Conference Surplus, Governance Deficit

Summary

• RIPE NCC's published figures do not show a cash surplus from its two main meetings. In 2024, EUR 250,000 of ticket income and EUR 309,000 of meeting sponsorship covered about 50.4% of EUR 1.11 million in meeting costs. In 2025, EUR 252,000 of ticket income and EUR 201,000 of meeting-specific sponsorship covered about 29.2% of EUR 1.55 million. The calculated uncovered amount rose from roughly EUR 551,000 to EUR 1.097 million.
• The 2025 meetings recorded 741 total entities for RIPE 90 and 597 for RIPE 91, or 1,338 attendance records. Those are not 1,338 unique people, member organisations or network operators: one person could attend both events, non-members can attend, and participation may be online or onsite. Attendance therefore cannot be treated as proof of member authorisation or representative governance.
• The gross 2025 meeting cost was about EUR 1,158 per attendance record, while the cost remaining after meeting ticket and specific sponsor income was about EUR 820 per record. These are scale indicators, not prices or claims that every attendee received equal value. The equivalent uncovered amount was about EUR 55 per year-end active member, but that denominator is also only a sensitivity measure rather than an invoice allocation.
• The proper test is not whether conferences are culturally valuable. It is whether each material increase has an explicit member mandate, a transparent contribution-margin account, sponsor-dependence disclosure and measurable outcomes for operator participation, policy work, service improvement, newcomer retention and regional reach.
• A conference can create an institutional surplus of visibility, relationships and legitimacy while still running a financial deficit. Without an outcome account, management receives the institutional upside, sponsors receive access and recognition, entities receive networking value, and the membership absorbs the residual cost without being shown which public-interest result justified it.

The word “surplus” hides two different accounts

A registry conference can be successful and loss-making at the same time. It can fill rooms, connect engineers, expose a routing problem before it becomes an outage, help a new operator find peers, give working groups a place to test consensus and strengthen the institution's standing across its service region. None of those gains appears as ticket revenue. They are an institutional surplus: a stock of relationships, attention, information and authority that remains after the event closes.

The financial account is different. Venue hire, production, connectivity, travel support, staff time, catering and contractor costs leave the bank account. Tickets and sponsorship return some money. The difference is funded from the wider membership income unless another source is identified. A good event may justify that subsidy. It does not cease to be a subsidy because entities enjoyed it or because the community considers physical meetings part of its tradition.

RIPE NCC's own reporting makes this distinction unavoidable. Its 2024 financial report records EUR 250,000 of meeting income and EUR 309,000 of sponsorship associated with the RIPE meetings. Its external-engagement retrospective says that ticket and sponsorship income covered 46% of RIPE 88's costs and 52% of RIPE 89's costs. The 2025 annual report places the cost of the two meetings at EUR 1.55 million, up from EUR 1.11 million in 2024, while the 2025 financial report records EUR 252,000 of meeting income and EUR 201,000 of sponsorship specifically for RIPE 90 and RIPE 91.

Those figures do not support a claim that the meetings generated a cash surplus. They support a more difficult conclusion. The events generated benefits that the organisation regarded as worth buying, while members financed a larger share of the bill. The central governance question is therefore not “did the conference make money?” It is “who defined the non-financial return, who authorised the residual cost, and what evidence shows that the return reached the networks whose fees paid for it?”

The question matters because a conference is capable of producing its own justification. A larger event attracts more sponsors. More sponsors support more production and hospitality. Better production and hospitality can attract more attendees. Higher attendance then appears as evidence that the event deserves a larger budget. This cycle may produce genuine value, but it can also reward scale independently of outcomes. Without an external control, activity becomes proof of necessity and visibility becomes proof of legitimacy.

The published margin deteriorated sharply in 2025

An event contribution margin is not a complete social-value account, but it is the first financial control. It asks how much of direct reported event cost was covered by event-related income and how much remained with the institution. Applying that simple arithmetic to RIPE NCC's published aggregates produces a clear change.

