AI lead scoring
Scores leads using their engagement, website activity and the history of past closed deals.
- It works from
- Lead engagement, website signals and historic closed deals.
- It gives you
- A lead score.
AI in Lumora
AI that takes the repetitive work out of daily operations
Lumora products include optional AI features that summarise, draft and suggest on the screen where the work happens. They run on Anthropic Claude or OpenAI using an API key, and a person reviews the result before records change.
- 01Optional features
- 02Anthropic Claude or OpenAI
- 03People confirm changes
- 04Works on your own records
How does AI work in Lumora products?
AI in Lumora products is a set of optional features, each tied to one task, such as summarising a lead or drafting meeting notes. The features run on Anthropic Claude or OpenAI using an API key. They read the records involved, suggest or draft a result, and a person reviews and confirms it before anything changes.
Lumora AcademyPro connects to Anthropic Claude or OpenAI with the academy's own API key. The defaults are Claude Opus 5.5 and GPT-5, and other current models can be chosen. Each AI feature can be switched off, and each person has a daily request limit.
Lumora Clockit uses Anthropic Claude, with a key the platform administrator enters and a model that can be changed in the admin settings. Without a key, transcripts and call statistics still work.
Principles
How does Lumora keep AI under your control?
The same rules apply to every AI capability in the products.
- 01
People stay in control
The Clockit scheduling assistant shows a proposal and acts only after you click Confirm. Academy grading suggestions and question drafts are saved only when a teacher chooses them.
- 02
Works on your own records
AI reads the records involved in the task: a lead's timeline, a lesson's text, your real free slots. Private and restricted lead notes are left out of lead summaries.
- 03
Runs on your API key
Lumora AcademyPro connects to Anthropic Claude or OpenAI with the academy's own key. Lumora Clockit uses an Anthropic key entered by the platform administrator.
- 04
Shows where answers come from
Meeting answers link to the moment in the recording, and CRM fields filled from a call come with the quote and timestamp. Search across meetings returns only what the person may open.
- 05
Confirmed actions are logged
Actions confirmed through the Clockit scheduling assistant are logged, so you can see what changed and when.
- 06
Tied to a real task
Each capability does one job on the screen where that job happens: the lead page, the grading form, the WhatsApp conversation, the meeting page.
Example
How does one AI suggestion become a change?
The Lumora Clockit scheduling assistant, moving a meeting.
Ask in plain words
A user types 'move my 3pm tomorrow to Thursday morning'.
The assistant checks real data
It looks up the user's own event types, free slots, bookings and focus plan through read-only lookups.
It proposes the change
The reply lists the free Thursday-morning slots and shows a reschedule card.
A person confirms
Nothing changes until the user clicks Confirm on the card.
The change runs and is logged
The booking moves, the invitee is emailed and the confirmed action is logged.
Capabilities
What can AI do in each Lumora product?
Every AI capability, grouped by product, with what it reads, what it produces and a before-and-after example from daily work.
6 capabilities · Customer relationship management
AI in BizPush
AI-drafted follow-ups
Drafts personalized follow-up messages based on call transcripts.
- It works from
- Call transcripts.
- It gives you
- A draft follow-up message.
Customer sentiment analysis
Reads the tone of customer emails, chat messages and support tickets.
- It works from
- Customer emails, chat messages and support tickets.
- It gives you
- The detected tone of the conversation.
Opportunity recommendations
Points out possible upsell and cross-sell opportunities.
Sales summaries
Writes short summaries of deal progress and pipeline risks.
- It gives you
- A summary of deal progress and pipeline risks.
Next-step suggestions
Suggests the next step a rep could take on a deal.
- It gives you
- A suggested next step.
6 capabilities · HR and people operations
AI in PeoplePro
Retention insights
Scores attrition risk to flag employees who may be disengaging.
- It gives you
- An attrition risk score.
Job description drafts
Drafts job descriptions based on the competencies of a role.
- It works from
- The role and its competencies.
- It gives you
- A draft job description.
Interview question assistant
Prepares interview guides from an applicant's resume and the competencies you are hiring for.
- It works from
- Applicant resume and target competencies.
- It gives you
- An interview guide.
Performance review summaries
Summarizes peer feedback, goals and manager notes into a draft review.
- It works from
- Peer feedback, goals and manager notes.
- It gives you
- A draft performance review.
