Medical congress intelligence, explained.

A locked abstract in Hall A and a program lead two rooms away

From a title on the agenda to a changed plan: the path of a late-breaker

1. The meeting's program is readTitles and session times come out. The team picks which abstracts matter and assigns a person to each room.
2. The clock runsThe team knows the title and the time. Until the embargo lifts, the numbers are unknown and nothing is guessed.
3. The result is read and checkedThe abstract goes public. Someone reads it, checks each number, and sets it beside the company's own data.
4. A person says what it meansThe competitive intelligence lead rates the impact, writes the "so what for us", and decides who else is told.
5. The team actsA daily debrief goes out, the team's chart of rival drugs by stage moves, and someone changes a plan or decides not to.

Step 3 is the subject of this guide. Step 4 stays with people.

Embargo rules, slide photos, data mining and insider cases: what catches newcomers out

Abstract, late-breaker, embargo and competitive intelligence lead

Twelve terms are defined in the one-page summary.

What medical congress intelligence covers, from the abstract book to the morning debrief

In shortCongress intelligence follows what is presented at scientific meetings, and sets it beside what the same rivals say in press releases, filings and trial registries. It runs on eight kinds of material. Each one carries different detail, becomes public at a different time, and deserves a different level of trust.

The description of a medical congress by EFPIA, the European drug industry's trade association, explains the shape. The scientific program, which it says covers up to 95% of the whole, is managed by the medical society with no industry input: lectures, debates, abstracts and posters. Companies run the rest, the sponsored talks called symposia and the exhibition, as separate satellite activity.9 Coverage has to keep those two apart. A number from a selected abstract and a statement from a company's own symposium are different kinds of evidence.

Size varies by field. The European Crohn's and Colitis Organisation's statistics page shows 1,528 abstracts submitted in 2024, and participation that peaked at 8,034 in 2019.10 United European Gastroenterology reports more than 14,100 people from over 122 countries at its 2026 week.11 A competitive team usually tracks a small slice of any of them.

Two jobs often sit beside this one. Medical insights work listens to what doctors say. Pipeline tracking follows rivals all year through registries and company pages. Congress intelligence is the sprint: a few days on a fixed agenda in which a rival's story can change. A record in a trial registry, a public database of clinical trials such as ClinicalTrials.gov, feeds it: the US rule requires a primary completion date to be updated within 30 calendar days of reaching it and a recruitment status within 30 days of a change.15

When a congress abstract goes public: the embargo rules of eight real meetings

In shortAn embargo is a schedule, and each society writes its own. Regular abstracts can be public weeks before the talk. Late-breakers lift on the day at some meetings and earlier at others, and data beyond the abstract often waits for the session. A team that plans from the session time alone will be late at some meetings and early at others. Read each meeting's own policy, every year.

The table below gives each society's rule as it published it, for the meeting named. Every one of these changes from year to year, so treat the dates as examples of the pattern and check the current policy before you plan.

What we read: six societies' own documents in full. We could not open ASCO's or the American Heart Association's own pages, so those two rows come from a summary of them with quoted passages. Another publisher confirms ASCO's May 21 release and May 29 opening; the rest of ASCO's row, and all of AHA's, rest on the summary alone.1,16,2

The societies: ASCO, American Society of Clinical Oncology. AACR, American Association for Cancer Research. AHA, American Heart Association. EHA, European Hematology Association. ECE, European Congress of Endocrinology. AAN, American Academy of Neurology. WCN, World Congress of Nephrology. ESMO, European Society for Medical Oncology.

The days are our own arithmetic from each society's dates. A zero still has a time attached: at ASCO the late-breaker lifts at 8:00 am Eastern whatever hour the talk is, and at AHA it lifts at the talk.1,16

One meeting, five clocks: the EHA 2026 embargo schedule

Three things that hold across most embargo policies

That last point matters for a public company, which may have to disclose results that move its stock. The World Congress of Nephrology may exempt one, case by case, when legal counsel advises that disclosure is required, and AAN asks authors in that position to notify it.18,13

AHA's 2026 guidelines, in the summary we read, say SEC-required announcements must be approved by AHA before release.16 A rival's short announcement in the spring can be followed by the full data at a meeting in the fall, and a team should have both on its calendar.

Who an abstract embargo binds, what a broken one costs, and how a schedule can move

In shortPolicies speak first to authors, sponsors and journalists. Some reach further: ASCO adds "others", and EHA's press form says third parties are obliged. The penalties fall mostly on the abstract and its authors. Ask your own counsel how each meeting's text applies to you, and plan as if circulating embargoed data inside your company were a problem, because it can be.

