Private equity deal screening
Categorize incoming company summaries by sector, business model or acquisition theme so the relevant deal team can review them.
STAYCHARTED AMT / PRIVATE EQUITY & VENTURE CAPITAL
Investment screening for PE and VC
Categorize deal flow by investment mandate, sector and diligence needs with StayCharted AI Model Trainer. Train on reviewed company summaries and keep your team in control.
FROM YOUR EXAMPLES
Subscription software for maintenance planning at manufacturing plants. Sells to operations teams.Investment theme → Industrial software
Where this fits in your work
Venture capital deal flow
Apply your fund’s themes to startup descriptions, organize inbound opportunities and identify records that need more information.
Due-diligence organization
Categorize extracted notes and questions by commercial, financial, legal or management topic for the appropriate reviewer.
How does AI fit into a PE or VC screening workflow?
StayCharted applies the categories your investment team has taught it to company summaries and notes. Start with one repeatable task: classify an investment theme, triage mandate fit or organize diligence questions. Keep those targets separate so each answer has a clear meaning.
| Workflow | What comes in | Category | Next step |
|---|---|---|---|
| Private equity sourcing | A company summary describes recurring maintenance services for industrial customers. | Industrial services | Send to the team covering that investment theme. |
| Venture capital intake | A startup description explains developer tools sold by subscription. | Developer infrastructure | Group with opportunities in the fund’s software theme. |
| Mandate-fit triage | A summary lacks enough information to establish the target customer or business model. | Needs more information | Ask an analyst to complete the record before screening. |
| Due-diligence notes | An analyst note asks how much revenue depends on the largest customer. | Commercial diligence | Organize the question for the commercial reviewer. |
From incoming opportunity to analyst review
- Prepare a company summary using information available at intake.
- Apply hard eligibility checks in your existing systems.
- Use your published model to suggest a screening category and confidence.
- Have the deal team review uncertain records and sample the remaining results, especially exclusions.
- Use approved corrections in the next training run.
PREPARE YOUR EXAMPLES
Start with a file your team has already labeled.
Start with a CSV or Excel file of company summaries and the screening category your team approved. Include business descriptions, customer segments, business models and other relevant information available at intake. Keep one company per row for company screening, or one note per row for diligence-topic classification. Use a separate model for each task.
Use one row per example with an approved answer. Keep the input representative of what the model will receive later, and agree on how to label ambiguous cases before training.
| Company summary | Approved investment theme |
|---|---|
| Software subscriptions for factory maintenance planning | Industrial software |
| A network of outpatient clinics providing specialist care | Healthcare services |
| Tools for software teams to test and deploy applications | Developer infrastructure |
From examples to reviewed results
Prepare
Resolve inconsistent labels and remove unnecessary sensitive information.
Train
Choose a model on your plan and train on representative examples.
Validate
Inspect held-out accuracy by category and the mistakes that remain.
Use and improve
Publish, fill a file or call the API, then review and feed corrections into the next version.
KEEP REVIEW IN THE WORKFLOW
Know where the model stops.
A screening category describes how a record fits your labels. Confidence is confidence in that category, not a probability of investment success or an estimate of returns. Review exclusions and consequential classifications, including confident ones. Your investment committee retains the decision; valuation, financial analysis and verification of claims remain separate work.
Evidence and implementation
Teach the mandate, not the outcome of the investment
Write a short definition of each category and resolve disagreements before training. Labels such as Within mandate, Outside mandate and Needs more information describe an initial screening task. Invested and Passed can hide valuation, timing, fund capacity and relationship considerations. A completed investment is not evidence that the company matched every screening criterion.
Keep hard thresholds in your existing rules
Use explicit checks for numerical requirements such as check size, revenue, ownership percentage or investment stage where it is a fixed field. Use classification for interpreting company descriptions and team-specific themes. A category prediction is not a reliable substitute for checking a financial threshold, calculating valuation or validating a reported figure.
Test on companies the model has not seen
Inspect accuracy and mistakes for each screening category. Keep repeated descriptions of the same company together when planning your evaluation, so near-identical records do not make performance look better than it is. Test a separate set of recent opportunities and review false exclusions as well as false matches. Do not include eventual investment decisions or later diligence findings in inputs that would be unavailable at intake.
Work from your deal pipeline
Export company summaries from your CRM, fill the file with suggested categories and confidence, and use your existing import process to bring results back. API access starts with Essentials for teams building an integration. StayCharted does not fetch company intelligence, extract pitch decks or provide a native connection to every investment CRM.
Handle confidential deal information deliberately
Use only information your team is authorized to process. Remove unnecessary founder contact details and confidential material before uploading. Personal-information checks help with supported patterns but do not detect every sensitive detail. Review workspace permissions, US hosting and the processing terms against your firm’s requirements. If an AI assistant reads information through a connection, its provider receives that information.
Update the model when your mandate changes
When fund strategy or category definitions change, review the labels, add representative examples and retrain. Check the new report before publishing a replacement. Feedback does not silently change the published model. Keep investment approval and the interpretation of diligence findings with your team.
File fills return category suggestions for your existing import process. API calls return results to your own software; your integration decides how to apply them. Plan allowances and workspace rate limits still apply.
Common questions
How can AI help with investment screening for private equity and venture capital?
StayCharted can learn screening categories from company summaries your team has reviewed, then suggest categories for new opportunities. Use it for investment themes, mandate-fit triage or diligence topics, with confidence to help prioritize review. It does not predict returns or make investment decisions.
What data do we need for a deal-screening model?
Provide a CSV or Excel file with relevant company descriptions and an approved category for each row. Define the categories first and use information available when a new opportunity arrives. Historical Invested or Passed labels may reflect factors beyond mandate fit; reviewed screening labels are a clearer starting point.
Can it check our fund’s investment criteria?
It can categorize narrative descriptions using examples of your criteria. Keep exact numerical thresholds and mandatory eligibility checks in explicit rules. Missing, outdated or unverified information still needs a person to resolve it.
Can StayCharted read pitch decks and financial statements?
For this text workflow, first extract the relevant text from PDFs, presentations or scanned documents into spreadsheet columns. StayCharted categorizes the supplied text; it does not perform document extraction, financial modeling or valuation.
Can it predict which companies will be successful investments?
No. A screening model predicts the category it learned from your examples. Its confidence is not a forecast of returns, business survival or investment success.
How should we measure a deal-screening classifier?
Check performance on companies not used in training, examine errors in every category and test recent opportunities. Pay attention to relevant companies incorrectly excluded. Repeated company descriptions and information learned after the screening decision can make a test misleading.
Can we use it with our investment CRM?
Use spreadsheet exports and imports, or build an API integration on Essentials or above. Your integration applies results to the CRM. Do not assume a native connector to your particular deal-management platform.
Which plan should an investment team start with?
Free includes text Classifier Models and one AI Classifier Model for testing reviewed summaries. Essentials adds API access and paid-plan workflows such as the Review Queue; Business adds Dedicated AI Models. Check current plan limits before processing a larger pipeline.
Explore the details
Try it on your own examples.
Free includes text Classifier Models and one AI Classifier Model. Essentials adds API access; Business adds Dedicated AI Models. Vision is an optional add-on on paid plans.