What every customer gets, underneath Portfolio.
The Shared Platform is the customer record, three engines, Touchpoints and Mira. Portfolio runs on it today; Engagements will when it ships. A record written by one product is read by the other with no integration in between, because it is the same code and the same record.
Every product reads the same customer.
Before any engine runs, there has to be one account, one set of contacts, one contract and one ARR figure that everything agrees on. That foundation is platform, not product, which is why a health score and a renewal quote never disagree about what the customer is worth.
Every record carries its origin, so you can always tell what came from your CRM and what was created here. Your CRM stays the system of record for the account; the fields it does not hold (health, renewal posture, delivery status) live here and write back only where you enable it.
Every conversation with the customer, on the record.
Touchpoints is the sidebar group where the customer's meetings, email, tasks, commitments and survey answers land. Signals reads it to score the account and Renewal Desk reads it to build the offer, so both work from the same history.
The assistant built into the platform.
Mira is our AI assistant and the Shared Platform's AI layer. She sits inside Portfolio rather than beside it, so she can read one customer’s whole history (the calls, the tickets, the survey answers, the renewal) and answer questions about it.
Nothing touches a customer record, and nothing reaches a customer, until one of your team sends it.
Thirteen come with the product, in four kinds: renewal negotiation, account planning and communications, customer evidence, and meeting analysis. Write your own beside them in your own words, and enable, pause or edit any of them later.
Ranked by severity with the ARR attached. Each one carries a confidence percentage, the signal it came from, and a sentence on why it surfaced. Approve or dismiss, and both are recorded.
She writes a brief before a meeting and a summary after, and keeps each version, so you can see what she knew last quarter. A follow-up email she drafts sits there until a person sends it.
Ask her a question, share the conversation with a colleague, and search the transcript of every recorded meeting. She reads through your own role and field permissions, so she cannot summarize a record you would be refused if you opened it yourself.
Your team runs it, without asking us.
Everything below is administered inside your own workspace, and every change is attributed to the person who made it.
Your administrators create roles and set their capabilities. They set the health components, weights and thresholds, and every previous version is kept, so you can see what a score meant last quarter. They connect and disconnect sources, decide which topics Mira may address, and build the surveys and playbooks. Approvals, ownership changes and access grants are visible across the workspace, with the enforcement detail on the trust page.
Your team writes surveys, playbooks, onboarding programs, agreements and price books as templates, and each version is kept. A trigger or a clock fires them: a milestone completed, a deal won, a health threshold crossed, a renewal approaching, or the hour you set. Approval rules say which changes need a second pair of eyes. None of it needs scripting or a ticket to us.
Ferent works in any language, including right-to-left languages such as Arabic and Hebrew. Your administrator adds and sets up the languages your teams use. Call them clients instead of accounts and that holds in every language, including on fields you added. Every override is audited.
Built to WCAG 2.1 AA. We check contrast at the token level and publish the ratios, keep a visible focus ring on every interactive element, never carry meaning in color alone, and let text follow your browser settings. An independent accessibility audit and a VPAT are not done yet. Ask and we will tell you where that sits. If you hit a barrier, you get a fix or a date, in writing.
See it on your own data.
Bring an account you are worried about, a forecast you do not trust, or an onboarding that went badly. Forty-five minutes.