Two ways to run the AI inside BoldReach: start instantly with our pre-paid credits — no API keys, no setup — or plug in your own provider's keys for zero markup. Switch between the two anytime, no migration required.
Most teams don't want to set up AI billing on day one — they want to use the product. Every BoldReach plan ships with a baseline allotment of AI credits that covers a typical team's voice notes, scoring, drafts, and forecasting. Hit the cap? Top up with metered credits at transparent per-task rates. Per-workspace budgets and 80% alerts come standard.
Once your usage grows beyond the included credits — or if your procurement team prefers — plug in your own AI provider keys. We route requests through your account, you pay the provider directly, we charge nothing on top. Useful when you have negotiated rates, regulated data flows, or just want full audit visibility against your own provider dashboard.
Different tasks need different model tiers. Voice transcription wants something fast and cheap. Lead scoring works fine on a mid-tier reasoning model. Long-form drafting wants the premium tier. BoldReach routes per task, picks the cheapest model that can handle it, and only escalates when needed — whether the bill goes to our credits or your keys.
Each task has a fallback chain. If your primary model rejects the request or times out, we route to the next configured option. If everything is down, the CRM degrades gracefully — manual scoring, manual drafting, voice notes still record locally and queue for transcription when the network returns.
BoldReach ships an AI usage dashboard that shows your spend per task, per workspace, per day — across credits and BYOK alike. Set per-workspace budgets. Get an alert before you hit them. Compare cost-per-task between paths to see which actually costs less for your specific traffic.
| Capability | BoldReach | Most CRMs |
|---|---|---|
| Day-one usage | AI credits included — no setup | Configure API keys before AI works |
| Long-term billing | Stay on credits or switch to BYOK | Vendor-bundled or vendor-locked |
| Markup on inference | BYOK = zero markup | Hidden in seat price |
| Cost visibility | Live per-task dashboard | Opaque |
| Fallback when AI is down | Multi-provider chain + manual mode | Feature breaks |