Pecan AI
No-code predictive models for churn, LTV, and demand forecasts
Overview
Pecan turns plain-language business questions into deployed churn, lifetime-value and demand-forecast models without a data-science team, now framed as a conversational predictive AI agent. Connects to Salesforce, HubSpot, Snowflake, Databricks and more; pricing is by quote, with Starter listed at $950 a month on AWS Marketplace.
Pricing
Pricing shown for reference only. These figures reflect RECATOOLS research as of 1 Oct 2026 and may be out of date or incomplete. This is not financial or purchasing advice — always confirm the current price on the provider’s official website before making any decision.
- 2 prediction batches/month
- 500M rows
- In-app support
- Google/Microsoft SSO
- 10 prediction batches/month
- 2B rows storage
- Essential enablement support
- Custom batch volume
- 5B rows storage
- Pro enablement support
- Any SAML/OIDC SSO
Use cases
What you can produce with Pecan AI
- No-code predictive model builder (churn, LTV, propensity, demand)
- Conversational predictive AI agent
- Prediction monitoring alerts and model-health guidance
- Connectors for Salesforce, HubSpot, Snowflake, Databricks, BigQuery and more
- Explainable prediction-driver output
- ISO 27001 and SOC 2 Type II certified
- Listed on AWS Marketplace
Pecan automates the unglamorous parts of predictive modelling — data preparation, feature engineering, algorithm choice, deployment — so an analyst who knows the business can ship a churn or lifetime-value model without hiring data scientists, and it explains which factors drive each prediction. G2 reviewers rate it 4.7 out of 5 across 44 reviews. It plugs into Salesforce, HubSpot, Snowflake and most warehouses, which suits sales, marketing and growth teams predicting churn or propensity to buy. The catch is cost and opacity: Pecan removed prices from its website in 2026, and the only published figure is $950 a month for Starter on AWS Marketplace, a plan limited to two prediction batches a month. ISO 27001 and SOC 2 Type II certified, but with nothing specific for ASEAN buyers. Worth a demo for mid-sized teams with clear predictive questions and the budget; small teams will find it expensive.
What we found
Pecan builds predictive models without a data-science team: describe a business question — which customers will churn, what a new customer will be worth, how much stock you will need — and it prepares the data, engineers features, trains and deploys the model, then explains which factors drive each prediction. In 2026 it repackaged that as a "predictive AI agent" you converse with, alongside DemandForecast.ai, a forecasting product launched in 2025. It connects to Salesforce, HubSpot, Snowflake, Databricks, BigQuery, Redshift and the main SQL databases, plus mobile-attribution sources such as AppsFlyer, Adjust and Firebase, and says it needs no personal data to work.
Pecan was founded in 2018 by Zohar Bronfman, now chief executive, and Noam Brezis, both with doctorates from Tel Aviv University. It raised a $66 million Series C led by Insight Partners in 2022, bringing its total to $117.5 million; we found no newer round. It holds ISO 27001 certification and an annual SOC 2 Type II assessment.
Pricing has gone behind a sales conversation. Until at least April 2026 Pecan's website listed Starter at $760 a month and Team at $1,400 on annual commitments; by September those figures had been replaced with "talk to sales". The only published prices now are on AWS Marketplace, where Pecan lists Starter at $950 and Business at $1,750 for one-month contracts. Plans differ mainly in volume: Starter allows two prediction batches a month and 500 million rows of storage, Team ten batches and two billion rows, and Business custom batches, five billion rows and any SAML or OIDC single sign-on. Two batches a month limits how often a small team can refresh its predictions.
On G2, Pecan averages 4.7 out of 5 from 44 reviews, almost all five or four stars. Pecan describes its monitoring as real-time alerts on training and prediction progress, with help-centre guidance on overfitting, data leakage and model health. We found no regional presence or pricing for Southeast Asia, so ASEAN buyers should expect to deal with its US and Israeli teams.
Compiled by RECATOOLS from public sources: vendor documentation and pricing, code repositories and published reporting.
About this listing
This entry was compiled from publicly available data including Pecan AI's official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with Pecan AI unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to Pecan AI directly →
Spotted something out of date? Suggest an update →
Alternatives to Pecan AI
More in Business & Finance
