A variable in models adjusted during training.
One of the earliest neural network models.
Tailoring user experiences with AI-driven recommendations.
Training AI models on general datasets before fine-tuning.
AI techniques to forecast outcomes.
Wordflow's mid-tier subscription for established marketing teams. Includes 300 tracked prompts, all 7 AI platforms, all AI model integrations, Wordflow Insights, team workspace, and access to the Daily Tracking Add-on. Priced at $299/month or $239/month on annual billing.
Instructions given to AI to guide its outputs.
Wordflow's classification system for tracked prompts. Every prompt is categorised as Informational (the user is learning), Commercial (evaluating options), Problem-solving (addressing a challenge), or Transactional (close to a decision). Intent categorisation maps AI visibility onto the marketing funnel, showing where brand visibility breaks down by buyer journey stage.
Wordflow's method for generating AI visibility data. Rather than pulling from static datasets, Wordflow submits each tracked prompt directly to each AI platform and captures the real response produced at that moment — reflecting exactly what a user would see if they asked that question today. The basis for all Wordflow campaign metrics.
An AI-powered answer engine that responds to queries with cited, sourced answers rather than a list of links. Perplexity is widely used for research and product discovery, making it a high-intent channel for brand visibility. One of the seven AI platforms tracked in Wordflow AI Search Campaigns, and one of the three platforms covered in the Wordflow Free Scan.
The practice of designing and refining inputs to AI models to produce accurate, useful, or targeted outputs. In a GEO context, prompt engineering informs which tracking prompts a brand should monitor — the queries real users ask that determine whether a brand is cited in AI-generated answers. Distinct from a single prompt; prompt engineering is the discipline of crafting them systematically and at scale.