A construction project can leave behind more than a finished building. The challenge is to make learning, decent work and community benefit part of the delivery system—and to prove what actually happened.
A building has two kinds of legacy
When a school, library or community centre opens, its visible legacy is easy to photograph. We can see the walls, the entrance, the structure, the landscape and the spaces where people will gather. The less visible legacy is harder to assess. Who acquired a new skill while it was built? Did a trainee receive meaningful supervision? Did a small supplier gain experience that could make it more competitive next time? Did the project create a lasting connection between an employer and someone seeking a route into working life?
These questions are often treated as an optional social programme, separate from the 'real' construction project. That separation misses an opportunity. Construction already requires people to learn, coordinate and solve problems together. A project creates a temporary workplace and a network of clients, designers, trades, suppliers and public institutions. With appropriate planning, some of that activity can support vocational learning and inclusion.
The difficulty is that good intentions are not the same as good outcomes. A contractor can advertise community value without providing useful training. A client can ask for apprentices without ensuring that their work is relevant, safe or properly supervised. A digital platform can collect names and promises without helping anyone complete a meaningful task.
This article proposes a different approach: treat social value as a verifiable part of project delivery, not a slogan attached to it. The focus is on public procurement and construction, with Norway as a practical regulatory reference point and SamfunnsBro as a potential community-partnership model. The proposals are exploratory; they do not describe a completed Jarlhalla pilot.
1. What social procurement means
Social procurement is the deliberate consideration of social outcomes when buying goods, services or works. Depending on the contract and applicable law, those outcomes may concern decent working conditions, accessibility, inclusion, training opportunities or participation by people facing barriers to employment.
The European Commission's Buying Social guide, published 18 June 2021, explains how public buyers can consider social issues from planning through contract management. Its social procurement overview identifies decent work and inclusion as possible objectives. These are policy approaches, not proof that every contract creates positive outcomes. [1, 2]
A buyer can express an aspiration to support skills without yet having a defensible procurement requirement. A useful requirement must identify what the contractor should deliver, why it relates to the contract, how it will be verified and what happens if performance falls short.
Social procurement is not permission to favour a preferred community organisation or hire exclusively from one postcode. It is a structured way of considering legitimate social objectives while respecting competition, equal treatment, transparency and the rules governing the purchase.
The starting question is: which learning or inclusion opportunities are genuinely created by the work being performed?
2. Permission, obligation and ambition are different
Public buyers operate within a hierarchy of legal requirements, permitted choices and voluntary project ambitions. Confusing these levels creates risk.
Article 70 of EU Directive 2014/24/EU allows contracting authorities to establish contract-performance conditions involving social or employment considerations, provided those conditions relate to the subject matter of the contract and are disclosed in the procurement documents. This is a legal possibility subject to conditions, not a general authorisation to add any social requirement. [3]
Norwegian procurement guidance similarly stresses the connection to the delivery. DFØ explains that award criteria and contract terms must relate to the goods, services or works being procured. A condition about unrelated corporate activities may not satisfy that requirement. [4]
An award criterion evaluates tenders. A minimum requirement defines an essential threshold. A contract-performance condition governs how the successful supplier performs the work. These mechanisms have different legal and evidential functions and should not be treated as interchangeable.
Supervised learning related to work performed under a building contract may be easier to connect to the delivery than a requirement that the contractor support an unrelated campaign. Even then, wording, scope and proportionality need legal and procurement review. For a real procurement, a qualified buyer must establish the current rules, thresholds, exemptions and contract structure before adopting proposed wording.
3. Norway's apprenticeship rules make the issue concrete
Norway provides a useful case because apprenticeship participation is not merely an abstract social ambition. DFØ's official guidance explains when contracting authorities must include apprenticeship-related conditions in certain services and construction contracts. [5]
Under Norway's conditional framework, relevant public contracts must require a supplier to be connected to an apprenticeship scheme, with at least one apprentice participating and at least 10 per cent of relevant work performed by apprentices. This does not apply to every construction contract. Application depends on contract scope, value, duration, trades and whether there is a particular need for apprenticeship places.
DFØ explains that the percentage concerns hours of work relevant to educational programmes with a particular need for apprenticeship places. It is not 10 per cent of the contract price, and it is not necessarily 10 per cent of all hours across every trade. The requirement for at least one apprentice is separate. [5]
A project cannot establish compliance by showing a photograph of an apprentice or counting someone who visited the site for a day. The team needs a defensible denominator, reliable time records and clear definitions.
