My Testing Process: How I Evaluated 8 Online Psychic Platforms Across 76 Sessions

Why I Built a Structured Testing Methodology

Most online psychic reviews are based on one or two sessions and heavily colored by the emotional state of the reviewer. Someone seeking comfort after a loss who has a moving reading will rate a platform very highly; someone skeptical who has a mediocre session will rate it poorly. Neither is useful data.

I wanted to build something more reliable: a multi-platform, multi-session research program with consistent methodology, documented predictions, and verified outcomes. Something closer to how consumer testing organizations evaluate products than how typical online reviews are produced.

The goal was a defensible answer to the question: which online psychic platform actually performs best, measured by the thing that matters most — accuracy?

Platform Selection

I selected the 8 platforms with the highest brand recognition and user volume in the online psychic category as of October 2025:

1. Kasamba
2. Keen
3. California Psychics
4. Purple Garden
5. Psychic Source
6. AskNow
7. Oranum
8. PathForward

All platforms were tested as new users. Fresh accounts were created on each platform to ensure I experienced the standard new-user onboarding, introductory pricing, and reader selection flow without any prior relationship with the platform.

Session Standards

Minimum Sessions Per Platform

I required a minimum of 5 sessions per platform before finalizing rankings. Most platforms received more (Keen: 16 sessions; Kasamba: 14). The minimum threshold ensures that no platform is ranked on outlier single-session data.

The Blind Testing Protocol

Every session followed the same opening protocol:

• Provide first name only
• State general topic area (e.g., “I have questions about a relationship situation”)
• Provide no additional context, backstory, or leading information
• Respond to reader statements with neutral acknowledgment only (“I hear that”), not confirmation

The blind protocol is standard for this type of testing because it maximizes the information value of specific claims. A reader who correctly identifies a named person, describes a situation, or predicts a specific outcome without being given context has demonstrated something meaningful. A reader reflecting back information provided is demonstrating nothing.

Multi-Format Testing

Where platforms offered multiple formats, I tested all of them: phone, chat, and video. This was important because format affects session dynamics. Kasamba is primarily chat; California Psychics and Keen are primarily phone; Psychic Source and Purple Garden offer video. Comparing platforms requires comparable format testing.

Prediction Logging System

During Sessions

For chat sessions: the chat log is verbatim. I saved every session immediately after completion.

For phone sessions: I maintained a written log during the call, noting specific claims in the reader’s words as closely as possible. Immediately after ending the call, I completed the log while memory was fresh.

What Qualifies as a “Specific Prediction”

For accuracy calculations, I counted only predictions that contained at least one of the following elements:

• A named person, place, or organization
• A timeframe (week, month, season, year)
• A describable physical characteristic
• A numerical value (amount, count, rating)
• A specific event or decision outcome

General statements (“a change is coming for you”, “you’ve been going through a difficult period”) were logged but excluded from accuracy calculations. They are genuinely too broad to evaluate.

Verification Protocol

30-Day Check

For each session, I set a calendar reminder for 30 days. At the check, I reviewed my prediction log and scored each specific claim as: Verified, Not Yet Verifiable, or Incorrect/Unverified.

90-Day Check

For longer-range predictions and “Not Yet Verifiable” items from the 30-day check, I completed a second verification at 90 days. Final accuracy rates reflect all verifications through the 90-day window.

Scoring Verified Predictions

A prediction was scored Verified if: the described outcome occurred, the named person or entity was correctly identified, or the described event happened within the stated timeframe (with a reasonable ±2 week window for timing claims).

Scoring Framework

Each platform received an overall score from four dimensions:

Accuracy carries the highest weight because it is the most important thing a psychic reading service delivers. Platform experience and pricing are meaningful but secondary — a beautiful app with inaccurate readers is not useful. Pricing fairness receives the lowest weight because most platforms are competitive and the gaps are modest compared to accuracy differences.

Customer Service Evaluation

I contacted every platform’s customer service at least twice:

• Once with a billing question (how charges appear, how to manage subscription)
• Once to test the satisfaction guarantee/refund process

Response time, resolution quality, and ease of process were documented and contributed to the Platform Experience score.

Independence and Disclosure

All testing was conducted independently. No platform provided complimentary access, advance notice of the research program, or any form of compensation during the testing period. Accounts were created as standard users; all introductory rates were those publicly available to any new user.

This site contains affiliate links. I earn a commission if you sign up through my links. This relationship was established after all testing was complete — the rankings were finalized before any affiliate conversations began. See my disclosure page for complete details.

See the Results

The output of this testing is the complete 2026 rankings and individual test result pages for all 8 platforms. The rankings represent my best current data on which best online psychic platforms deliver genuine value.

Frequently Asked Questions