The words differ. ASCO, as quoted in the summary we read, says that before publication "the author, coauthors, sponsor of the research, journalists, and others must not make the information public".1 ESMO's confidentiality line names "the author, co-authors, institute(s), research sponsor(s) and ESMO".4

EHA's form states that the policy covers abstracts "regardless of the source from which the information is obtained", that third parties are obliged to abide by it, and that if an embargo is broken "both the third party and the person involved will be held responsible and liable".12 That form is written for the media, so read it for its direction, and take advice on its reach.

The World Congress of Nephrology goes furthest on money: before presentation, the presenting author, co-authors, research sponsors and journalists may not "use the information for trading purposes or provide it to others who can use it for trading purposes".18 That one line joins the congress's rule to securities law, the subject of section 06.

What breaking an embargo can cost, in the societies' own words

The morning a schedule moved: ASCO's early release in May 2026

A published hour can change. ASCO reported that it inadvertently released details of six embargoed studies through a podcast at 5:00 am Eastern on May 21, 2026, twelve hours before the planned 5:00 pm release, and that it then published those abstracts early.1 We read this in a summary of ASCO's pages, and we found no other publisher's account of it.

For a coverage team the lesson is practical. A clock built from last month's schedule would have kept six public abstracts locked after ASCO had released them, and a team that read the podcast's details before the society released the abstracts would have had a harder question for counsel. Someone has to watch the society's own notices during the meeting.

What a careful team does about embargoed material

Photos, recordings and data mining: what congress terms allow a company to do

In shortCongress content has an owner. Each congress's own rules govern what you may capture and what you may mine, and getting permission is your job. Societies commonly hold the abstract's copyright, restrict recording of sessions and posters, and may require permission for commercial mining of even their published abstracts. Read those terms before anyone sends a phone into the hall or points software at an abstract book.

Three separate permissions tend to get blurred: whether you may read something, whether you may copy it, and whether you may process it at scale for a commercial purpose. An abstract that is free to read on a society's website can still be restricted on the second and third.

Competitive intelligence has an ethics code of its own. SCIP, the field's professional society, lists "to accurately disclose all relevant information, including one's identity and organization, prior to all interviews" in its code. We take that wording from two articles that quote it, because we could not read SCIP's own page.20,21 At an exhibit stand, that means wearing the badge and saying who you work for.

US trade secret law defines improper means to include "misrepresentation", while the EU's directive treats observation of a product or object made available to the public as lawful.22,23

Hearing trial results at a congress, and when a remark at an exhibit stand becomes a legal problem

In shortHearing a result that a society has made public is ordinary. The hard cases are a sentence at a stand or a corridor aside that may not be public anywhere. The published US actions we read concern insiders who traded ahead of an announcement. The working test for a congress team is one question: can we point to where this is public? If not, it goes to counsel and into no report.

As general information, the US and EU rules run like this. In the US, trading while aware of material nonpublic information, meaning facts an investor would want that are not yet public, breaches the antifraud rules when it also breaches a duty, and Rule 10b5-2 says one such duty arises "whenever a person agrees to maintain information in confidence".24

Regulation FD (Fair Disclosure) is a separate rule that binds the companies issuing securities: it covers selective disclosure to market professionals and holders likely to trade, and it treats a Form 8-K, a report filed with the SEC, a wire press release or an open webcast as public.25 It does not make a rival's analyst into a regulated party, and it does tell you what public looks like.

In the EU, the Market Abuse Regulation defines inside information as precise information that has not been made public and would be likely to have a significant effect on prices. Article 14 bars insider dealing, recommending it, and unlawful disclosure, and Article 10 treats disclosure as unlawful unless made in the normal exercise of employment, a profession or duties.26 A competitor's listed shares can fall under those bans, so a remark passed around a program team has a legal context even if no one there buys a share.

The third case is the one that mentions a conference, and the record shows why it matters. The investigator learned of the planned announcement on June 16, 2022, more than four months ahead of it, and bought after learning. The release says the stock rose 60% after the announcement.

Without admitting or denying the allegations, the defendant consented to a judgment, subject to court approval, with disgorgement, meaning giving up the profits, and an equal civil penalty, each of $1,520,455.6 An SEC enforcement official said in 2014 that the agency "will continue to vigorously investigate and prosecute those who illegally trade on this information".8

Testing a field note: can we point to where this is public?

