Quick answer
An MSCA Doctoral Network is a doctoral training programme, not a collection of unrelated PhD projects. The proposal needs a coherent research challenge, a consortium that creates genuine international, interdisciplinary and intersectoral value, and a training and supervision system that develops the doctoral candidates beyond the technical needs of their individual projects.
The 2026 call opened on 28 May and closes on 24 November 2026. It has an indicative budget of EUR 593 million and is expected to fund more than 130 projects and around 1,950 doctoral candidates.
Official source: MSCA Doctoral Networks 2026 call.
Start with the programme-level question
The proposal should have a reason to exist at network level.
A common weakness is to begin with several good supervisors who each propose a PhD topic. The coordinator then tries to connect those topics under a broad theme. This can produce a consortium, but not necessarily a network.
A stronger sequence is:
shared research and training challenge -> network-level objectives -> candidate projects -> shared methods and training -> sectoral and international exposure -> durable capability
The individual doctoral projects should be necessary parts of the programme rather than isolated projects that happen to share a grant agreement.
The consortium is part of the scientific argument
MSCA Doctoral Networks can include universities, research organisations, infrastructures, companies, SMEs, public bodies and other socio-economic actors. The proposal should explain why each partner changes what the network can do.
Useful roles include:
- complementary scientific expertise;
- access to specialist infrastructure or datasets;
- industrial or non-academic research environments;
- clinical, policy or user perspectives;
- training capabilities unavailable at other beneficiaries;
- secondment environments that expose candidates to a different mode of research or innovation.
The partner list should therefore read like a capability map, not a reputation list.
Design doctoral projects from the network architecture
Each doctoral candidate needs a credible individual project. The projects also need to interact.
Look for meaningful interfaces:
- shared samples or platforms;
- sequential outputs where one project informs another;
- common methods that allow comparison across systems;
- shared datasets or models;
- complementary experimental and theoretical questions;
- joint validation environments;
- network-wide training linked directly to the research challenge.
The evaluator should be able to see why the candidates are stronger because they belong to the network.
Training should be programme-specific
Generic training menus are easy to write and easy to forget. A stronger programme starts from the capabilities the candidates need to become excellent researchers and professionals.
Separate the design into three layers:
Research-specific training
Methods, instruments, models, data techniques and domain expertise needed for the scientific programme.
Transferable capability
Communication, project management, open science, research integrity, entrepreneurship, IP, leadership, teaching or data stewardship where relevant.
Exposure and career development
Secondments, industry interaction, interdisciplinary work, international mobility, mentoring and individual career development planning.
The best training plan feels like an operating system for the network rather than a calendar of courses.
Supervision must work across institutions
Doctoral Networks place strong emphasis on supervision and career guidance. A network-wide supervisory structure should therefore do more than satisfy governance requirements.
It should define how supervisors coordinate, how candidate progress is reviewed, how secondments are prepared and assessed, how problems are escalated and how good practice is shared across beneficiaries.
Industrial and Joint Doctorates have additional structural requirements, including joint supervision arrangements in the relevant cases. The exact rules must be checked in the current Work Programme and call documents.
Read: MSCA Supervision, Training and Knowledge Transfer
Excellence is research plus training architecture
The scientific objectives need to be ambitious and methodologically sound. The network also needs to show that the training environment is itself excellent.
That means the proposal should connect:
- the research challenge;
- the candidate projects;
- the methods and infrastructure;
- the training programme;
- the supervision system;
- the international/intersectoral/interdisciplinary dimensions.
If those elements are described separately, the proposal becomes harder to evaluate as one programme.
Read: How to Write the MSCA Excellence Section
Impact should operate at several levels
A Doctoral Network can create impact for:
- individual candidates;
- participating organisations;
- the scientific field;
- the non-academic sectors involved;
- the European research and innovation ecosystem.
The proposal should explain mechanisms, not simply list benefits.
For example, an industrial secondment matters because it exposes a candidate to constraints, tools or validation environments that change how they approach the research problem. A shared training school matters because it creates a capability the beneficiaries could not provide individually. A common data or methodology standard matters because it can outlive the funding period.
Read: How to Write the MSCA Impact Section
Recruitment is part of quality
MSCA Doctoral Networks require open, transparent and merit-based recruitment. Vacancies must be advertised widely, including through EURAXESS, and the proposal should show that recruitment is organised professionally.
Do not treat this as an administrative paragraph. Candidate selection determines the scientific quality of the programme. The recruitment plan should therefore be credible, fair and capable of attracting the profiles the network actually needs.
Governance should remain operational
Large consortia often over-engineer governance diagrams. Keep the structure readable.
At minimum, make clear:
- who owns scientific coordination;
- who owns training coordination;
- how candidate supervision is monitored;
- how beneficiaries make decisions;
- how conflicts are handled;
- how risk and quality are reviewed;
- how the supervisory board operates.
The governance model should help the network function, not simply look formal.
Final network test
Before submission, ask:
- Is there one programme-level challenge connecting all candidates?
- Could any doctoral project be removed without changing the network logic?
- Does every beneficiary add a distinct capability?
- Is the training programme specific to this network?
- Are secondments scientifically or professionally necessary?
- Does supervision work across organisations, not only locally?
- Can the evaluator see what persists after the grant ends?
A strong Doctoral Network should leave behind more than completed PhDs. It should create people, methods, relationships and research capacity that continue to generate work after the funded programme ends.
Preparing a Doctoral Network?
Arvacore supports coordinators with programme architecture, consortium logic, candidate-project integration, Excellence and Impact review, training and supervision structure, work-package design and final consistency review.
Explore MSCA proposal support or request a fit check.
Independent Arvacore guidance. Always check the current MSCA Work Programme, call page, Guide for Applicants and official FAQs.