The company's R&D Day program in 2026: disclosure and decisions

Working program of the company's R&D Day: how to select research, protect confidential data, prepare posters and demos, and set the next stage of projects.

A company's R&D Day is needed when executives and research teams are ready to see the portfolio as a whole. Authors show not only a successful result but also the initial uncertainty, the validation method, limitations, and a request for the next stage. The business gets a comparable picture, while teams get substantive feedback and a decision owner.

At Aventura, we start such a project with boundaries. First, we determine which research can be shown to the entire company, which requires a private room, and which cannot be presented at the event before an intellectual property review. Then we assemble the program, participant routes, posters, demos, and the technical plan.

How does R&D Day differ from Demo Day and a hackathon?

In this article, we use a working distinction between formats. R&D Day reviews the current research and development portfolio. Teams explain the hypothesis, method, evidence, uncertainties, and the next research step. Demo Day showcases prepared solutions, Innovation Day gathers or selects ideas, and a hackathon creates a prototype within a limited time. These formats can coexist, but they should not replace one another.

The boundary matters for selection. If participants arrive with a finished product and ask for permission to run a pilot, that is closer to a corporate Demo Day. If the programme is broader than a research portfolio and includes idea collection, talks and demos, the right scenario is an Innovation Day at the company. If the team builds a solution right at the event, you need the scenario corporate hackathon. R&D Day runs earlier or in parallel: it helps to understand what has already been proven and what uncertainty remains to be tested.

The OECD Frascati Manual 2015 describes research and development through novelty, a creative approach, uncertainty, a systematic approach, and the ability to transfer or reproduce the result. For an R&D Day application, this framework can be translated into five simple questions: what is new here, what hypothesis was tested, what was uncertain, how the work was carried out, and whether another team can repeat the method.

If a project does not pass such a filter, it should not be devalued. Engineering improvements and scaling can be moved to a separate track. Then the jury does not compare fundamental research with already implemented automation on the same scale.

Decisions determine the R&D Day program

In short: Before choosing the stage, venue, and speakers, the company determines the decisions that should remain after R&D Day. For each project, this can be continuation of the experiment, additional validation, merging teams, access to data, preparation of a patent application, a technical pilot, or discontinuing a direction. Without this framework, the day turns into an exhibition without follow-up.

Первый документ проекта - карточка результата дня. В ней фиксируют владельца портфеля, состав принимающих решения, типы допустимых решений и срок следующей контрольной точки. Мы строим программу от этой карточки назад.

Участнику заранее объясняют, какой запрос он может вынести на комитет. Формулировка «нужна поддержка» слишком широкая. Рабочие запросы звучат конкретно: предоставить тестовую площадку, назначить эксперта по материалам, открыть обезличенный набор данных, провести IP-review или подтвердить ресурсы для следующего эксперимента.

For event production, the mechanics are suitable for a business event: a single headquarters, decision owners, a transition scenario, and a change log. The organizer does not evaluate the scientific validity of the project. We connect experts, authors, management, and technical services so that a decision is made at the right moment and recorded.

How to select projects?

Отбор проходит в два контура. Сначала технические эксперты проверяют, относится ли заявка к исследованиям и достаточно ли в ней доказательств для обсуждения. Затем программная команда выбирает подходящий формат: короткий доклад, постер, демо, закрытый разбор или цифровая галерея. Харизма автора и качество ролика не должны заменять содержание.

Заявка может состоять из семи полей:

1. problem or research question;
2. initial uncertainty and hypothesis;
3. method and data used;
4. obtained result;
5. limitations and negative results;
6. level of disclosure;
7. request for the next stage.

На первом проходе мы предлагаем скрыть должность автора и название подразделения, чтобы оценщики смотрели на описание проекта. Полностью слепой такая оценка не станет, но внутренняя иерархия будет меньше отвлекать от содержания. После технической проверки программная группа распределяет проекты по трекам: discovery, methods, engineering, translation, scale и shared research infrastructure.

Не каждый сильный проект обязан выходить на главную сцену. Complex methodology works better in a poster session. Sensitive details are discussed in a closed room. A demo is needed only where the audience must see how the prototype behaves, not a beautiful interface.

Disclosure is divided into three access levels

In short: Every project must have an information owner and a confirmed disclosure level before editorial work on the presentation. The team separately checks trade secrets, patentable features, partner agreements, personal data, and export and industry restrictions. Recording, photos, and the shared archive are opened only for approved materials and audiences.