Your company's own employees add a further layer. Someone who holds confidential information from a consulting role, an advisory board or a past employer may have a duty that outlasts the job. Your compliance team should ask about this when it builds the coverage plan, well before anyone is on a plane.

How far to trust a congress number: what abstracts leave out and how often papers differ

In shortA congress abstract is a short, early document chosen by a committee. Many never become full papers: about 37% of abstracts in one large review reached full publication, with an estimated 69% of randomized and controlled trials published by ten years. When papers do appear, results can differ. A brief should say what kind of document a number came from.

Publication is linked to the results themselves. The same review found that results labeled positive were more likely to reach full publication (risk ratio 1.31), and so were oral presentations (1.46) and randomized designs (1.51).27 A team that treats every congress number as a future paper will overrate the good news. Journals do accept a full report after an abstract or a poster at a meeting, so a presentation does not stand in the way of a paper.29

How congress abstracts and later papers disagree

Neither series comes from drug trials in the fields a company usually tracks, and we found no such figure for late-stage drug trials specifically. The plastic surgery authors defined a change as a shift in direction or more than 10% in a result, and they recommend marking abstracts as preliminary.28 That is good advice for an internal brief too: name the document type beside the number.

Words that lean: spin in trial reports

Reading the words takes care as well. A 2026 review of 71 randomized myeloma trials found spin, meaning reporting that could mislead by emphasizing benefit or playing down unfavorable findings, in 25 of them, 35.2%. It was most often optimistic language and emphasis on secondary or subgroup findings, and it was tied to trials that missed their primary endpoint.31 One specialty, one study. For a congress team, the takeaway is to pull the primary endpoint first, then read what the abstract's conclusion says about it.

Setting a rival's result beside your own trial: the arithmetic of "not head-to-head"

In shortTwo numbers from two trials share no randomization, so a difference between them may come from the trials themselves. Line up the same endpoint at the same time point, subtract each trial's own placebo, print the populations, and say what you could not check. In the example, a 1.0-point gap sits far inside the rival trial's own 95% confidence interval, the range the true effect is likely to lie in.

Inside one randomized trial, patients in both arms came from the same recruited group, so the gap between arms can be credited to the treatment.

Across trials, patients were enrolled separately under separate rules. Setting the treated arms side by side is prone to bias and offers no advantage over observational data, as a founding paper on the method put it. Comparing each drug's gap over its own placebo preserves each trial's randomization, and those authors still warn of limited strength of inference.32

Health technology assessment bodies, public bodies that review what a medicine adds over current care, use a range of formal indirect comparison methods.33 A competitive brief is only a first read, and it needs to say so.

The fictional comparison, worked by hand

All numbers are the guide's fictional case, and our own arithmetic from them.

What has to match before a comparison means anything

The nearest regulatory text is about a stricter use: an arm from another trial used as the comparison group, called an external control.

FDA's February 2023 draft guidance, still a draft, says such use "would only be appropriate" when comparability exists on "participant eligibility criteria, treatment administration, patterns of care (e.g., location of treatment sites), recording of concomitant medications", and how outcomes were assessed.34 It also calls data collected in different time periods a particular concern when care for a disease changes over time. A competitive table goes to no regulator, and the list still works as a reading guide.

Words matter too. "47% with an inadequate response to at least one advanced therapy" and "52% who had failed at least one advanced therapy" sound alike, and a trial's definitions may differ. "A more refractory population", which here points to the larger share who had failed an advanced therapy, is therefore a judgment, even when the numbers are right. A careful brief prints both phrases and lets the reader see the gap.

Choosing 46 abstracts from 2,318: the arithmetic of a coverage plan with three people

In shortCoverage is a selection problem. Start from the questions your program needs answered, turn them into a tracked list, then place people on a grid of days and rooms. Every blank cell is a decision, and it is better made on purpose. In the example, the team tracks about 2 in 100 abstracts, and one session on Day 3 has nobody in it.

The Medical Affairs Professional Society describes a useful way to start for insight work: teams define broad key insight topics, then draft narrower key insight questions to guide field interactions.35 A congress plan borrows the idea. "Will the rival's maintenance data arrive before our filing?" tells an analyst what to read and where to sit. "Cover immunology" tells them nothing.

Ranking sessions by what a late read would cost your program

A plan should also record what your own people are doing. In the fictional example the program lead is presenting her own data in Hall C at 11:30 am. She cannot cover anything then, and she should not be the person reading rival numbers in her own session's prep. Good plans also name a reviewer for field notes, reachable during the meeting's hours and its time zone.