Meeting year	Reported cost	Ticket income	Meeting-specific sponsorship	Combined recovery	Calculated uncovered amount
2024	EUR 1.110m	EUR 0.250m	EUR 0.309m	50.4%	EUR 0.551m
2025	EUR 1.550m	EUR 0.252m	EUR 0.201m	29.2%	EUR 1.097m

The recovery percentages and uncovered amounts are calculations from published figures, not audited line items labelled “meeting margin.” That distinction is important. Cost-allocation methods may change. Some staff cost may sit in another activity. Aggregate sponsorship reported elsewhere may include regional events or other engagement programmes. The table deliberately uses only the sponsorship that the reports identify with the two main meetings, and it does not invent a value for staff time or indirect overhead that is not attributed there.

Even with those qualifications, the direction is material. Reported meeting cost rose by EUR 440,000, or about 39.6%, in one year. Ticket income was almost flat. Meeting-specific sponsorship fell by EUR 108,000. The amount left after those two income sources nearly doubled. If the higher cost bought a correspondingly larger public benefit, members should be able to see it. If location alone drove much of the increase, as the annual report says, members should be able to see the choice that was made among reach, accessibility, venue standard and price.

The figures also correct a tempting narrative about sponsorship. Sponsorship did not make the meetings self-funding. In 2025, total sponsorship across the RIPE NCC was EUR 421,000, but only EUR 201,000 was identified with RIPE 90 and RIPE 91. Combining all sponsorship with the two-meeting account would overstate cost recovery because other sponsored activities also incurred costs and served different purposes. The economically honest comparison uses like with like.

Nor should the uncovered amount be described as waste. A member-funded conference can be a rational investment, just as a training programme, security control or public measurement platform can be rational without charging its users the full cost. The point is that rational investment requires an investment thesis. “The community values meetings” is not enough to explain why a 39.6% cost increase was the best use of an additional EUR 440,000, why the member-funded portion expanded, or what result would cause management to choose a less costly format next time.

Attendance records are not a constituency

RIPE 90 in Lisbon recorded 569 onsite and 172 online entities, for a total of 741. RIPE 91 in Bucharest recorded 418 onsite and 179 online entities, for a total of 597. Across the two meetings, that is 1,338 attendance records. The same reporting identifies 125 onsite and 36 online newcomers at RIPE 90, and 99 onsite and 48 online newcomers at RIPE 91. It also records broad country representation.

These are useful reach measures. They tell members that the events were not empty, that remote access mattered, that newcomers appeared and that entities crossed national borders. They do not tell members how many unique people attended across the year. They do not identify how many represented fee-paying members, how many operated networks, how many came from vendors, governments, civil society, academia or the RIPE NCC itself, and how many held multiple affiliations. They do not reveal whether small members, remote operators or underrepresented economies were present in proportion to the burdens and risks they carry.

The distinction is more than statistical tidiness. A meeting room is a self-selected population. Travel budgets, visa access, English fluency, employer support, family obligations and confidence in specialist forums all affect who appears. Online access lowers some barriers but does not remove time zones, bandwidth, language and the advantage held by people paid to participate. A full room can therefore represent intense engagement among a narrow professional layer rather than broad authorisation by the membership.

General Meeting participation makes the mismatch visible without proving its cause. RIPE NCC reported eligible-member turnout of 5.3% in May 2025 and 4.1% in October. Those percentages should not be directly compared with the 1,338 meeting-attendance records as though the populations were the same. The meetings include non-members and repeat entities; the General Meeting electorate is composed of eligible members under a different definition. The evidence supports only a governance warning: high event activity can coexist with low formal member participation.

That warning should change how conference success is described. Attendance is evidence of demand for attendance. It is not evidence that the membership authorised budget growth. A newcomer count is evidence that new people entered the room. It is not evidence that they returned, contributed to policy, improved an operating practice or remained involved six months later. Country count is evidence of geographic breadth. It is not evidence that the people attending had authority to speak for networks in those countries.

The registry should preserve those measures while refusing to make them carry claims they cannot support. The next layer is organisational and functional: unique organisations, member status, operator role, network scale, country, participation mode and repeat attendance. The layer after that is outcome: working-group contribution, operational change, service feedback, policy implementation, election participation, mentorship, training completion and later community work. Privacy can be protected through aggregation. What cannot be protected is the assumption that a headcount is a mandate.

