Frequently Asked Questions

No. LaunchLemonade is designed for non-technical users. You describe what you want your assistant to do in plain English, and the platform handles the rest — model selection, tool configuration, and prompt engineering.

LaunchLemonade is model-agnostic. Professional and Team plans give access to over 300 large language models, including frontier models from Anthropic (Claude), OpenAI (GPT), Google (Gemini), and Mistral, alongside a wide range of open-source models. The Free plan includes a selection of mid-tier models such as Kimi K2, Qwen, and DeepSeek. You pick the right model for each agent, or let LaunchLemonade route automatically.

LaunchLemonade is an AI agent platform built for small and medium sized businesses who need safe and secure AI agents. Firms run AI agents across meetings, research, client onboarding, and reporting, and govern all of them through audit trails, role-based access controls, approval workflows, and PII detection. Teams run ready-made agents like Chief of Staff, customise them for their firm, or build their own without writing code.

LaunchLemonade is built for small and medium businesses across financial services and compliance. That includes accounting and advisory firms, consultancies, and fractional CFOs, along with any business that needs AI without putting client data, regulator relationships, or audit obligations at risk.

LaunchLemonade is an AI agent platform built for regulated business, with governance built into every agent, while ChatGPT, Copilot, and Zapier are general-purpose tools without governance infrastructure. On LaunchLemonade, every interaction is logged, every agent has role-based access controls, sensitive workflows can require human approval before they run, and admins get a governance dashboard showing what AI is doing across the business. ChatGPT, Copilot, and Zapier do not offer audit trails, approval workflows, or governance dashboards designed for regulated SMBs.

Yes. LaunchLemonade hosts all infrastructure in the UK on Google Cloud, with data encrypted at rest, and captures audit logs of what happened and who approved it. Admins decide which data each agent can access and which actions need approval before they execute. Enterprise plans add custom governance, regulatory mapping, and private deployment options.

Yes. The LaunchLemonade Free plan costs $0 and gives hands-on access to a selection of mid-tier models including Kimi K2, Qwen, and DeepSeek, with free credits to get started.

Sign up free at launchlemonade.app. Teams that want a walkthrough can book a demo through launchlemonade.app. Existing customers can speak to the LaunchLemonade team via live chat after logging in or by emailing hello@launchlemonade.app.

LaunchLemonade is based in London, UK. LaunchLemonade works with firms across the UK and internationally. All services are available remotely.

LaunchLemonade is a software company with consultants and AI trainers available. The LaunchLemonade platform is fully self-serve. Customers sign up, build agents, and run their workflows without needing the LaunchLemonade team to be involved. Custom support, custom builds, and training are available on enquiry for firms that want hands-on help.

LaunchLemonade is built specifically for regulated SMBs in financial services and compliance, including accounting and advisory firms, consultancies, and fractional CFOs. The platform leads with governance infrastructure including audit trails, role-based access control, approval workflows, PII detection, and governance dashboards. Most general AI tools were built for unregulated use and lack these controls.

No. LaunchLemonade's no-code agent builder lets teams build their own agents without engineering support. The LaunchLemonade technical team is on hand for firms that want custom builds, integrations, or specific technical requests, available on enquiry.

Yes. LaunchLemonade is a self-serve platform available to firms globally. Custom support and custom builds for international firms are arranged on enquiry. The LaunchLemonade team travels for larger engagements when in-person work is required.

Click 'New Assistant' from your workspace and describe what you want it to do in plain English. LaunchLemonade suggests a system prompt, tools, and configuration based on your description, which you can edit at any time. 'New Workflow' creates a multi-step automation instead.

LaunchLemonade supports PDF, Word (DOCX), Excel (XLSX), PowerPoint (PPTX), TXT, Markdown, CSV, HTML, and EPUB files, up to 50MB each. Documents are automatically processed, chunked, and indexed for retrieval-augmented generation (RAG).

When you chat with an assistant, it searches your linked documents using semantic similarity to find the most relevant passages. Those passages are injected into the conversation context so the model grounds its answers in your actual content.