Eight ways congress coverage goes wrong, and which ones reach a decision

In shortThe expensive errors are ordinary: the wrong clock, a slide number in a brief ahead of its release, an abstract's number carried forward as final, a subgroup used as the headline, a comparison across different endpoints. No public study counts how often these happen in company teams, so measure your own on a past congress.

Forwarding is one more habit to watch. A brief that was accurate and careful at 9:00 am can lose its caveats by the time a brand team forwards one line of it at noon. Anything an internal brief says about a rival's medicine may later shape something said outside the company, and anything said outside goes through medical, legal and regulatory review first. Keep the label attached to the number.

Why the hours after a late-breaker matter: investor calls, stock moves and a missing benchmark

In shortTrial results can move markets, and a sponsor may schedule an investor call the same day. Company leaders, investors and doctors start asking questions at once. That is why teams care about hours. Nobody publishes what a late brief costs, or the size of this market, so measure your own time from public to told.

A study of 24 investigational compounds at large US-listed companies, from January 2011 to May 2013, found a median day-of-announcement abnormal return, meaning the share move after adjusting for the wider market, of +0.8% for positive results and −2.0% for negative ones. After two days the figures were +0.4% and −1.7%.36

The study is old, small and limited to large firms, and it shows a real effect: results are treated as news on the day they come out. Regulation FD's list of public channels includes open webcasts and wire press releases, which is why the call and the release are often on the calendar.25

What nobody has published about congress coverage

The reasons to move fast are mostly practical. A program lead who hears a number before a meeting can ask better questions at it. A medical affairs team can prepare for what doctors will ask that day. A strategy lead can read an investor call knowing what the abstract said. And a team that speaks first with a checked number leaves less room for a wrong one in a hallway.

What AI can and cannot do with a congress abstract, and which rules reach this work

In shortIn the studies we read, language models handled simple stated numbers well and complex ones poorly. Every extracted number should open on its source, and a person should decide what it means. Few AI rules reach internal competitive summaries. FDA's draft and draft EU Annex 22 concern other work, and the EU AI Act's dates moved in 2026.

The authors of that proof-of-concept study say to read the results with caution.38 A separate evaluation of seven models found them close on simple binary outcomes and poor when "tallying the results requires inference".39

A third study, on clinical note summaries, reported a 1.47% hallucination rate, meaning made-up content, and a 3.45% omission rate across 12,999 sentences; all its authors worked for the company behind the method.40 None of these tested congress abstracts, and we found no public test of congress work. A congress abstract's response rates fall on the easier side, and a mean change in a disease score falls on the harder one. Models improve quickly, so measure the one you are offered on your own abstracts.

The AI rules people ask about, and how far each one reaches here

The tools themselves deserve the same care as the data. Ask where the abstract text goes, whether a provider trains on it, and who can see a held note. No regulator has written a rule about a competitive summary that we could find, which leaves a company's own policy in charge.

Who covers a medical congress today, and how one late-breaker takes nine and a half hours by hand

In shortCoverage is usually a few people across rooms, a planner or a spreadsheet, and sometimes an agency writing nightly notes. By our illustration, most of the time between a result and the program team goes on the evening write-up. Your own split will differ, and it is quick to measure.

From CI lead to compliance counsel: five seats that cover a meeting

The 9.5 by-hand hours of one late-breaker, longest step first

A planner, an agency, in-house analysts or software that reads at the lift

LB-07 in Botminds, from a locked title at 8:39 am to a decision at 8:52 am

In shortBefore the lift, the system knows a title, a time, a room and who covers it. At 8:40 am the abstract is read, 14 fields are pulled with citations, the source is confirmed public, and it is compared on the same endpoint. Three facts-only lines reach one pre-approved channel at 8:43 am. The rating, the meaning and any other audience wait for Dana Okafor, the CI lead.

1

The Vienna agenda as a day-by-room grid, before the lift

Dana's home is the congress itself. Wednesday, October 7 is laid out by hall and hour: nine sessions with tracked abstracts, each chip marked as a competitor, your data, embargoed or academic, with the initials of who covers it. The red line is the time now. On the right, the embargo watch counts down to LB-07 in Hall A, with three more late-breakers to come.

2

What the system holds while an abstract is under embargo

Open the event before 8:40 am and the abstract's pane is locked. The record holds four things: the published title, the session, who covers it, and where the summary will go. The screen says the abstract is under embargo until 8:40 am and that nothing about it can be used or shared. The event is a placeholder with a clock on it, and it was added from the title alone.