A per-attendance figure is a diagnostic, not a ticket price

Dividing 2025's EUR 1.55 million reported meeting cost by 1,338 attendance records produces approximately EUR 1,158 per record. Subtracting EUR 252,000 of ticket income and EUR 201,000 of meeting-specific sponsorship first produces an uncovered amount of EUR 1.097 million, or about EUR 820 per attendance record.

Neither figure means an attendee “cost” exactly that amount. One person may appear in both meeting totals. Onsite and online participation have different marginal costs. A meeting has fixed production and venue expenses that do not fall proportionately when attendance falls. Sponsors and ticket purchasers may overlap with fee-paying members. Staff may perform work before and after the event that benefits the community beyond the days of the programme. The arithmetic is a scale test, not an invoice model.

It is nevertheless useful because it creates a question that an aggregate total avoids. What would members expect to observe from an activity carrying an uncovered amount of roughly EUR 820 per attendance record? If the answer is new operational knowledge, the report should identify adopted practices or documented incidents avoided. If the answer is policy development, it should identify which proposals advanced and how the meeting changed them. If the answer is access, it should show which organisations gained durable participation rather than merely registering once.

If the answer is institutional trust, it should measure the groups whose trust changed rather than citing the people already willing to attend.

The same amount divided by 19,863 active members at year-end is about EUR 55 per member; divided by 20,647 active LIR accounts, it is about EUR 53 per account. Those values are also sensitivities, not assessments on an invoice. The membership and LIR-account counts are year-end stocks, while meeting costs and income accumulated across the year. A legal entity may hold more than one LIR account. New and closing accounts did not all exist for twelve months. Members pay different total amounts once sign-up and resource charges are considered.

The sensitivity still clarifies the choice. A EUR 1.097 million residual is small beside RIPE NCC's EUR 38.952 million of total 2025 expenditure, but it is not immaterial. It is roughly the amount that could fund a set of technical hires, a sustained measurement project, security improvements, targeted fee relief or multiple regional interventions. The relevant question is not which imaginary alternative management should have chosen. It is whether the meeting decision was compared with alternatives using the same evidence standard.

Location is a governance choice disguised as logistics

The 2025 annual report identifies location as a key driver of the higher meeting cost. That is plausible. Venue markets differ. A city that is easy for one part of the service region may be expensive for another. Production vendors, hotel capacity, security, taxation and local transport vary. Hybrid participation adds a technical layer that a purely onsite event did not carry twenty years ago.

Calling location a driver does not end the inquiry. It reveals the decision that needs to be governed. Meeting location distributes at least four things: the institution's direct spending, entities' travel cost, visa and border risk, and the opportunity to participate onsite. A lower venue invoice can impose higher travel costs on attendees. A central aviation hub can reduce time for some and exclude others through visa requirements. A prestigious location may help sponsorship while weakening access for small operators. The cheapest room is not necessarily the cheapest meeting when private entity costs are included.

The selection account should therefore publish a short comparison before commitment. It need not expose commercially sensitive bids. It should show the candidate regions, expected direct cost band, expected ticket recovery, visa accessibility, travel-connectivity index, carbon and travel burden, local operator reach, remote-participation plan and the weighting applied to each factor. After the meeting, actuals should be compared with those assumptions.

This would make cost growth contestable without reducing location choice to a race to the cheapest city. Members could see that an expensive option was chosen because it materially improved access to a neglected part of the service region, reduced aggregate entity cost, or served an explicit rotation rule. They could also see when a costly location failed to deliver the promised distributional benefit.

The principle is simple: logistics should not become a zone of delegated authority merely because it is operationally complicated. When location determines a material share of a member subsidy and changes who can enter the room, it is part of governance.

Sponsorship buys more than a logo

Sponsorship is often treated as an uncomplicated reduction in member cost. A company pays; the meeting receives funds; members finance less. That benefit is real. It should be measured against the access and signalling that the sponsor receives.