WIPO links trade secret protection to reasonable measures to maintain secrecy. Among such measures, an organization may use confidentiality agreements, labeling, and restriction of physical or technological access. The European Patent Office notes that public written or oral disclosure may affect the assessment of novelty.

For the event, it is convenient to use three levels:

This approach is more precise than the binary “show or hide.” The European Research Executive Agency frames the principle of open science as maximum openness with necessary closure due to intellectual property, personal data, confidentiality, and security.

Before filming, the team prepares a zone map. Posters and screens must not contain access keys, real client data, names without justification, closed diagrams, or accidentally open work chats. The client determines the legal basis and rules, and we translate them into the script, navigation, and briefing for the film crew.

Company R&D Day program

A strong program alternates short talks, posters, demos, meetings with business customers, and closed expert deep dives. The main stage provides navigation across the portfolio, but deep work happens in small formats. The finale is dedicated to decisions, owners, and timelines, not just awards and a group photo.

A working sequence might look like this:

1. Quiet start and digital gallery. Participants familiarize themselves in advance with permitted materials, check the route and accessibility requests.
2. Portfolio opening. The R&D head identifies priority unknowns and explains the types of decisions for the day.
3. Short research talks. Each author presents the problem, method, evidence, limitation, and request.
4. Poster session. Authors are at their stands according to the schedule, while participants follow thematic routes.
5. Lunch with problem-focused tables. Seating is organized around tasks, not departments.
6. Negative results session. Teams review disproven hypotheses and the conditions under which the method does not work.
7. Demo session. Prototypes are shown after a readiness check and with a fallback scenario.
8. Reverse pitches from the business. Business units formulate research tasks and available resources.
9. Closed expert rooms. Authorized participants discuss sensitive details.
10. Portfolio committee. For each project, the next step, owner, and review date are recorded.

For the main stage, parallel sessions, and registration, you can use the experience of organizing conferences. But R&D Day cannot be assembled as a continuous series of talks. Researchers need time at the posters, experts need separate rooms, and leaders need a short decision block.

A poster reveals one story, a demo - one effect

In short: A poster conveys one central story, and a demo shows one verifiable effect. The author does not transfer the entire report to the stand and does not reveal all the prototype's functions. A unified template, editorial review, clear chart labels, and backup recording help participants quickly understand the logic of the result without revealing anything unnecessary.

We ask the author to build the poster around one main story, shorten the text, and use fonts that are readable from a distance. We check contrast and color scheme accessibility according to W3C recommendations. For a corporate R&D Day, the poster answers seven questions: problem, hypothesis, method, data, result, limitations, and the next experiment.

The author should have two explanations. The first is intended for a specialist in their own field. The second is for a competent colleague from another function. This helps transfer knowledge between R&D, product, manufacturing, and business without reducing it to a marketing slogan.

The demo undergoes a lightweight readiness check. The team tests equipment, network, accounts, data, security, the stop criterion, and the fallback scenario. NASA requires a Test Readiness Review for tests with new or modified equipment and/or procedures; the review confirms and documents safety, technical, and operational readiness. For a corporate event, we use a shortened analogue of such control.

For AI projects, we ask to separately name the application context, data, metrics, limitations, human oversight, and known errors. NIST AI Risk Management Framework links responsible work to four functions: Govern, Map, Measure, and Manage. For the event, this means a clear owner, a described application context, verifiable metrics, and a risk treatment plan.

How to evaluate research and development?

In short: Evaluation separates research quality, potential impact, and the feasibility of the next stage. The presentation format receives a small separate weight and does not compensate for weak evidence. Criteria are published before applications are submitted, and technical evaluation is carried out before the event. Audience voting can be kept as an additional signal, but not as the only outcome.

The European Commission uses pre-announced criteria for research applications: scientific quality, impact, and quality and efficiency of implementation. For an internal R&D Day, the following 100-point rubric is suitable:

Awards are also better split by type of contribution: best evidence, strong negative result, reproducible method, research infrastructure, mentorship, and accessible poster. Then the company does not create a single ranking in which a mature product always beats early-stage research.

Venue, hybrid format, and accessibility

The venue should support quiet poster zones, safe demonstrations, closed rooms, a main stage, and clear routes. A hybrid participant gets a digital gallery, remote questions, and access to permitted materials. Accessibility is planned before registration opens: the physical path, subtitles, keyboard navigation, and document format are checked.

If the program includes posters, demos, and closed reviews, one hall is usually not enough for the zones to run simultaneously. Posters require even lighting and wide aisles. Demos need electricity, network, and a controlled area. Closed expert review should not take place behind a makeshift screen next to the main stage.

