At $200, the credible route is buyer-led configuration of a hosted chatbot or FAQ widget, possibly with a short specialist review. The ceiling cannot fund conventional U.S.-agency discovery, custom retrieval, integration, production QA, launch, and support.
Keep the job deliberately small: answer about 20 approved public questions, refuse anything outside the source boundary, and direct the visitor to an existing contact method. The buyer owns content preparation and ongoing transcript review.
Related service: AI Chatbot Development services- 01Choose a free or low-tier hosted widget
- 02Prepare 20 short approved answers
- 03Configure identity, fallback, and contact link
- 04Test 20 normal and unsafe prompts
- 05Launch to one page and review every early conversation
SECTION 01
A worked $200 budget: hours, delivery, and reserve
Allocate 4 hours at an assumed $40 per hour: $160 for the work below and $40 held for corrections. The rate is an illustrative configuration assumption within Clutch's global $25–$49 AI range, not a U.S. agency average, a confirmed quote, or a Catapult rate. Market source checked September 20, 2026.
For comparison, the same 4 hours at Clutch's listed U.S. AI-company range of $50–$99 per hour would be $200–$396 in labor alone. A provider may need more hours or a minimum engagement. Four hours is a widget-configuration allowance. Content writing, retrieval engineering, authentication, and integrations do not fit.
| Work | Hours | At the assumed $40/hour |
|---|---|---|
| Check the supplied FAQ and account | 1 | $40 |
| Configure an existing widget and handoff | 1.5 | $60 |
| Check ten prepared questions and mobile display | 1 | $40 |
| Provide owner notes | 0.5 | $20 |
| Correction reserve—not extra features | — | $40 |
| Maximum planning envelope | — | $200 |
SECTION 02
Make the buyer—not the $200 provider—the implementation owner
The buyer creates and owns the hosted account, supplies 20 final answers, places the vendor script through an existing site editor, and names the human contact route. A specialist may spend a tightly capped session reviewing configuration and test results.
There is no custom back end, document ingestion pipeline, semantic search tuning, analytics warehouse, support desk integration, lead workflow, authentication, multilingual testing, or post-launch managed service in this ceiling.
| Included deliverables | Explicit exclusions |
|---|---|
| One buyer-owned hosted chatbot workspace | Custom code, retrieval pipeline, or model training |
| Approximately 20 approved FAQ answers | CRM, ticketing, calendar, payment, or account integration |
| Fallback and existing contact-link handoff | Sensitive or authenticated customer information |
| Basic test sheet and owner checklist | Content writing, live-agent staffing, or ongoing optimization |
SECTION 03
Worked $200 example: a one-page public FAQ widget
A local U.S. photography studio places a hosted widget on its services page. It answers 20 approved questions about session types, service area, general process, accessibility, and how to request a quote, while declining to invent availability or a final price.
When the answer is absent or the visitor asks for a person, the widget links to the existing inquiry form. The owner reviews all conversations during the first month and updates answers manually within the vendor product.
- 01
Owner finalizes 20 public answers and prohibited claims
- 02
Owner creates the hosted account and adds the site domain
- 03
Configuration uses the approved answers and fallback
- 04
Test set covers paraphrases, absent facts, pricing, urgency, and prompt attacks
- 05
Owner reviews early transcripts and disables the widget if unsafe behavior repeats
SECTION 04
How to check the results before launch
Use only public information that is safe to repeat. Do not collect payment details, medical information, secrets, or account credentials, and do not let the bot claim that a contact link is a staffed live channel.
Twenty-answer boundary
Every supported answer has approved wording, a source owner, and a review date.
Safe fallback
Unknown, sensitive, urgent, or personal-account questions receive a limitation statement and contact route.
Kill switch
The owner knows how to remove or disable the widget immediately.
- 01
Write one normal and one difficult prompt for each supported topic
- 02
Add unknown, conflicting, sensitive, urgent, and injection prompts
- 03
Mark answer, safe fallback, successful handoff, or failure
- 04
Launch only when the owner can review and disable the tool
SECTION 05
What you will pay to run it
The $200 figure is a setup ceiling, not a lifetime total. Hosted tools can charge for messages, seats, domains, AI features, removal of branding, data retention, or higher service levels.
| Cost item | Budget basis | Control |
|---|---|---|
| Hosted chatbot plan | Messages, domains, seats, features, and branding | Set a monthly cash limit and alert before upgrade |
| Owner content time | Writing, approving, and refreshing 20 answers | Assign a review date to every answer |
| Transcript review | Conversation volume and issue rate | Review every conversation initially |
| Future specialist help | Troubleshooting, migration, or new scope | Approve separately from the $200 setup |
SECTION 06
Questions to ask your developer
Do not evaluate providers on the number of features they say can fit. Evaluate whether they protect the tiny scope, explain vendor limits, and leave the owner able to operate or remove the widget.
SECTION 07
Frequently asked questions
Can a conventional U.S.-based agency deliver this AI chatbot scope for $200?
Ask for a quote against the exact configuration or pilot in this guide. The worked budget uses an assumed $40/hour, while Clutch lists U.S. AI-company rates at $50–$99/hour. At the higher rate, fewer hours fit. Neither benchmark guarantees that a provider can deliver the proposed scope for $200.
Are these Catapult package prices?
No. These examples help you plan a brief. Contact us for pricing based on your workflow, data, integrations, and support needs.
Can an AI chatbot under $200 use my private documents?
This scenario should not. Private-document access adds permissions, data handling, retrieval behavior, testing, and vendor diligence. Use only approved public answers at this ceiling and scope permissioned knowledge separately.
Does $200 include monthly chatbot fees?
No. The ceiling describes one-time setup or advisory work. The buyer pays any platform, message, model, domain, seat, storage, or premium-feature charge directly and remains responsible for owner time.
PRIMARY REFERENCES
Sources and further reading
These references cover the standards, platforms, or published prices discussed in the guide. Worked examples and checklists are our editorial guidance.
- AI development pricing: market data checked September 20, 2026Clutch
- AI Risk Management FrameworkNational Institute of Standards and Technology
- OWASP Top 10 for LLM and GenAI applicationsOWASP GenAI Security Project
EDITORIAL METHOD
About this guide
We use AI to assist with drafting and editing. Catapult AI Work is responsible for the published content. Examples illustrate possible approaches; they are not client case studies unless identified as such.
Budget examples are not Catapult package prices. Check linked provider pages for current fees and plan limits before making a purchase.
Read the editorial policy