Our Fall release closes out a year of work built on a premise most medical AI ignores: the people inside a medical organization do very different jobs, and intelligence only gets used when it knows which one is yours.
Earlier this year I wrote about a shift in when intelligence arrives: from the quarterly report you pull to the continuous signal that gets pushed to you. That was the Spring story, and it still holds. Our Fall story is something quieter that we cared about all year: who the intelligence recognizes.
The medical organization is not one user. An MSL preparing for a call, a field medical leader who owes a director a monthly report, and a medical strategy lead concentrating engagement three weeks before a congress work in very different ways, on different deliverables, and answer for different things when someone pushes back. What they share is a scientific mission and the evidence underneath them. What they do with it, hour to hour, is not the same job.
Most medical AI ignores this. The industry has sold medical affairs a great deal of general-purpose intelligence and then asked every team to adapt itself to the tool. Answer a question when asked, and you have met the bar. That bar is now table stakes. The harder and far more useful problem is knowing which question matters before anyone asks it, and knowing what the person asking is on the hook for. That is a different kind of knowledge, and you cannot retrofit it onto a general assistant. You have to build to the job.
So we did. This release gives each team its own named solution, and each solution runs on one shared intelligence backbone. One platform underneath, but what each team meets is theirs. That structural point is the actual news, more than any single feature, so let me walk it the way our customers actually experience it: as the job each of them is accountable for.
An MSL is rarely at a desk and frequently without signal, and the work that matters is the scientific exchange, not the documentation around it. We started building for that job in March with the Sorcero Capture App, on the observation that MSLs were losing hours to writing things down instead of talking to physicians.
This release closes the loop between capturing and preparing. A stakeholder profile now assembles field notes, advisory board transcripts, and publication activity into a single record feed, and produces a pre-call brief from it: what this stakeholder cares about, what they have asked before, and where they stand on the clinical questions the team is tracking. The Capture App is now on iPhone as well as iPad, with offline capture and automatic enrichment as records sync, because a role spent moving between rooms and dead zones should not depend on a tablet and a good connection.
The accountability here is the call itself. Walking in prepared, with the last six months of context already assembled, is the difference between a scientific conversation and a reintroduction.
One level up, the artifact under scrutiny is the monthly report. A field medical lead does not get to say the intelligence is somewhere in the platform. They have to stand behind a document in front of a director.
Medical Insight Reports give each intelligence question a continuously refreshed summary and a clear view of what changed against the prior snapshot, and the output moves directly into a document, so the report starts from a draft rather than a blank page. The measure that matters to this person is time and defensibility. On earlier versions of this workflow customers cut weekly reporting time by over 72% and ad hoc analysis time by up to 90%. The point is not only the hours saved,but hours saved assembling the thing rather than thinking about what it means.
A medical affairs leader is accountable for a scientific narrative they cannot directly observe, and that narrative moves whether or not anyone is watching. In June we introduced the Therapy Resonance Profile and the Strategic Alignment Framework to move that accountability off quarter-old reporting and onto leading indicators drawn from the evidence as it changes. This release makes the underlying objects operational rather than illustrative.
Strategic Imperatives now function as working objects, not slideware. Stakeholder Alignment scores how an individual maps to each imperative, shows the trajectory of that position over time, and exposes the evidence behind both. A leader can ask which relationships are moving toward the strategy and which are drifting, and get an answer that traces to source rather than to a hunch.
Congress is where that question gets answered fastest, and it is where the platform now does the most. Congress Intelligence launched in March with centralized access to 3.5 million abstracts and posters from more than 4,300 conferences, retiring the sixty-day wait for a manual post-congress debrief. This release makes congress data a first-class citizen: abstracts and late-breaking data now exist as a dedicated record type, unified with KOL publication profiles and analyzable alongside every other source. The result is that planning assumptions, field captures, and abstract data sit in the same dataset. A single question can span what the team expected before an event, what it heard during it, and what it means for the strategy afterward.
If each team meets its own solution, the risk is three good experiences that do not talk to each other. Sorcero Wizard, entering beta for current customers by the end of the month, is the connective tissue. It is a conversational layer with native awareness of platform objects, including Insights, Collections, and Stakeholder profiles, and it holds context across an exchange so a line of inquiry can be narrowed, redirected, and drilled into without being restated.
That reads differently depending on who is asking, which is the whole idea. An MSL keeps narrowing on a stakeholder without re-identifying the subject. A field lead interrogates a monthly report and then asks what changed since the last one. A strategy lead moves from a congress abstract to its presenting author to that author's prior engagement history in a single thread. It is one capability that recognizes three jobs, not three features wearing a trench coat.
Underneath the team solutions is one backbone, and this is where a platform argument is either won or lost. Every insight traces to the source. Governance, role-based access, and audit hold across every team, which is the only reason AI is usable in a regulated medical organization at all, backed by independently audited SOC 2 Type II, GxP, 21 CFR Part 11, GDPR, and Data Privacy Framework compliance.
Two things about that backbone matter more than any feature. First, everything in this release is in production on real enterprise data today, not a roadmap commitment or a services engagement dressed as software. New customers see working intelligence on licensed external evidence in their first weeks, before their own pipelines are even connected. Second, because every team runs on the same connected evidence layer, insight compounds across the organization instead of fragmenting into tools that do not speak to each other. A medical organization can start with the team whose problem is most urgent and expand across teams, therapeutic areas, and data sources, with each solution composing onto the same backbone rather than bolting a new silo beside it. That is what turns a year of individual capabilities into something that gets more valuable as it touches more of the organization.
A system of record put the evidence in one place and made it traceable. A system of intelligence went further and told you what the evidence means, where you stand and why. The frontier now is a system of action, where the platform shows a team where to focus and helps them act on it. The Therapy Resonance Profile and the Strategic Alignment Framework are the start of this next step, partly through push intelligence and the engagement prioritization shipping today, and partly through a road we are honest about still building.
What I keep coming back to is that none of this originated in a strategy session. It came from the same conversations I opened with, where the gap between what the science said and what the team could act on was measured in quarters. The roadmap is set by what medical affairs teams keep asking for, and the measure of whether we got this right is plain: can a team act a quarter earlier than they could before. When they can, the rest of the mission follows, which is the right science reaching the right clinicians, and the right patients, sooner.
This release is simply the moment the premise stopped being a thesis and started being how our customers open the platform on a Tuesday morning.