If research centers are located in different cities, we connect a hybrid format. The remote presenter checks the sound, screen sharing, backup recording, and access rights in advance. The general broadcast must not accidentally show private windows or internal notifications.

W3C advises taking accessibility into account before, during, and after the event. In practice, this means an accessible route, an accommodation request form, captions, microphones for questions, materials with a clear structure, sufficient contrast, and a digital gallery that can be used with a keyboard.

A rehearsal tests the entire participant journey

In short: The rehearsal follows the participant's journey and the content lifecycle. The team checks registration, access rights, posters, demos, transitions, closed rooms, committee decisions, and publication of approved outcomes. Every risk has an owner and a backup. A table read of the script is not enough if prototypes and systems have not been tested on the actual site.

The project headquarters needs owners for six workstreams:

• R&D content and technical selection;
• intellectual property and legal restrictions;
• information and physical security;
• program, moderation, and decisions;
• venue, equipment, and hybrid;
• internal communications and final materials.

At Aventura, we consolidate their decisions into a run of show. It specifies readiness times, access permissions, escorts, filming restrictions, demo stop points, backup files, and the person who authorizes the transition to the next block.

Separately, we check photo and video coverage of the event. The video crew receives a map of permitted zones and a list of materials. Recording of the closed technical room is not carried out by default. The public video is checked by the information owners before posting.

What to do after R&D Day?

After the event, a decision register remains, not just a photo album. For each project, the status, owner of the next action, required resources, and review date are recorded. We recommend scheduling checkpoints at 30, 60, and 90 days. At these checkpoints, the team reviews which experiments have continued, which methods have been reused, where an IP or security review was required, and which initiatives have been stopped for a clear reason.

Attendance and satisfaction help evaluate the service, but do not replace portfolio results. It is more useful to count the number of scheduled reviews, merged teams, opened data accesses, technical pilots, and closed initiatives with a documented rationale.

Materials also get a lifecycle. The shared gallery stores only the approved version. Private files are kept in a system with role-based access. The owner periodically checks who still needs the material and when access should be closed.

We write about working observations, preparation, and new business formats in Aventura's Telegram channel.

Frequently asked questions

In short: Clients most often ask how many projects to take on, whether to invite external guests, whether every author needs a stage, and whether demos can be recorded. The answer depends on the portfolio and the information mode. First the company determines the decisions and disclosure levels, then chooses the formats. The number of presentations adapts to the depth of discussion, not the other way around.

How many projects should be included in R&D Day?

The number depends on the formats. The main stage can accommodate a limited number of short talks. The other strong projects can be shown in poster sessions, the digital gallery, and private rooms. It is better to give the author time for questions than to assemble a long queue of presentations without discussion.

Can clients, partners, and universities be invited?

Yes, if a separate level of materials has been prepared for the external audience. A guest's access to the venue does not mean access to the entire portfolio. Partnership agreements, intellectual property, and personal data are checked before sending the invitation.

Are awards needed?

Awards help recognize different types of contribution, but they should not replace decisions about projects. We recommend recognizing evidence, reproducibility, negative results, infrastructure, and mentoring, not just the most impressive prototype.

What to do if the demo doesn't work during rehearsal?

The team decides whether the issue can be safely fixed before the event. If not, the live demo is replaced with a verified recording or a poster. The backup goes through the same disclosure check, because the video may contain confidential data and interfaces.

Should all presentations be published after the event?

No. For each item, the access level, retention period, and owner are set in advance. The shared archive should not automatically receive confidential applications, source data, and working versions of presentations.

How to measure the value of R&D Day?

Look at decisions and portfolio movement: assigned experiments, merged teams, access to resources, reuse of methods, completed reviews, and the share of completed next steps. These metrics complement attendance and participant feedback.

If a company is preparing an overview of its research portfolio, at Aventura we can bring the selection, access modes, posters, demos, expert rooms, and the decision committee into a single scenario. Request a quote via the form, to discuss the project lineup, audience, and technical constraints.

Sources

• OECD: Frascati Manual 2015 - five criteria for R&D
• WIPO: protecting trade secrets
• European Patent Office: Guidelines for Examination 2026, disclosure
• European Commission: how research projects are evaluated
• European Research Executive Agency: Open Science
• W3C WAI: accessible presentations and events
• NASA: Test Readiness Review
• NIST: AI Risk Management Framework 1.0