Udir explains that an apprentice has a training contract and employment relationship with an approved training enterprise. A volunteer workshop, site visit or informal mentoring conversation is not automatically a formal apprenticeship. [6]
4. Apprentice hours are not the same as learning outcomes
Imagine two hypothetical projects recording the same apprentice hours. In the first, an apprentice repeats a narrow task with little feedback. In the second, the apprentice receives planned exposure to relevant work, regular supervision, documented competencies and structured conversations about progression. The reported hours may be identical, while the learning experience differs greatly.
The International Labour Organization's Quality Apprenticeships Recommendation No. 208, adopted in 2023, addresses structured training, suitable compensation, occupational safety, supervision, assessment and recognition of skills. It is an international recommendation, not a replacement for Norwegian law. [7]
A responsible project therefore needs two reporting layers. The first concerns contractual compliance: eligible hours, approved arrangements and relevant records. The second concerns training quality: learning plans, supervisor availability, completed modules, feedback and evidence of progression.
A positive satisfaction survey cannot cure failure to meet a binding contract term. Conversely, technical compliance with an hours requirement does not prove long-term employment or social inclusion.
Projects should avoid claiming that training caused later employment or income gains without an evaluation capable of supporting that conclusion.
5. Fair competition is part of social sustainability
There is a temptation to define social value as keeping work local. Local relationships can be valuable, but public procurement cannot simply become a mechanism for favouring familiar businesses or preferred partners.
DFØ explains that award criteria must be objective, disclosed, linked to the delivery and consistent with procurement principles. Suppliers need to understand what will be evaluated. [4]
A better approach asks what capability the project needs rather than naming a particular supplier, town or organisation as the required beneficiary. Contract-related training conditions should be considered only when they are lawful, proportionate and open to qualified bidders.
A future SamfunnsBro network could help people discover public opportunities or connect with training organisations. It should not promise privileged tender access, certify apprenticeship compliance or act as a gatekeeper for public contracts.
Social value loses credibility if the process used to create it is unfair.
6. From a project brief to a skills brief
Construction projects already develop technical briefs, work packages, schedules, risk assessments and information requirements. A practical innovation would be a corresponding skills brief before procurement or mobilisation.
It would identify relevant trades, legitimate supervised learning opportunities, when those tasks occur, what approvals are required and who is responsible for supervision. It would also identify constraints: safety-critical work, restricted areas and activities where an inexperienced participant cannot safely contribute.
A skills brief is not a substitute for a formal training contract, curriculum or educational plan. It makes the opportunities and limits of a specific project visible to the relevant organisations.
In a refurbishment, activities might include surveying, careful material removal, carpentry, insulation, technical installations and commissioning. Some may support approved apprenticeships; others may support observation, career guidance or classroom exercises outside the site.
The categories must remain distinct. A student completing a BIM exercise is not automatically an employee. A visitor attending a site presentation is not necessarily an apprentice. A volunteer at a community event should not be recorded as performing paid construction work. Clear labels protect participants and prevent inflated claims.
7. A five-stage delivery model
Define the need. The client identifies legitimate outcomes such as apprenticeship participation, accessible career information or stronger connections between training providers and employers. These ambitions must be checked against procurement rules before becoming requirements.
Map real opportunities. Designers, contractors and training representatives examine the programme for appropriate tasks. They identify likely hours, disciplines, supervisory capacity and the effects of schedule changes.
Assign responsibility. The contractor remains responsible for its workers, safety and contractual obligations. Approved training enterprises and educational authorities retain their statutory roles. A community intermediary may support communication but cannot assume authority it does not possess.
Record performance. Eligible apprentice hours, learning milestones and supervision activities are documented through agreed processes. Sensitive personal data are separated from public reporting. Disputed records are investigated.
Evaluate and improve. The client checks contractual compliance and examines whether additional activities produced useful outcomes. Problems are recorded alongside successes.
This model creates safe stop points. If an activity lacks an appropriate supervisor, authorised training arrangement or lawful basis, it should not proceed merely to satisfy an impact target.