At a technical community event, sponsorship can buy brand recognition among a specialised audience, recruitment exposure, hosted social contact, speaking-adjacent visibility, association with the registry's public-interest standing and repeated access to people who influence operational norms. Formal policy power may remain untouched while informal agenda access shifts. A sponsor does not need to purchase a vote to receive strategic value from proximity.

This is not an argument against commercial support. Operators, equipment vendors, cloud companies and service providers are part of the Internet's functioning ecosystem. Their engineers often contribute knowledge that a registry meeting needs. Excluding them would produce a poorer forum. The governance obligation is to distinguish contribution from dependence.

The published aggregate cannot currently do that. It shows meeting-specific sponsorship income but not concentration by sponsor, sector or benefit tier. Members cannot tell whether a large share came from one company, whether the same firms dominate across years, whether in-kind support is valued consistently, or whether particular benefits create access unavailable to ordinary members. Sponsor names and tier logos disclose presence, not economic dependence.

A proportionate account would publish concentration bands without revealing confidential contract detail. It could show the share supplied by the largest sponsor, the largest five, vendor versus operator versus other sectors, cash versus in-kind support, returning versus first-time sponsors and the benefits attached to each tier. It should also identify protected spaces: programme selection, policy formation, fellow selection, moderation and member voting must remain independent of sponsorship status.

The financial comparison reinforces the point. Meeting-specific sponsorship declined from EUR 309,000 in 2024 to EUR 201,000 in 2025 while reported cost rose. That does not prove a loss of sponsor confidence; packages, timing and accounting may have changed. It does show that sponsorship cannot be assumed to scale automatically with ambition. If management expands the event on an expectation of commercial recovery, the budget should state the downside borne by members if that recovery does not arrive.

The self-incentive is institutional, not personal

Criticism of conference growth often collapses into claims about staff travel or individual preference. That is an unnecessarily weak account of the incentive. An institution can rationally favour a larger meeting even when every employee acts honestly.

Events make work visible. A software improvement may prevent failures that never become public; a well-produced conference produces photographs, attendance totals, sponsor lists, videos and immediate praise. Meetings strengthen relationships with governments, vendors, community leaders and prospective partners. They give management a stage on which to explain the organisation and receive direct feedback. They create recurring deadlines around which teams can organise. These are institutional rewards.

The costs are distributed differently. Much of the residual is absorbed by a broad fee base. The member deciding whether to entity may face only a small apparent share. The opportunity cost is fragmented across activities that were not selected. The people receiving concentrated event value are more likely to attend the forum where its value is discussed. This is a classic condition for expansion without bad faith: concentrated benefits, diffuse costs and a metric—attendance—that favours the activity's continuation.

The answer is not to presume that every expansion is empire-building. It is to use controls that do not depend on motive. A multi-year cost ceiling, explicit member approval for departures, comparable location options, event contribution margins, sponsor concentration and post-event outcomes would discipline an excellent team and a weak one in the same way.

This is also why management's own evaluation cannot be the only test. A team that designed the meeting should report what happened, but an independent member committee or audit function should verify the financial bridge and the outcome definitions. The committee need not judge programme content. It should judge whether the claims attached to the expenditure are supported by comparable evidence.

Technical consultation is not budget authorisation

RIPE's open community tradition gives technical discussion unusual legitimacy. Engineers can propose, dispute and refine operating policy in public. Consensus can be more effective than a narrow shareholder vote when a rule must work across autonomous networks. The strength of that model should not be used to blur the membership corporation's financial authority.

A working-group discussion about what the Internet needs is not the same act as members approving how much of their fees should be spent on a conference. A plenary room can test technical ideas without representing all fee payers. A programme committee can select excellent talks without deciding the appropriate institutional subsidy. The RIPE community and the RIPE NCC membership overlap, but they are not identical legal or economic constituencies.

Keeping those mandates separate protects both. Technical entities should not be forced into corporate budget politics in order to contribute engineering knowledge. Members should not be told that questioning event cost is an attack on open technical coordination. The community can advise on programme purpose and participation quality; the membership can authorise cost envelopes and financial trade-offs; management can execute within that authority.