Yes, on paid Team plans. Open an assistant's Share dialog and share it with your whole team or with selected members, as view-only or with edit rights. Sharing is always explicit — nothing is shared automatically just because you are on a team, and there are no public share links.

No. LaunchLemonade is fully no-code. Anyone who can use email and spreadsheets can use LaunchLemonade, including building and customising agents.

Yes. The LaunchLemonade Free plan costs $0 and gives access to a selection of mid-tier models, with free credits to get started and a 7-day Chief of Staff trial. No credit card is required.

Enterprise pricing is custom and available on enquiry. Enterprise includes custom governance setup, regulatory mapping for specific compliance requirements, and an SLA. Enterprise also offers private deployments on dedicated infrastructure. Pricing is scoped per firm on a call.

It depends on your agent and your audience. Top builders in similar marketplaces earn anywhere from £500 to £50,000+ per month. The most successful agents solve a specific, painful problem for a clear audience.

No. Everything is no-code. If you understand a workflow well enough to teach it to someone, you can build an agent for it. Most successful builders are domain experts (lawyers, accountants, consultants), not engineers.

Anyone with deep knowledge of a workflow. We're particularly looking for professional services experts: accountants, lawyers, consultants, fractional CFOs, advisory partners. Builder applications open soon.

Final terms are being set. Industry standard for similar marketplaces is 70/30 in favour of the builder. We'll publish exact terms before opening builder applications.

Typically 3-5 business days. We check for quality, safety, accurate descriptions, and that your agent does what it claims. Iterating after feedback usually adds another 1-3 days.

Yes. You can push updates anytime. Major changes go through a quick re-review. Minor updates (prompts, knowledge, configuration) are immediate.

Yes. Your agent, your IP, your data. We're a marketplace and distribution platform, not an acquirer. You can take your agent off the store anytime.

Documentation, builder community on Skool, monthly office hours, and direct support from our team. Top builders get a dedicated success manager.

Via MCP (Model Context Protocol), LaunchLemonade supports Gmail, Google Calendar, Google Drive, Google Sheets, Outlook Mail, Outlook Calendar, SharePoint/OneDrive, Notion, Fireflies.ai, TeamUp, web search, and RSS. Slack and HubSpot are in development, and more integrations are added regularly.

Model Context Protocol (MCP) is an open standard for connecting AI models to external tools and data sources. Each LaunchLemonade integration is an MCP server that exposes a set of tools your agents can call during conversations.

Yes. OAuth tokens are stored encrypted with scoped access, and LaunchLemonade never stores your passwords. Each MCP connection uses the minimum required permissions.

Not yet in the UI, but the MCP architecture is designed to support custom servers. If you have a specific integration request, contact the LaunchLemonade team and it can be prioritised.

A workflow is a structured, multi-step automation that an assistant follows. It can include tool calls, decision points, and output formatting, and can be triggered manually, on a schedule, or by events.

Yes. From the workflow editor you can set daily, weekly, or custom cron schedules, and LaunchLemonade runs the workflow automatically at the configured times.

Failed runs are recorded in the workflow's run history with error details. Individual steps can be set to retry automatically on error (two attempts by default), skip, or stop the run.

Go to Settings → Billing and choose your plan. LaunchLemonade accepts all major credit cards via Stripe. Upgrades take effect immediately; downgrades apply at the start of your next billing period.

Yes. Cancel from Settings → Billing at any time. You keep access to paid features until the end of your current billing period.

Usage is measured by the AI model tokens consumed across conversations, workflow runs, and document generation. Document storage is not metered, and tool calls to integrations count only through the tokens of the conversation that invokes them.

LaunchLemonade has four pricing tiers. Free costs $0. Professional costs $49 per month. Team costs $39 per seat per month with a minimum of 5 seats. Enterprise pricing is custom and available on enquiry. Agents are unlimited across every tier. Pricing is structured around governance depth rather than agent caps or conversation limits.