Workforce capacity analytics
Forecasts team capacity and headcount needs by department.
HR policy assistant
Answers employee questions about leave policies, benefits and the company handbook.
- It works from
- An employee's question, with the company's leave policies, benefits information and handbook.
- It gives you
- An answer to the question.
6 capabilities · Employee learning and development
AI in GrowTeam
AI course drafts
Drafts modular courses from company SOPs, manuals, documents or recorded webinars.
- It works from
- Company SOPs, manuals, documents, or links to recorded webinars and videos.
- It gives you
- A draft course split into short modules, with quizzes.
AI quiz and scenario drafts
Drafts case studies, multiple-choice questions and practical knowledge checks.
- It gives you
- Draft questions, case studies and knowledge checks.
Personalized learning paths
Adjusts learning paths to each employee's role, experience and current skill gaps.
- It works from
- Employee role, past experience and skill gaps.
- It gives you
- A suggested learning path.
Knowledge-gap identification
Highlights skill gaps at team level so they can be addressed before they affect projects.
AI feedback on assignments
Gives written feedback on employees' written submissions and code tasks.
- It works from
- Written submissions and code tasks.
- It gives you
- Feedback on the submission.
Intervention alerts
Tells managers when an employee is struggling with a topic or falling behind schedule.
- It gives you
- An alert to the employee's manager.
5 capabilities · Institute management & student portal
AI in AcademyPro
New
Lead summary and next step
Reads a lead's details and shared timeline and writes a short status summary, the lead's interests and objections, and one concrete next step.
- It works from
- Lead name, stage, source, course of interest, centre, score, enquiry notes and up to 30 most recent shared timeline entries (private and restricted notes are left out).
- It gives you
- A summary of two or three sentences, lists of interests and objections, and a suggested next step, shown on the lead page.
BeforeA counsellor opens a lead with 14 calls, messages and notes over three weeks.
With LumoraA short summary with the lead's stated interests and objections and one suggested next step to take.
Runs on Anthropic Claude or OpenAI, with the academy's own API key (bring your own key). Default models: Claude Opus 5.5 (claude-opus-5-5) and OpenAI GPT-5 (gpt-5). Any current Claude model can be set instead, for example Claude Fable 5.1, Claude Opus 5, Claude Sonnet 5.5 or Claude Haiku 5.5; on OpenAI any chat model that supports JSON-schema replies.
New
WhatsApp reply drafts
Drafts a WhatsApp reply in the language the person last wrote in, leaving [placeholders] for facts it was not given. It is never sent automatically.
- It works from
- The last 20 WhatsApp messages with the person, the lead's course and stage, the counsellor's name and an optional instruction on what the reply should do.
- It gives you
- Editable reply text in the WhatsApp conversation panel.
BeforeA parent writes in Hindi asking about weekend batches and fees.
With LumoraA polite Hindi draft answering the question, with [fee amount] and [batch start date] left for the counsellor to fill in.
Runs on Anthropic Claude or OpenAI, with the academy's own API key (bring your own key). Default models: Claude Opus 5.5 (claude-opus-5-5) and OpenAI GPT-5 (gpt-5). Any current Claude model can be set instead, for example Claude Fable 5.1, Claude Opus 5, Claude Sonnet 5.5 or Claude Haiku 5.5; on OpenAI any chat model that supports JSON-schema replies.
New
Question bank drafts
Writes draft questions on a topic in the chosen types and difficulty, with answers and explanations. Drafts are validated like typed questions and saved only when a teacher picks them.
- It works from
- Topic, subject, difficulty, number of questions, question types and optional teacher notes.
- It gives you
- A list of draft questions (choice, true/false, numeric, fill-in, matching, short answer) with correct answers and explanations.
BeforeA teacher needs 10 medium questions on simple interest.
With LumoraTen drafts, including multiple-choice and numeric items with tolerances and worked explanations, ready to tick and save.
Runs on Anthropic Claude or OpenAI, with the academy's own API key (bring your own key). Default models: Claude Opus 5.5 (claude-opus-5-5) and OpenAI GPT-5 (gpt-5). Any current Claude model can be set instead, for example Claude Fable 5.1, Claude Opus 5, Claude Sonnet 5.5 or Claude Haiku 5.5; on OpenAI any chat model that supports JSON-schema replies.