This separation also improves dissent. A member may believe physical meetings are indispensable while opposing a particular venue or cost increase. An online entity may value the content while wanting less hospitality expenditure. A sponsor may support the forum while accepting stricter separation from programme decisions. Governance should allow those positions instead of presenting a binary choice between celebration and cancellation.

The relevant mandate should be written in advance. Members should know the baseline cost, the allowed annual increase, the conditions for an exception and the outcome measures. A material departure should return for approval or at least an advisory vote with a management response. The vote should be accessible outside the meeting itself, so the people affected are not required to travel to govern travel expenditure.

What the event account should contain

A credible event account can fit in a few pages if the organisation chooses the right units. It should begin with a reconciled financial bridge:

1. Direct venue, production, connectivity, catering, security, contractor and fellowship costs.
2. Staff travel and a consistently allocated share of staff time.
3. Ticket income by onsite, online and concession category.
4. Cash and consistently valued in-kind sponsorship.
5. The resulting member-funded amount, with the prior year on the same basis.
6. Approved budget, actual result and explanations for material variance.

The second part should describe distribution rather than raw attendance. Unique entities should be separated from attendance records. Unique organisations should be shown, with privacy-preserving classifications for members, non-members, operators, vendors, public bodies, civil society, research and the registry itself. Onsite and online participation, geographic reach, newcomers, repeat participation and fellowships should be linked to the service region and to stated inclusion goals.

The third part should test outcomes. A technical conference will never reduce to one indicator, but it can still identify observable results: policy proposals advanced; operational guidance published; service changes committed and completed; security or routing incidents analysed; measurement datasets created; cross-network response groups formed; newcomers who contributed again within twelve months; and member organisations that first participated through regional or online access.

The account should include negative findings. A session may be popular and produce no durable action. A fellowship cohort may attend and not return. A hybrid platform may attract registrations but little interaction. A costly location may fail to broaden representation. Reporting those results would not prove failure. It would show that the institution is learning rather than advertising.

Finally, the account should state what will change. Continue, expand, redesign or stop are all legitimate conclusions. A repeated commitment to “build on success” without thresholds is not evaluation; it is an annual renewal request written after the renewal has already been assumed.

A better scorecard separates reach, authority and operating value

The meeting needs three scorecards because one measure cannot carry all claims.

Reach asks who could participate and who did. It includes unique people and organisations, online access, countries, newcomer conversion, operator representation, small-member participation, visa refusals, fellowship reach and cost borne by entities. Reach measures whether the forum crossed barriers.

Authority asks who authorised the expenditure and who influenced decisions. It includes the member vote or approved envelope, General Meeting participation, programme independence, sponsor concentration, conflict disclosures and the separation between community advice and corporate budget authority. Authority measures whether visibility was converted into legitimacy honestly.

Operating value asks what improved in the Internet's coordination layer. It includes implemented policy, registry-service change, shared security practice, routing coordination, incident response, documentation, tools, datasets and sustained collaboration. Operating value measures whether the event changed something beyond the event.

These categories prevent common substitutions. A wide country count cannot compensate for an unauthorised cost increase because it belongs under reach, not authority. A member vote cannot prove technical value because it belongs under authority, not operating value. A strong engineering outcome cannot erase sponsor concentration, though it may justify a subsidy if members approved it with that information.

The scorecard should also identify the intended beneficiary. Some meeting value accrues to the open Internet community rather than members alone. That can be a valid purpose for a public-interest registry, but it should be named and bounded. If non-member public benefit is the objective, management should say how much member money is assigned to it and why the members' mandate permits that transfer.

Remote participation should be costed as infrastructure, not decoration

The 2025 totals show substantial online participation: 172 at RIPE 90 and 179 at RIPE 91. Hybrid access is therefore not a minor broadcast service. It is part of the meeting's operating model.

Online participation can lower travel and visa barriers, but counting a connected browser as equivalent to being in the room would be misleading. Remote entities may find it harder to enter informal discussions, attract a chair's attention, meet maintainers or build the trust through which technical cooperation often moves. Time zones can turn a full programme into an overnight obligation. A high-quality stream is not the same as reciprocal participation.

