Table of Contents
- Do These Platforms Really Offer Unlimited Usage?
- What's the Real Cost at Scale, Including Overage Fees?
- How Do Revenue Shares Impact Your Margins?
- Which Platform Offers the Most Predictable Scaling?
- Key Due Diligence Questions Before Platform Commitment
- The Path Forward for Scaling AI Coaching Businesses
- Related reading
Do not index
Do not index
Most AI coaching platforms market "unlimited" usage at their headline price, but the reality for creators scaling past a few hundred paying subscribers is more complex. Hidden fair-use policies, revenue shares, and metered credit systems can dramatically impact your margins as usage grows.
We've analyzed the actual cost structures of five major platforms to help you understand what happens when your AI coaching business scales beyond the initial launch phase.
Do These Platforms Really Offer Unlimited Usage?
The short answer: most don't, despite their marketing language.
Coachvox markets "unlimited chats" at $99/month, but their terms include an undisclosed fair-use threshold. Per Coachvox's support article checked September 2026, if usage becomes uneconomic for them, they can add quotas, throttle responses, queue requests, shorten context windows, switch to cheaper models, offer paid credits, or restrict features. They claim this affects fewer than 1% of creators, typically with advance notice.
Personify advertises "unlimited messages" on their Pro plan ($39/month or $29/month annually). However, no published fair-use rules or overage schedules were found, and their terms explicitly state that public marketing materials don't govern the actual application, leaving the contractual meaning of "unlimited" unclear (per Personify's site, checked September 2026).
Delphi has used "unlimited chats" in older marketing materials. Their current pricing shows training-word caps (5 million on Builder, 12 million on Scaler) but no published monthly message limits or overage rates for hosted conversations. Their API is rate-limited to 120 requests per 60 seconds (per Delphi's docs, checked September 2026).
The platforms that are transparent about limits use metered systems. Pickaxe operates on dollar-denominated AI credits: the Gold plan ($37/month) includes $15 in monthly credits, Pro ($147/month) includes $50. When credits run out, tools pause until you purchase more or enable auto-reload.
CustomGPT includes specific query credits: Standard ($99/month) provides 1,000 query credits, Premium ($499/month) provides 5,000. Once you hit the limit, the agent stops responding entirely. Additional blocks of 2,500 queries cost $375/month.
What's the Real Cost at Scale, Including Overage Fees?
The hidden expense most creators miss isn't the AI usage, it's the revenue share many platforms charge when you use their native billing systems.
Here's how the economics work at 500 subscribers sending roughly 30 messages per month each (15,000 total monthly messages):
Platform | Advertised Usage | Overage / Fair-Use Behavior | Revenue Share (Native Billing) |
Coachvox | "Unlimited" at $99/month | Fair-use threshold; can throttle, queue, or add quotas | 10% + 2.9% Stripe fee |
Personify | "Unlimited" at $39/month | No published fair-use policy | Not disclosed |
Delphi | No published message limits | No published overage policy | 15% (Builder/Scaler) |
Pickaxe | Credit-based: $50 included (Pro) | Tools pause when credits exhausted | 8-10% depending on plan |
CustomGPT | 5,000 queries (Premium) | Agent stops at limit | No revenue share disclosed |
For a creator generating $500K annually from 500 subscribers, Delphi's 15% revenue share equals $75,000 yearly, far exceeding any reasonable AI usage cost. Coachvox's 10% share plus fees would cost $65,000 annually at that revenue level.
Pickaxe's community estimates suggest typical coaching conversations cost roughly $0.0005-$0.005 per message, though long-context sessions can run higher. At 15,000 monthly messages, you're looking at $7.50-$75 in pure usage costs, well within their Pro plan's $50 monthly credit allowance.
CustomGPT's Premium plan would require significant add-on blocks for 15,000 monthly messages. If each message consumed one credit (actual usage varies), you'd need roughly $1,999/month in additional query blocks, making it prohibitively expensive for high-volume coaching applications.
How Do Revenue Shares Impact Your Margins?
Most platform comparisons focus on software subscription costs while ignoring the revenue share component, but at scale, the revenue share typically dwarfs the AI usage expenses. For a deeper breakdown of how those percentages get negotiated, see our piece on negotiating Delphi and Coachvox's revenue share.
Consider two scenarios with 400 paying subscribers at $1,500/year average subscription price ($600,000 annual recurring revenue):
Scenario A, platform with 15% revenue share: annual platform cost $90,000, AI usage (estimated) $4,000-$8,000, total platform expense $94,000-$98,000.