8. A hypothetical community library renovation
Imagine a municipality planning to renovate a small public library. The works involve the building envelope, internal finishes, technical systems and accessibility. They could also provide opportunities to explain the construction industry to students and career changers.
Before procurement, the client reviews relevant social considerations and whether statutory apprenticeship conditions apply. It identifies necessary records and avoids selecting a preferred contractor or intermediary in advance.
After award, the contractor develops a supervised learning schedule. Approved apprentices participate through their employers and training arrangements. Their work is planned around real tasks, curriculum requirements and site safety.
Separately, the municipality could consider a public learning event. An architect might explain how a drawing becomes a construction detail; a tradesperson could discuss material reuse; a mentor might describe career pathways. These events would have their own safety and organisational arrangements and would not automatically count as apprentice hours.
At handover, the client reports what it can verify: contractual conditions, documented participation, completed events and feedback. It does not claim the project solved youth unemployment or transformed the local economy.
This example is entirely illustrative. It is not a real Jarlhalla commission, municipal agreement or completed SamfunnsBro programme.
9. Where SamfunnsBro could contribute
Jarlhalla's public SamfunnsBro description presents a community initiative connecting people, knowledge, work and collaborative projects. Its named pathways include MentorBro, ArbeidsBro, KunnskapsBro, SkaperBro and BedriftsBro. That description is evidence of a stated concept, not evidence that the procurement-to-skills workflow proposed here has been implemented. [8]
MentorBro could introduce experienced professionals to people seeking general career guidance. Formal supervision of apprentices would remain with authorised training organisations and employers.
ArbeidsBro could explain the difference between employment, apprenticeships, internships, volunteering and project collaboration, with clear terms. It should point users toward authoritative routes rather than promise jobs.
KunnskapsBro could maintain reviewed resources about qualifications, construction safety and technology. SkaperBro might support non-safety-critical learning exercises, while BedriftsBro could provide a transparent way for organisations to express interest in lawful partnerships.
These are proposed roles, not operational commitments. Their value would depend on governance, participant safeguards, qualified partners and measurable results. A useful bridge connects people to real opportunities without pretending to control the institutions responsible for them.
10. Digital construction tools can support skills mapping
A project schedule already describes activities and timing. A BIM model may help explain what will be built and how disciplines interact. A controlled skills register could link broad learning opportunities to work packages without embedding personal profiles in the model.
For example, a renovation package might be classified as suitable for supervised observation, an approved apprenticeship task or expert-only activity. Competent professionals would make and review those classifications as risks and sequencing change. An AI model should not turn them into automatic safety approvals.
A dashboard could show planned opportunities, verified apprentice hours, completed learning milestones and unresolved supervision issues. It should remain a management aid rather than a substitute for original records, training documentation or contract administration.
Technology need not be elaborate. A well-controlled spreadsheet may be more reliable than an ambitious platform with unclear permissions and weak data quality.
The guiding principle is traceability before automation. Each reported outcome should have a responsible owner, a source and an agreed definition. Without these basics, an attractive dashboard can amplify confusion rather than create accountability.
11. Inclusion must not become exploitation
Participants need to understand whether an opportunity is paid work, formal training, volunteering or an informational event. The terms should be communicated before they commit. Nobody should be encouraged to perform productive construction work without the employment, safety and compensation arrangements that apply.
The ILO's quality-apprenticeship framework emphasises structured learning, appropriate compensation, safe conditions and social dialogue. Norwegian educational guidance also makes clear that apprentices are workers in their training enterprises, not simply visitors gaining exposure. [6, 7]
Inclusion requires practical accessibility. Information should be understandable, and participation should not assume equal digital confidence, transport access or free time. Organisers should ask participants which barriers matter rather than inventing a generic disadvantaged-person profile.
Safeguards are especially important for younger participants and anyone entering hazardous workplaces. Site access, insurance, supervision and the distinction between observation and work must be addressed by competent parties.
A good programme creates a genuine route toward capability and dignity. It does not use people's aspirations to produce inexpensive labour or attractive publicity.
12. Safety and professional responsibility are non-negotiable
Construction learning takes place in a domain where errors can have physical consequences. A project should never trade away safety or professional responsibility to improve social-impact numbers.
Work at height, electrical installations, structural alterations, lifting operations, hazardous substances and occupied-building interfaces may require specific qualifications, controls or authorisations. The appropriate arrangements depend on the task and applicable rules.