New
Grading suggestions
Suggests marks and student feedback for work graded by hand, judged only against the instructions, rubric and model answers, and lists concerns such as possible copying.
- It works from
- Assessment title and instructions, rubric criteria, question prompts and model answers, and the student's answers or uploaded file text (long work is shortened).
- It gives you
- Suggested total, per-criterion or per-question marks with reasons, feedback to the student and concerns, shown beside the grading form. Nothing is saved automatically.
BeforeA teacher opens a 3-page written assignment with a 4-criterion rubric.
With LumoraSuggested points for each criterion with a one-line reason each, draft feedback, and a note to check one paragraph that may be copied.
Runs on Anthropic Claude or OpenAI, with the academy's own API key (bring your own key). Default models: Claude Opus 5.5 (claude-opus-5-5) and OpenAI GPT-5 (gpt-5). Any current Claude model can be set instead, for example Claude Fable 5.1, Claude Opus 5, Claude Sonnet 5.5 or Claude Haiku 5.5; on OpenAI any chat model that supports JSON-schema replies.
New
Lesson doubt tutor
Answers a student's questions about a lesson, grounded in that lesson's material, and explains the method instead of giving answers to graded work.
- It works from
- The lesson text, the course name, the last 12 messages of the conversation and the student's question.
- It gives you
- A chat answer in the lesson page (and through the mobile API), kept as a conversation per lesson.
BeforeAt 11 pm a student is stuck on a step in a lesson's worked example.
With LumoraA step-by-step explanation with a short worked example, saying when the answer goes beyond the lesson.
Runs on Anthropic Claude or OpenAI, with the academy's own API key (bring your own key). Default models: Claude Opus 5.5 (claude-opus-5-5) and OpenAI GPT-5 (gpt-5). Any current Claude model can be set instead, for example Claude Fable 5.1, Claude Opus 5, Claude Sonnet 5.5 or Claude Haiku 5.5; on OpenAI any chat model that supports JSON-schema replies.
5 capabilities · School management
AI in SchoolPro
Student performance insights
Spots learning plateaus across subjects and suggests where a teacher could step in.
- It gives you
- Suggested teacher interventions.
At-risk student identification
Flags students who may be at risk by combining attendance drops, missed homework and unusual test scores.
- It works from
- Attendance, homework and test scores.
- It gives you
- An at-risk alert.
Attendance pattern diagnostics
Shows patterns of chronic absence early.
- It works from
- Attendance records.
Academic recommendations
Suggests worksheets and practice exercises for each student's weaker areas.
- It gives you
- Suggested worksheets and practice exercises.
Parent message drafts
Drafts progress updates and fee reminders for parents.
- It gives you
- Draft messages to parents.
5 capabilities · Enterprise resource planning (ERP)
AI in Clarity
Cash flow forecasting
Projects future cash flow from historical receivables and planned spending.
- It works from
- Historical receivables and planned spending.
- It gives you
- A cash flow forecast.
Budget recommendations
Points out departments that are overspending and suggests budget adjustments based on run-rates.
- It works from
- Department spending and run-rates.
- It gives you
- Suggested budget adjustments.
Anomaly detection
Flags unusual invoice lines, possible duplicate payments and inventory variances.
- It works from
- Invoices, payments and inventory records.
- It gives you
- Flagged items.
Procurement recommendations
Suggests when to buy in bulk, using seasonal demand and vendor lead times.
- It works from
- Seasonal demand and vendor lead times.
- It gives you
- Suggested purchasing windows.
Plain-language business questions
Answers business questions typed in plain language.
- It works from
- A question typed in plain language.
BeforeA manager types: "Show me our highest margin product lines this quarter".
5 capabilities · Event marketplace and management
AI in SmartEvents
Event recommendations
Suggests venues, decor themes and entertainment packages that suit the event type and budget.
- It works from
- Event type and budget.
- It gives you
- Suggested venues, themes and packages.
Budget estimation
Estimates per-head costs for catering, AV and venue hire.
- It gives you
- A per-head cost estimate.
Vendor matching
Matches event requirements with well-rated local vendors who are free on the date.
- It works from
- Event requirements.
- It gives you
- Matching vendors.
Event planning assistant
Drafts run-of-show agendas, staffing checklists and guest communication scripts.
- It gives you
- Draft agendas, checklists and scripts.