Scenario B, flat-rate platform with disclosed AI usage: annual platform license $2,364 (BuddyPro at $197/month), AI usage at ~$20/subscriber/month = $96,000, total platform expense $98,364.
While the total costs appear similar in this example, the revenue-share model consumes roughly 15% of gross revenue while the flat-rate model with disclosed usage costs runs at approximately 16% of gross revenue. The critical difference lies in predictability and profit retention: revenue-share models scale platform expenses directly with your success, while usage-based models let you keep margin improvements as your audience grows.
If your subscribers pay $1,500/year and cost $240/year in AI usage ($20/month), you keep $1,260 per subscriber after platform costs, an 84% margin that stays consistent as you scale.
Which Platform Offers the Most Predictable Scaling?
Predictable cost structures come from platforms that separate software licensing from usage expenses and avoid revenue shares altogether.
The most transparent approach is found on platforms like BuddyPro, where experts pay a flat annual license ($197/month when billed annually) plus disclosed AI usage costs of roughly $15-30 per subscriber per month, about 20% of a typical $1-2K/year subscription price. No revenue share, no hidden fair-use policies, no sudden throttling.
This model gives you full visibility into unit economics. The transparency allows creators to accurately forecast expenses and retain 100% of the profit after disclosed platform costs.
The metered platforms like Pickaxe offer reasonable transparency for lower-volume applications, though coaching conversations with long context windows can push costs above standard estimates. CustomGPT's query-credit system becomes expensive quickly for conversational AI coaching. For the full breakdown of what Coachvox actually charges, see Coachvox's pricing for coaches.
The "unlimited" platforms with undisclosed fair-use policies create the most uncertainty. You might operate smoothly for months, then face sudden throttling or additional charges as your usage crosses invisible thresholds.
Key Due Diligence Questions Before Platform Commitment
Before committing to any AI coaching platform, especially those with ambiguous "unlimited" offerings or revenue-sharing arrangements, ask their sales team these specific questions.
Fair-use policy clarity: request the exact monthly message threshold that triggers fair-use restrictions. If they won't provide specific numbers, ask for documented examples of when restrictions were applied to existing customers and what alternatives were offered.
Revenue share calculations: for platforms charging percentage fees, ask whether the percentage applies to gross subscription revenue, net revenue after refunds, or only revenue processed through their billing system. Clarify if there are minimum revenue thresholds or caps on monthly revenue-share payments.
Usage cost visibility: request detailed breakdowns of AI usage costs per message or conversation. Ask if pricing varies by model type, message length, or context window size. For credit-based systems, ask for real customer examples of credit consumption rates in coaching applications.
Migration and data portability: understand what happens to your subscriber data, conversation history, and custom training if you need to switch platforms. Ask about export formats and any restrictions on data portability during contract transitions.
Getting these answers in writing before signing helps avoid costly surprises as your coaching business scales.
The Path Forward for Scaling AI Coaching Businesses
The platform choice that makes sense at 50 subscribers often becomes problematic at 500. Revenue-share models that seem reasonable at $50K annual recurring revenue become significant profit drains at $500K.
Look for platforms that offer transparent usage-based pricing without revenue shares, clear documentation of any limits, and pricing models that let you retain more margin as you grow. The most successful AI coaching businesses we've seen prioritize platforms where they can predict costs accurately and keep the overwhelming majority of what their subscribers pay after disclosed platform expenses.
The economics become even more compelling as you optimize your coaching methodology. Creators who develop efficient conversation patterns, streamline their AI prompts, and build effective subscriber onboarding can often reduce per-subscriber AI costs over time while maintaining service quality. On transparent platforms, these efficiency gains translate directly to improved margins. On revenue-share platforms, the percentage fee remains constant regardless of how efficiently you operate.
Consider also the long-term trajectory of AI costs. As language models become more capable and cost-effective, platforms with transparent usage pricing allow you to benefit from these improvements. Revenue-share models, by contrast, maintain their percentage regardless of underlying AI cost reductions.
For creators ready to scale past the early stages, understanding these cost structures upfront prevents expensive platform migrations later. The most sustainable approach involves choosing platforms that align their success with yours through fair, transparent pricing rather than revenue extraction that grows with your business success. For a broader look at how these numbers compare across the market, see this overview of AI coaching platform pricing.
Related reading
Written by