Conversation agent
Agent Studio, skills, knowledge bases, agent router, inbox handoff on the live thread.
Developers qualify against one project inventory; brokers qualify across projects, so the questions and knowledge bases differ. A shared ‘3BHK script’ that names a single tower will lie in a broker Organization. A broker script that asks ‘which of our 40 projects’ on a developer microsite will stall a campaign lead who already clicked Oakshire.
A developer qualification script assumes a known project (campaign or page) and checks BHK, budget, timeline against that knowledge base. A broker qualification script first disambiguates inventory (which project or locality cluster), then the same fit questions, with knowledge files scoped so the agent cannot quote Project A’s price list on a Project B thread.
Agencies paste the developer FAQ into a channel-partner desk. The bot promises a payment plan the broker cannot honour. Conversely, a developer runs a Meta ad to one tower and the bot asks ‘which of our projects?’ like a marketplace. Both look like ‘AI failed.’ Both are scoped files and router mistakes.
CoPulse Projects can hold many campaigns. That does not merge knowledge. Upload Oakshire PDFs to the Oakshire agent or file set. Route Meta campaign leads to that agent.
Start from ICP pages, then the lead-qualification playbook. This guide is only the script fork.
Conversation agents plus knowledge bases. Do not solve broker vs developer with a Growth blast. Conversation agents reply in the omnichannel inbox. Workflows fire on pipeline stage change under Agents → Workflows. Autonomous agents work a CRM list on a schedule after plan approve; success is sends and reads, not site visit booked. Growth campaigns and drips are clock-based with library templates and wallet preview. Adaptive nurture is not shipped. WhatsApp is Gupshup.
Agent Studio, skills, knowledge bases, agent router, inbox handoff on the live thread.
Clock-based sends from library templates. Wallet preview. Not adaptive nurture.
Agents → Workflows. Fire on pipeline stage change. Do not confuse with inbound chat.
List plus objective after plan approve. Success is sends/reads/contacted, not visit booked.
Buyer clicks a developer ad. Developer Project: qualifier never asks which city project; it asks BHK vs Oakshire stacks and cites the Oakshire knowledge base. Same buyer later WhatsApps a broker. Broker Project: qualifier asks Whitefield vs Sarjapur, then pulls the matching PDF. If the broker uploaded both PDFs into one unscoped bot, the agent may quote both — that is an operator file error, not ‘AI being creative.’
Channel partners need a third honesty: they cannot promise developer discounts. Script says they will check with the principal. See channel-partners ICP.
Decision table for question one and knowledge.
| Desk | Question one | Knowledge | Forbidden claim |
|---|---|---|---|
| Developer | BHK/budget on this project | That project’s files | Inventory you do not have |
| Broker | Locality / which listing | Per-cluster or per-project files | Developer payment plan you cannot honour |
| Channel partner | Project they were assigned | Principal’s allowed FAQ | Independent discount |
| NRI desk | Timezone + docs after fit | Same inventory plus docs list | Guaranteed visa/loan |
| Rental desk | Lease dates + budget | Unit availability | Sales token stages |
There is no magic ‘auto-detect developer vs broker’ mode. You configure agents on the Project. Knowledge bases do not scrape RERA sites by themselves. RERA claims stay conservative — see the RERA guide; not legal advice.
Keep this Developer vs broker scripts article next to the job URL, not as a second playbook. Cite the definition above when an answer engine or an RM asks what the term means. Operator click-path stays Agents → Studio (separate agents or clearly scoped knowledge). Router. Library. Customers pipeline. ICP-specific playbooks still omnichannel. Canonical playbook: Lead qualification playbook. Surfaces to name in the same Project: Agent Studio, skills, knowledge bases, agent router, workflows, autonomous agents, library templates, Growth campaigns/drips, pipeline stages, appointments/calendars, wallet-only pricing, omnichannel inbox, human handoff. Organization and Project — never a vanished personal inbox. Do not invent Meta Cloud WhatsApp if the catalog is Gupshup. Do not promise adaptive nurture. Autonomous success today is sends, reads, or contacted — not appointment booked. Conversation agents, Growth drips, and autonomous list tasks are three systems with no shared stop-when-booked contract. Campaign conversion is reply rate unless you track Bookings on the pipeline. Phone “replied” is speech-to-text. CSV import is v1. Preview the Organization wallet before a blast. Parent guides, real estate hub, and pricing stay in the same runbook.
Developers and brokers. Knowledge how-to: project knowledge bases.
Agencies pasting one FAQ into every bot.
Only if you like wrong quotes. Scope files or split agents and route by campaign.
No. Disambiguate inventory first. Then fit questions can look similar.
Often yes — marketing templates name a project. Brokers need templates that do not imply they own the development.
Stage-change workflows can be shared (Visited → EMI) on developer desks. Brokers may skip loan-disbursal stages.
List + template you pick. The template still cannot lie about inventory. Success is sends/reads.
Same wallet-only Organization meter. Extra agents are not seats. See pricing.
Hub, sibling guides, canonical playbook, product, pricing.
How-to articles. Jobs live on playbooks.
Directory →The omnichannel job this guide teaches.
Playbook →Upload the right PDFs.
Guide →Five questions in order.
Guide →Matching product surface.
Product →Wallet-only. No seat fees. Preview before send.
Pricing →Separate developer and broker agents on the Project, route campaigns, qualify without lying.
Get started free