3

At 8:40 am the cell turns red and the abstract is read

The chip flips from locked to high-impact competitor. Sam Patel is in Hall A, listening, and he has nothing to type yet. The other chips stay locked on their own clocks: 2:15 pm and 2:35 pm today, and 10:20 am tomorrow.

4

Read, extracted, checked, compared and summarized, with a time on each step

The steps run in view, each with its minute. One line comes from the Compliance Guard, the agent that checks each source: "congress abstract · public since 8:40 am". That is the system's answer to the question from section 06, can we point to where this is public? At a real congress, that answer comes from the society's published schedule, so your team should check it.

5

Three facts-only lines in the program channel at 8:43 am

Dr. Maya Chen reads three lines between her own sessions. HWL-217 met its Phase 3 induction endpoint, 27.4% vs 9.8% at week 12. Veltrimab's VLT-302 reached 24.9% vs 8.3%. Marker-positive patients, 38% of the trial, reached 33.2% vs 9.1%. Each line has a chip that opens its source. Read line two closely: "a more refractory population" is a judgment resting on 52% vs 47%, and the two trials may define that history differently (section 08). A team that wants strictly facts should edit that phrase in its summary template.

6

Every number opens on the passage it came from

A click on a citation opens event E-1041 with the abstract's results sentence highlighted. Page 41 of the abstract book sits in the document view.

7

Fourteen extracted fields, each with a citation and a confidence

Asset, sponsor, mechanism, indication, phase, patients, route, primary endpoint, three results, the period they cover and two next steps. Each field shows a number from 0 to 1 for how sure the extraction is, and a chip for its source document. Fields under 0.85 go to a person, as section 15 shows. Marker-positive is the lowest here, at 0.93.

8

Week-12 remission side by side, with a drafted so-what

The comparison lines up the same endpoint at the same time point, +17.6 against +16.6 points over placebo, and +24.1 in the subgroup, and prints a label saying the trials are separate and the comparison is not head-to-head. The draft then says what it may mean: "HWL-217 now matches Veltrimab on the headline induction number." That sentence is a draft for Dana. Section 08 shows why she might soften it: 1.0 point sits well inside the rival's own interval. She can edit every word.

9

Dana rates the impact and chooses who else is told

She confirms High, and picks the lists: the Veltrimab program team (14) and Immunology strategy (9), 23 people. Medical affairs and business development stay off. "Nothing is sent until you approve" sits at the top of the card.

10

The alert as it will read, with her name on it

A final dialog shows the text, the lists and the head-count. Sending it is her act: the dialog says "Signed by Dana Okafor", hours before the rival's 2:00 pm call. The event moves to published, and the landscape marks the change.

Held notes, registry moves, the landscape board and the signed morning debrief

In shortThe rest of the screens cover what happens around one late-breaker: every event in one list, a field that needs checking, a note held for counsel, a registry change, a mechanism-by-phase board, a catalyst calendar, briefs and debriefs, a dashboard, and settings your CI lead and counsel own.

A remark from an exhibit stand, held and unused

On October 6, Sam Patel adds a note from the Orsolis stand. He picked up a public study handout, and a staff member said "the unblinded interim look was better than we expected". Nothing public says so. The Compliance Guard holds the note, marks the sentence, marks the public part in a different color, and routes it to Priya Ranvelle, compliance counsel. Only she can release the clean part. The Guard cannot release a held note, and neither can Dana.

A field below 0.85 confidence, and a note from an open session

Oral presentation OP-112 has 12 fields, and one is flagged in red at 0.82: the place and month of the 52-week data. That came from a field note, from a question and answer at the microphone. It goes to a person to check. The note itself passed the public-session check, with the room, the presenter's remark and two questions from the floor written in Sam's own words. Notes are typed: no slide photos or recordings are part of this step.

Every event in one list, and a registry record that moved seven months

Abstracts, registry changes, press releases, filings and notes all end up as one kind of record, with a stage from detected to published. The registry event is a good example: version 14 of a study record moved primary completion from June 2027 to November 2026, because an amendment now reads the primary result when the first 400 participants reach week 52. A rival's maintenance readout may therefore land before your filing. The system kept both versions and shows the difference.

