Video budgets vary by an order of magnitude between a 12,000 USD founder explainer and a 220,000 USD national campaign, and most production agency sites do a poor job of helping prospects place themselves on that spectrum. A prospect lands on a beautiful reel, watches three pieces, and still has the same questions: do you do my kind of work, what would something like the second video cost, and which director would lead it. SleekAI reads project posts tagged by format, industry, and budget band, then answers "do you make brand films for B2B SaaS" with specific projects and rough scope.

The bot pulls from wp_posts, ACF fields, and custom taxonomies via the data-source wizard. Tag projects by format (brand film, commercial, documentary, product launch), industry, and approximate budget band, and the bot filters answers on any combination. A prospect on the commercial cluster gets a different conversation than a prospect on the documentary cluster, each scoped by display conditions.

Pre-qualification is critical for video. The system prompt encodes the studio's minimum project size, the formats the team does and does not take, and rough budget bands per format. Off-fit prospects get a polite no; qualified prospects book a discovery call with the chat transcript attached. Conversation logs include model name, token usage, and page URL so the studio can audit which formats and industries actually convert and tune the prompt over time.

Video production budgets span an order of magnitude, and most studio sites do almost nothing to help prospects place themselves on the spectrum. A B2B SaaS marketing director with a 100,000 USD brand film budget watches three pieces on a reel and has no way to tell whether the studio behind them is a fit at that budget, a stretch, or out of reach. They submit a brief, wait, and either get a polite no after the discovery call or simply never hear back because the brief was too small to prioritize.

Both outcomes hurt the studio: the new business team spent a discovery call slot on a misfit, or the prospect walked away with a worse impression than if the site had let them place themselves correctly in chat. A bot that reads every project tagged by format and budget band closes that gap honestly. The prospect asks "is 100,000 USD realistic for a brand film like the one I just watched" and gets a real answer with comparable references, not a deflection to a contact form.

Format and industry filtering is the second gate. Most production sites group projects by client logo, not by format taxonomy, and prospects scroll reels trying to infer whether the studio handles their kind of work. The bot filters case studies on the same taxonomy the portfolio already uses, so a documentary prospect gets documentary references and a commercials prospect gets commercials references.

The transcript is the compounding payoff: reading a week of conversations tells the studio which formats inquire most, which budget bands recur, and which reel pieces actually pull weight. That feedback loop is hard to get from contact form submissions, where the prospect's real taste references and budget anchoring get compressed into a one-line subject before anyone reads it.