The event account should separate viewing, active session participation, questions, working-group contributions and later involvement. It should disclose platform and production cost, then test which remote features produce durable participation. Moderation, persistent chat, remote-first question queues, asynchronous notes, multilingual support and scheduled regional replay discussions may matter more than higher video resolution.

That evidence can support a rational format choice. Some sessions may need an onsite core; some may work better as free online Open Houses, which RIPE NCC also runs; some regional work may be more effective in smaller local forums; and one large annual physical meeting might produce nearly the same operating value as two. The correct mix cannot be inferred from tradition or cost alone.

Treating hybrid capability as infrastructure also creates resilience. A registry community should be able to consult and coordinate during travel disruption, political crisis, public-health restrictions or an institutional dispute. The capability should therefore be assessed for continuity and inclusion, not merely as an amenity attached to an onsite event.

The meeting must compete with its alternatives

Every material activity has an opportunity cost even when the annual budget can absorb it. The 2025 uncovered meeting amount of approximately EUR 1.097 million could not simultaneously fund another use. It is not necessary to nominate a preferred alternative to insist on comparison.

Management could present three packages: the proposed two-meeting plan; a lower-cost hybrid or one-meeting plan; and a distributed regional plan. Each would show direct cost, expected income, entity private-cost implications, geographic reach, continuity, sponsor risk and anticipated outcomes. Members could then see what they were buying with the incremental amount.

The alternatives should be real enough to choose. A deliberately unattractive “cheap” option does not create accountability. Nor should every technical detail be subjected to a vote. Members can approve a purpose, cost band and measurable trade-offs while leaving procurement and programme execution to professionals.

Competition among formats would also expose hidden assumptions. If two large meetings are necessary because policy work depends on six-month cycles, the evidence should show which policy work would stall under another format. If regional events cannot substitute because they fragment the community, the account should show what cross-regional result is lost. If sponsorship depends on scale, the downside should appear in the financial model rather than as an unsupported reassurance.

This is the practical meaning of a member mandate. It is not a plebiscite on catering. It is the right to see the material options, their distributional consequences and the outcome promised for the extra expenditure.

Conference value should flow back to the networks

The strongest defence of a registry meeting is operational. The Internet is built by autonomous organisations that must coordinate without a central network owner. Engineers need places to compare failures, challenge proposed policy, understand routing incidents, maintain shared norms and build enough trust to respond when formal contracts are too slow. A high-quality meeting can reduce coordination cost across the service region.

That defence becomes weaker when the flow back to operating networks is invisible. An employee may attend, learn and leave without the organisation changing anything. A vendor may gain leads while member networks receive little. A policy discussion may involve experts but not the operators who will bear implementation cost. A presentation may remain an archive video rather than becoming usable guidance.

The registry should therefore measure transfer. Did attendees brief their organisations? Did published notes reach non-attending members? Did a recommended practice enter deployment? Did a member service change after structured feedback? Did smaller operators gain access to expertise they could not otherwise afford? Did a cross-network contact formed at the meeting later help resolve an incident?

Some of this evidence can be gathered through samples rather than surveillance. A three- and twelve-month follow-up with participating organisations, linked to specific sessions and commitments, would be more valuable than a satisfaction survey completed before attendees leave the venue. Working groups can report which meeting interventions changed a proposal. Service teams can report which issues entered and exited their backlog.

The objective is not to make every conversation auditable. Informal contact is part of conference value precisely because its future use is uncertain. The objective is to prevent uncertainty from becoming an unlimited budget claim. A portfolio can contain exploratory activity if members know its cost, purpose and review horizon.

A governance compact for RIPE meetings

The financial evidence points toward a compact rather than a cap. A hard cap could make the event cheaper while excluding exactly the entities who need support. A compact would bind expansion to authority and evidence.