The mechanism-by-phase board and the catalyst calendar through June 2027

Once Dana publishes an event, the board moves. HWL-217 sits in Phase 3 in the Cytokine Q row, flagged "Today", beside Veltrimab in the Cytokine X row. Below it, a calendar runs from October 2026 to June 2027: the investor call, the summit's 52-week data, the registry's new completion date, the earliest top-line maintenance results in February 2027 and your own planned filing.

A six-section Day 3 debrief, signed before it goes out

The debrief workflow runs at 6:00 pm on each congress day, or earlier if Dana asks. It gathers the sessions covered, matches notes, leaves out anything held, and drafts six sections. She signs it, and it goes to 38 people by 7:00 am the next morning, a design target. A later session gets a short update.

Dashboard: events by source, time to alert, analyst hours

The dashboard reports on sample data. It shows 122 events in 30 days, a median of 1.1 hours from source to alert, against a by-hand line of 9.5 hours, and impact as rated by the CI lead. A table shows what each source returned: congress abstracts have a 46% keep rate, because the team chose them, and registries have 0.8%. These are figures from the sample, on fictional data.

The settings a CI lead and compliance counsel keep for themselves

What to watch, when to alert, which sources, who approves. Three compliance policies are locked. The embargo summary is itself a rule, one a team can switch off. Coverage is assigned from the grid, with the questions to answer written down, as Part A recommends.

Six agents read congress sources; a CI lead and compliance counsel decide

In shortSix agents, pieces of software that each do one job within set limits. One kind of message goes out alone: a facts-only embargo summary to one pre-approved channel, under a rule the CI lead approved. Ratings, the so-what, other audiences and the debrief need her signature, and only compliance counsel can release a held note.

Who signs what at a congress: the CI lead, counsel and the rest of the team

The six kinds of output are the embargo summary, which goes out under the rule, then the impact rating, the so-what, an alert to any other list, the daily debrief, and the release of a held note. A team that wants a person on the summary too can switch the rule off, at the cost of the three minutes.

What congress intelligence takes in, what it produces, and how it fits an enterprise

In shortIn go the program, abstracts at their release, public registries, press releases, filings, journals, a licensed database, field notes and your own approved reference cards. Out come a session tracker, cited summaries and events, a signed debrief, and a live landscape. Design targets are stated as targets, and measured in your own run.

Design targets for congress coverage, and how to read them

Roles, audit trail, connections and security for a congress team

Congress rules are yours to follow. Each congress's own rules govern what your team may capture and what software may mine, and getting permission is your job. We cannot give that permission for a society, and we make no claim that the product meets any congress's rules.

Ask each society whether its terms allow commercial mining of its abstract book, and record the answer on the sources page in settings before you switch that source on. The product watches abstracts on the schedule your team puts on file, and a person should re-check that schedule the week before and during the meeting. It says what an abstract reports, and your team decides what that means.

Pricing a late congress brief from your own records, in one week

In shortNobody publishes the cost of a late congress brief. You can count analyst hours, agency fees, and travel for sessions a record could have covered. The larger costs sit outside a spreadsheet: a program lead caught short in a meeting, or one sentence that should never have been passed on. Measure your own in a week, from your last two congresses.

Seven congress costs, where each one hides and how to count it

What a slow congress desk loses that no calculator counts

Your return on congress automation: analyst hours, agency spend and hours earlier to the program team

In shortCount the hours your team spends typing and checking during congresses and on the year-round watch, and any coverage fees you would stop paying. Report time to brief next to the money, without pricing it. If the money does not pay on its own, say so: the case may rest on speed and record-keeping.

Your hours and dollars back across a year of congresses

How ready your congress team is: schedules, rights, field notes and who signs

In shortEight questions, each scored 0 to 2, plus one for reflection. Scores of 0 to 6 say build your base, 7 to 11 say run a pilot, and 12 to 16 say you are ready to implement. Nothing you choose leaves this page.

What to ask a congress intelligence vendor about clocks, rights, citations and field notes

In shortAsk to see things happen on a past congress. Ask to try opening an embargoed abstract, for the first-read time on one, to click a number through to its passage, and for the society's permission. Ask us too. Warning signs sit beside each answer.

A nine-week congress intelligence pilot: replay one past meeting, then shadow one live day

In shortPick a meeting you already covered, and get the society's permission first where its terms require it. Key the answers by hand for thirty abstracts, replay the program, and compare field by field. Test the clock and the held-note path with counsel. Then run alongside your team on one live congress day, and decide on numbers you set before you began.

The seven measures that decide a congress pilot

Congress coverage and abstract embargoes: the questions people search for, answered

In shortTwelve short answers. Each stands on its own.