Package suggestions
Suggests bundles of equipment and services that vendors can offer as discounted packages.
- It gives you
- Suggested packages.
5 capabilities · Translation management
AI in LingoMaster
Context-aware AI translation
Translates with industry terms, idioms and brand voice taken into account.
- It gives you
- A draft translation.
Terminology and glossary enforcement
Applies glossaries so product names, technical terms and company acronyms are translated consistently.
- It works from
- Source text and your glossaries.
Automated quality scoring (LQA)
Scores translations for accuracy, fluency and tone.
- It works from
- A translation.
- It gives you
- A quality score.
AI translation suggestions
Offers human translators alternative phrasings in a formal or casual register.
- It gives you
- Alternative phrasings.
AI-assisted review and proofing
Points out grammar issues, cultural missteps and formatting inconsistencies.
1 capabilities · Gamified learning and engagement
AI in Gamely
AI question writer
Drafts questions with answers, explanations, a skill tag and a difficulty level, which the teacher reviews before saving.
- It works from
- A topic (up to 150 characters) and grade, or pasted text up to 10,000 characters; question type or a mix; count up to 50; language.
- It gives you
- Ready-to-edit questions in any AI-supported question type; invalid ones are dropped automatically.
BeforeTopic: \"Photosynthesis\", grade 7, 10 questions, mixed types.
With LumoraTen draft questions (multiple choice, true/false, ordering, fill-in) with explanations, added to the question set for review.
Runs on The organization's own choice of 13 providers via laravel/ai (OpenAI, Anthropic, Gemini, Azure OpenAI, Bedrock, Mistral, Groq, xAI, DeepSeek, Cohere, OpenRouter, Ollama, OpenAI-compatible); bring your own key.
Before and after
Which manual work does AI remove?
Examples taken from the capabilities above.
Lead follow-up (Lumora AcademyPro)
Manual workA counsellor reads 14 calls, messages and notes from three weeks before calling a lead back.
With Lumora AIA short summary with the lead's interests, objections and one suggested next step.
Question banks (Lumora AcademyPro)
Manual workA teacher writes 10 medium questions on simple interest by hand.
With Lumora AITen draft questions with answers and worked explanations, ready to tick and save.
Grading written work (Lumora AcademyPro)
Manual workA teacher marks a 3-page assignment against a 4-criterion rubric from scratch.
With Lumora AISuggested points per criterion with a reason each, draft feedback and a note to check a paragraph that may be copied.
Rescheduling (Lumora Clockit)
Manual workChecking calendars by hand to move tomorrow's 3pm meeting.
With Lumora AIFree Thursday-morning slots and a reschedule card; one click confirms it and emails the invitee.
Meeting write-ups (Lumora Clockit)
Manual workWriting up a discovery call from memory after the next meeting has started.
With Lumora AINotes organised under the sales-discovery template, with action items and a recap email draft.
CRM updates after calls (Lumora Clockit)
Manual workReps type budget, timeline and decision maker into the CRM after each call, or skip it.
With Lumora AIProposed field values with the quote and timestamp they came from, ready to approve.
Questions about AI in Lumora
Anthropic Claude and OpenAI. Lumora AcademyPro works with either, with Claude Opus 5.5 and GPT-5 as defaults; Lumora Clockit uses Anthropic Claude, with the model set in the admin settings.
For Lumora AcademyPro, the academy connects its own Anthropic or OpenAI key. For Lumora Clockit, the platform administrator enters an Anthropic key.
No, not by default. The scheduling assistant acts only after you click Confirm, and grading suggestions and question drafts are saved only when a teacher picks them. CRM fields filled from calls wait for the host's approval unless the workspace turns that review off.
Yes. In Lumora AcademyPro each AI feature can be switched off and each person has a daily request limit. The rest of each product, such as AcademyPro lead scores or Clockit transcripts, works without an AI key.
No. The application sends the information a feature needs, such as a lead's timeline or a call transcript, to Anthropic or OpenAI through the provider's API and shows the result in the product.
No. The lesson doubt tutor answers questions from the lesson's own material and explains the method instead of giving answers to graded work.
The capability list on this page shows every AI feature by product. It is updated as AI capabilities are added to other products.
See the AI features on a task you know
Tell us which tasks take your team the most time. We will show where the AI features help, and where they do not.