Persona Testing
”project management software"	"My remote team keeps missing decisions in Slack”
Google Search Console keywords	ChatGPT/Claude/Perplexity conversational queries
Ranking position (#3, #7)	Mention rank in AI responses (#2 of 6 mentioned)
Click-through rate	Discovery rate (did AI mention you at all?)

Both matter: SEO for Google search, persona testing for AI discovery.

---