The profetiQ framework translates messy market signals into confident, auditable decisions.
We harmonize the entire internet signal stream in 23 languages and surface the actions worth taking. The quant framework weights every comment, event, and data point across six quarters of history so teams see where to move next.
How the AI Engine captures market signals and categorizes them into strategic imperatives
Car A has better features than Car B.
Every single signal through the web is captured and categorised by our fine tuned LLM model. Our AI algorithm, qualitiative analysis and sophisticated mathematical formulation gives a significance measure to each signal based on its certainty, rarity and recency.
We identify these signals world wide in 23 different languages (English, French, German, Spanish, Chinese Simplified, Chinese Traditional, Arabic, Russian, Japanese, Portuguese, Italian, Dutch, Swedish, Norwegian, Danish, Finnish, Polish, Turkish, Korean, Hindi, Indonesian, Malay, Thai)
We analyse the semantic information contained in every strategic signals and group them under strategic imperatives that would drive customer pull or requires structural operational shifts.
Brand Attractiveness and Brand Strength then calculated as a pull of 1000 index points for each make and model based on their aggregated significance measures.
The profetiQ system does not treat all comments, events, or data points equally. Every strength, weakness, opportunity, and threat is assessed by three ideas: how confident the source is, how unusual the topic is, and how recent it is. More recent signals count more; older ones steadily fade.
High-confidence, recent feedback keeps its weight near the top of the stack.
Niche topic plus moderate confidence; still meaningful because it is unusual.
Low confidence and stale timing rapidly down-weight the impact.
Time plays a different role depending on whether we are measuring attractiveness, strength, or market conditions. Brand Attractiveness is sensitive to short-term shifts and Brand Strength reflects structural proof points; both are expressed as 0-1000 factors, not 0-1. Market Force sits between them with a half-life tuned to macro change.
Short memory for sentiment pulses: a burst of buzz or backlash shows up quickly and fades quickly.
Long memory for structural proof points like engineering credibility, safety, and long-term reliability.
Macro conditions evolve by quarter, so regulatory, demand, and competitive currents fade more slowly.
Each comment or data point is tagged with a detailed strategy phrase and grouped into a strategy_high_level family such as Product, Technology, Customer Experience, or Market and Policy. Executives see familiar categories while keeping a line of sight back to each piece of evidence.
- Powertrain bets
- Software experience
- Safety stack
- Ownership journeys
- After-sales support
- Fleet promises
- Regulatory shifts
- Local incentives
- Competitive entry
- Press angles
- Influencer takes
- Community chatter
This highlights where the brand is over- or under-exposed - dense negative evidence in cyber security or a lack of recent positives in customer experience.
Within each market and segment, profetiQ compares competing alternatives and spreads them along a 0-1 relative scale to show position in the pack. Brand Attractiveness and Brand Strength are then expressed as 0-1000 factors for fast executive reads, while Market Force highlights tailwinds and headwinds in demand, regulation, and competition.
A 0.72 relative position means “towards the top of this cohort,” while 742 and 688 show absolute brand factors for quick executive reads.