Profonti runs in your cluster, on your database, with your data — under a perpetual license. You can't be locked out, forced to migrate, or charged escape fees. Your data is never combined with another customer's, never used to train any model. Privacy is the default, not an upsell tier.
Every feature. Every channel. Every answer cited.
Profonti turns the systems you already run on into a single, queryable repository — owned by you, isolated by architecture, cited on every answer.
This is the complete feature inventory. The first three blocks are the trust posture you should evaluate Profonti on. The rest is what your team uses day-to-day to get answers out.
Click a number and see the exact rows it came from — schema, query, timestamps. No hallucinations. No "where did this come from?" meetings. Audit-grade by construction.
"Active customer" means what your CFO says it means, not what an LLM guesses. Profonti carries your definitions through every answer, every digest, every alert — the same vocabulary across the company.
Some categories below — connectors, identity resolution, alerting — show up in every Profonti engagement. Others (mobile, slide export, anomaly detection) are turned on per-customer as the engagement matures. Everything ships in the same repository; nothing requires a separate product purchase.
One repository, every channel
- Browser
- Conversational analysis. Plain-language questions, cited answers, CSV export.
- Weekly digest sectioned by goal — the CEO board-meeting artifact. Operator-configured alerts when thresholds change.
- Slack / Teams
- The same questions, the same citations, in the channels your team already lives in.
- Mobile
- Field operators see the same answers desktop users see. Installable PWA — no app store, no separate account, no separate vendor.
- Slides
- Auto-generated PPTX from any question. Monday-board decks build themselves and stay in sync with the underlying data.
What you get
- Private repository in your cluster, your account — Profonti code on a perpetual license. Yours to run, ours to maintain.
- Source connectors for Postgres, MySQL, Snowflake, BigQuery, S3, HubSpot, Salesforce, Stripe, Cypher, Yoodli, REST APIs, and arbitrary file drops.
- Cross-source identity resolution — one canonical Customer / Franchise / Product across every system you connect.
- Conversational analysis with cited answers and per-question CSV export.
- Alerts + weekly digest by email, Slack, Teams, or SMS — sectioned by your stated goals.
- Definition lookup — your canonical meanings travel with every answer, every digest, every alert.
- Mobile PWA + PPTX export — same answers on every device, every deck.
- Anomaly detection — Profonti tells you when the numbers stop looking like they used to.
- Engineer-on-call — the team that built your engagement is the team that answers your alert.
Marketing words vs what we deliver
| What others promise | What Profonti delivers |
|---|---|
| "Generative AI for your data." | Generative answers, with the rows. |
| "Single source of truth." | Source citation on every number. |
| "Your data is secure." | Your data is yours — per-customer cluster, never combined. |
| "Custom dashboards." | No dashboards to maintain — ask the question fresh. |
| "Consultants will build it." | Engineers build it, run it, and answer when it breaks. |
Connectors, in slightly more detail
Out-of-the-box, Profonti can ingest from any combination of these sources:
- Operational databases — Postgres, MySQL, MSSQL, Snowflake, BigQuery, Redshift
- Cloud object storage — S3, GCS, Azure Blob (Parquet / CSV / JSON / NDJSON)
- SaaS systems — HubSpot, Salesforce, Stripe, Cypher Learning, Yoodli, and any REST API with a documented schema
- File drops — SFTP and S3 watch-folders for partner CSV / Parquet
- Bespoke — if the source exposes data, we'll write the connector. Custom connectors land in your repository and are yours under the perpetual license.
Source ingest is incremental and idempotent. Profonti tracks watermarks per source so re-runs are safe; identity resolution runs over the full graph each cycle, so cross-source links stay current as upstream systems change.
Why this list is the whole list
You're not seeing the same inventory dressed up with vendor jargon — "Cognitive Insights Engine", "AI-Powered Recommendation Tier", "Premium Compliance Module". Profonti has one product surface; the features here are all you get because they're all there is. Engagements differ in what data they ingest, what goals they answer, and which channels are turned on — not in which features were unlocked by a contract addendum.