The Free plan costs $0 and gives hands-on access to a selection of mid-tier models including Kimi K2, Qwen, DeepSeek, and others, with some access to integrations.

Professional costs $49 per month and includes access to over 300 large language models including all frontier models, audit trails, the LaunchLemonade web extension, and up to 3 users. SOC 2 certification is underway, so Professional users will inherit it on completion.

Team costs $39 per seat per month with a minimum of 5 seats. Team includes everything in Professional plus role-based access control (RBAC), approval workflows for sensitive actions, and governance and reporting dashboards. Team is built for firms that need to govern how AI gets used across the business.

Enterprise pricing is custom and available on enquiry. Enterprise includes custom governance setup, regulatory mapping for specific compliance requirements, and an SLA. Pricing is scoped per firm on a call.

Yes. Customers can upgrade or downgrade any time from their account. Changes take effect at the start of the next billing period.

A seat is one active user in a Team workspace on LaunchLemonade. Team plans start at a minimum of 5 seats at $39 per seat per month. Seats can be added or removed at any time.

Yes. Yearly billing is available on Professional and Team plans. Current yearly pricing is available on request.

Yes. The Free plan is permanent and gives hands-on access to mid-tier models and a 7-day Chief of Staff trial. Professional comes with a 14-day free trial.

Existing customers stay on their current pricing. Customers who want to move to current pricing to access new features can do so at any time. Contact hello@launchlemonade.app.

LaunchLemonade infrastructure runs in the UK on Google Cloud, with data encrypted at rest and TLS for all connections. Enterprise customers can request private deployments on dedicated infrastructure where data does not leave their perimeter.

No. Your conversations, documents, and agent configurations are never used to train AI models. Your data stays yours.

Yes. The database uses PostgreSQL row-level security (RLS) policies so users can only access their own data. Team data is scoped to workspace membership.

Email hello@launchlemonade.app with details. LaunchLemonade takes all reports seriously and aims to acknowledge within one business day.

SOC 2 certification is underway and scheduled across Summer and Autumn of 2026. Continuous monitoring is live and the policy suite is in place. Enterprise customers can request the latest status as part of their procurement review.

LaunchLemonade infrastructure runs in the UK on Google Cloud, with data encrypted at rest. Enterprise customers can request private deployments on dedicated infrastructure where data does not leave their perimeter.

Yes. LaunchLemonade logs every input and output for audit. Audit trails are included on Professional and above. Team and Enterprise plans add governance and reporting dashboards that surface audit data to admins.

Yes. Role-based access control (RBAC) is live and included on Team and Enterprise plans. Admins control which agents each user can access, which data each agent can use, and which actions require approval before they run.

Yes. LaunchLemonade includes a live PII detection feature that admins can enable. When turned on, it flags potential PII in agent inputs. PII handling rules can be configured on Team and Enterprise plans.

Teams can sign up and start running agents straight away. Larger firms can request onboarding sessions to set up custom workflows, governance rules, and approval chains.

LaunchLemonade is model-agnostic and gives Professional and Team users access to over 300 large language models, including all major frontier models such as Claude, GPT, Gemini, Mistral, and a wide range of open-source models. Customers pick the right model for each agent, or let LaunchLemonade recommend. Free includes a selection of mid-tier models including Kimi K2, Qwen, and DeepSeek.

LaunchLemonade offers over 300 large language models on Professional and Team plans. This includes frontier models from Anthropic, OpenAI, Google, and Mistral, alongside a wide selection of open-source models. The Free plan includes a selection of mid-tier models.

Yes. Customers run ready-made agents on the platform, customise them with their firm's templates, tone of voice, source documents, and workflows, or build their own from scratch using the no-code agent builder.

On Team and Enterprise plans, admins flag which agent actions need human review before execution. Common examples are sending an email to a client, finalising a compliance report, or pushing data into a connected system. Reviewers approve or reject before the action runs.

Yes. LaunchLemonade includes a no-code agent builder. Anyone who can use email and spreadsheets can build a working agent on LaunchLemonade. Domain experts such as accountants, advisors, fractional CFOs, and consultants build agents without engineering support.