First, RIPE NCC should publish a comparable two-year event margin using consistent cost allocation, including the member-funded residual. Second, it should obtain member approval for a rolling multi-year envelope and define the percentage or absolute variance that requires a new decision. Third, it should publish a location-options account before contracting, with distributional and entity-cost effects. Fourth, it should disclose sponsorship concentration and protected decision spaces. Fifth, it should replace raw attendance as the headline proof of value with the reach, authority and operating-value scorecards.

Sixth, the organisation should report both the unique-person and unique-organisation denominator, not only attendance records. Seventh, it should measure newcomer retention and operator participation after the event. Eighth, it should compare the selected format with at least one credible alternative. Ninth, it should state which activity would be reduced if the meeting exceeds its envelope rather than treating the wider budget as an automatic shock absorber. Tenth, an independent member or audit body should verify the financial bridge and outcome definitions.

None of these controls requires the RIPE meeting to become a profit centre. Full cost recovery could produce its own exclusion: high ticket prices would favour large vendors and wealthy organisations, undermining the event's open technical function. The target is not zero subsidy. The target is an authorised subsidy whose beneficiaries and results are visible.

Fee waivers and fellowships should appear in that account as deliberate access policy, not as missing income. The report can show the number and broad distribution of supported entities, the foregone ticket value, direct travel support and later participation without naming individuals. That treatment protects inclusion from the mistaken conclusion that every unrecovered euro is inefficiency. It also lets members distinguish money spent to overcome structural barriers from cost generated by venue, production or hospitality choices. A subsidy is easier to defend when its intended recipient and public purpose are stated before the event.

The decisive test is what happens when the evidence is weak

An accountability system is not defined by the data it publishes. It is defined by the decision that follows. If cost rises, sponsor recovery falls and outcome evidence remains weak, what changes?

The answer cannot always be cancellation. A single year may be affected by an unusually expensive location, contract timing or a strategic need to reach a part of the service region. But repeated exceptions should trigger redesign: fewer physical days, a different city band, one large meeting, more regional sessions, revised hospitality, a different sponsorship model or stronger remote participation.

Equally, strong evidence should protect the budget. If a costly meeting demonstrably brings neglected operators into policy work, produces deployed security improvements and reduces coordination failures, members should be able to authorise it without pretending tickets cover the cost. Transparent subsidy is more legitimate than fictional self-sufficiency.

The review horizon should also match the claimed benefit. Ticket recovery and attendance can be reported within weeks, while newcomer retention, implementation and institutional trust need months. A preliminary account can close the financial year, followed by a twelve-month outcome note before the next comparable commitment is approved. That timing prevents management from declaring success before the harder evidence exists, yet it does not postpone financial disclosure until memories and responsibility have faded.

It also lets members distinguish a one-off expensive meeting from a format whose costs and weak outcomes persist across several cycles.

This is where the conference surplus and governance deficit meet. The surplus is the authority, access and social capital created by convening the community. The deficit is the missing mechanism that returns that value to the principals who finance the institution and gives them a meaningful choice before cost expands.

A mature registry does not need to apologise for gathering its community. It needs to show that convening is a means rather than an end. Attendance demonstrates that people came. Sponsorship demonstrates that companies saw value. A balanced account must demonstrate the third proposition: that member-funded expenditure produced operating and governance value that could not have been obtained more fairly or efficiently through another format.

Sources and method

The financial comparison uses RIPE NCC's audited or officially reported annual figures. The 2024 meeting cost, ticket income, meeting sponsorship and event-specific recovery percentages come from the RIPE NCC Financial Report 2024 and the organisation's 2024 external-engagement retrospective. The 2025 cost, participation, newcomer, member, LIR-account and General Meeting figures come from the RIPE NCC Annual Report 2025. Ticket income, sponsorship and total-expenditure figures come from the RIPE NCC Financial Report 2025.

Calculated recovery, uncovered cost and unit figures are explicitly identified as arithmetic rather than audited event-profit measures. Attendance is counted as records across two meetings, not unique people. The active-member and LIR-account divisions use year-end denominators and are presented only as sensitivity measures. No conclusion assumes that every member paid equally, that every entity received equal value, or that meeting income and expenditure share a perfect cost-allocation boundary.