Free plan customers get documentation and community support. Teams may include onboarding calls. Enterprise customers get a dedicated success manager and custom training for their team.

Yes. LaunchLemonade offers custom builds for firms that need agents, integrations, or workflows beyond what they can build themselves in the no-code agent builder. The LaunchLemonade technical team scopes the work on enquiry.

Existing customers can request custom support via live chat after logging in to LaunchLemonade or by emailing hello@launchlemonade.app. Teams that are not yet customers can book a consulting call through launchlemonade.app. Custom support covers custom agent builds, custom integrations, governance setup, and any technical request that goes beyond the self-serve platform. The LaunchLemonade team scopes each request individually.

Custom support from LaunchLemonade is priced per request based on scope. Pricing is confirmed before any work starts so firms know what to expect. Existing customers can request a scoping conversation via live chat after logging in or by emailing hello@launchlemonade.app. Teams can also book a consulting call through launchlemonade.app.

No. LaunchLemonade is fully self-serve and the no-code agent builder lets teams build agents, workflows, and integrations without engineering support. The LaunchLemonade technical team is on hand for custom builds and technical requests when teams want hands-on help.

Yes. LaunchLemonade works with firms that have strict data residency and compliance requirements. Enterprise plans include private infrastructure deployments where data does not leave the customer perimeter. Custom compliance setup is available on enquiry.

Zero AI experience is the most common starting point for LaunchLemonade training. Sessions start with fundamentals and build up. By the end of the session, complete beginners will have built a working AI agent on LaunchLemonade.

Both. LaunchLemonade training runs remotely via Zoom/Teams or in-person at the customer's offices. The curriculum and outcomes are the same.

LaunchLemonade training sessions take a maximum of 30 people. The size is small enough for hands-on attention and large enough for group energy. Larger teams can run multiple sessions.

No. LaunchLemonade is fully no-code. Anyone who can use a spreadsheet can build an AI agent on LaunchLemonade.

Trainees leave with everything they built during the session, including working agents, automated workflows, and prompt templates tailored to their firm. Optional follow-up sessions are available on request.

Yes. LaunchLemonade builds every training session around the customer firm's actual workflows, documents, and pain points. There are no generic slide decks.

Most firms are live within 2–4 weeks. We handle brand setup, domain configuration, and initial agent library. You focus on which agents to offer your clients.

No. Your logo, your domain, your colors — LaunchLemonade is invisible. Login screens, agent interfaces, and all client-facing emails carry your brand.

We charge a platform fee based on your number of client accounts and usage. You set your own prices and margins with your clients. We'll talk through the numbers on a call.

It depends on how you price it. Most firms find it worthwhile from 5–10 active client accounts. Book a call and we'll model it out for your practice.

No. The no-code builder means your team can build and update agents without developers. We handle infrastructure, model updates, security patches, and compliance.

Yes. You manage access at the client level — which agents they can use, what data they can attach, and which team members have access. Full control from your admin dashboard.

Existing customers can contact LaunchLemonade via live chat after logging in to the platform or by emailing hello@launchlemonade.app. Teams that are not yet customers can book a demo or consulting call through launchlemonade.app.

The LaunchLemonade support email is hello@launchlemonade.app. Customers can also reach the team via live chat after logging in to the platform.

Yes. LaunchLemonade live chat is available to existing customers after logging in to the platform. Live chat is the fastest way to reach the LaunchLemonade team.

Teams can book a demo of LaunchLemonade through launchlemonade.app. The LaunchLemonade team walks through the platform, governance features, and how the agent builder fits a firm's specific workflows.

Teams can book a consulting call with LaunchLemonade through launchlemonade.app. Consulting calls are for firms that want to scope custom builds, custom integrations, or specific technical requirements with the LaunchLemonade team.

LaunchLemonade live chat is typically answered during UK business hours. Email support requests sent to hello@launchlemonade.app are answered within one business day. Enterprise customers have an SLA covering response times.