A learning opportunity should be rejected or redesigned if it cannot be supervised safely. The same applies when a contractor lacks the capacity to provide meaningful instruction. A nominal placement with no appropriate training is not a satisfactory substitute.
Design decisions remain with responsible professionals. A mentor may explain a fire strategy without becoming the appointed fire-safety designer. A student may model a wall assembly without creating an approved construction detail.
These boundaries should appear in the skills brief, induction material and participant information. They make collaboration more useful by clarifying what people may contribute and where expert review is required.
A mature programme treats a decision to stop an unsafe activity as evidence of responsible management, not a failure to meet an aspirational target.
13. Data protection belongs in the trust architecture
Connecting people to opportunities may involve names, contact details, work histories, qualifications and sensitive circumstances. Collecting information creates responsibilities.
The European Data Protection Board identifies lawfulness, purpose limitation, data minimisation, accuracy, storage limitation, security and accountability as central GDPR principles. The European Commission likewise explains that organisations need an appropriate legal basis and should process only information necessary for a defined purpose. [9]
A pilot should not create a broad database of people's personal difficulties merely because that might support its social mission. When a partner needs to verify a qualification, the process should establish who is authorised to do so and what evidence must be retained.
Public reporting should generally use aggregated results and carefully reviewed case studies. Photographs and testimonials need suitable permissions and safeguards. Consent for publicity is not a substitute for the legal basis governing operational records.
Employers, training providers, contracting authorities and community intermediaries may have different data-protection responsibilities. These must be assessed before a real pilot begins.
Trust grows when participants understand who holds their information, why it is needed and how errors can be corrected.
14. Measure outputs, outcomes and aspirations separately
Outputs are countable activities: eligible apprentice hours, mentoring sessions, information events or learning modules. Outcomes are meaningful changes, such as demonstrated competence, completion of a recognised training stage or a confirmed transition to another opportunity. Aspirations are broader goals such as stronger communities.
A credible report includes denominators and definitions. 'Twelve participants attended' needs a definition of attendance. 'Ten per cent apprentice work' requires an eligible-hours calculation.
Indicators might include applicable requirements met, verified hours, completed training activities, documented referrals and participant feedback. Quality indicators should examine supervision, safety concerns, complaints and unresolved issues.
A complaint is not automatically programme failure; a functioning reporting channel may reveal problems otherwise hidden. No single score captures all dimensions. A small set of defined indicators and an auditable record is more useful than an unsupported social-value headline.
15. Beware of double counting and invented impact
A project can generate an opportunity, an employer can provide it and a community intermediary can help someone discover it. If all three organisations claim to have created the same job, the combined impact report becomes misleading.
Define attribution before the pilot begins. A contracting authority can report compliance; an employer can report training delivered; a referral partner can report an introduction. These are related contributions, not identical outcomes.
Likewise, someone attending three workshops should not automatically be counted as three unique participants. A mentoring conversation is not a completed apprenticeship.
Claims about additionality need particular care. If an apprentice was already employed and scheduled to work on a project, the project should not automatically claim it created a new apprenticeship place. It may report participation, but the claim that the opportunity would not otherwise exist requires evidence.
Converting volunteer hours into a large financial impact figure can also create false precision. Any valuation needs transparent assumptions, appropriate boundaries and independent review.
Good reporting is not a competition to produce the largest number. It is a method for learning what the programme actually achieved.
16. Clear ownership matters more than new bureaucracy
The public client and contractor have different responsibilities. The client defines the procurement, applies the law, specifies relevant conditions and monitors performance. The contractor plans and delivers the works, manages its workforce, supplies records and fulfils the contract. Training enterprises and educational authorities retain their statutory roles.
A community organisation can assist with outreach only within a properly defined arrangement. It cannot certify apprenticeship compliance simply because it has spoken with participants.
Every deliverable needs an owner. Who checks hour records? Who investigates discrepancies? Who validates training arrangements? Who approves public reporting? Who handles participant complaints? What happens when an activity conflicts with safety, schedule or legal requirements?
An effective process answers these questions before the first public announcement. It should also allow correction of inaccurate reports and preserve original evidence.
The objective is not to create another layer of promotional administration. It is to make the delivery system more capable of producing legitimate social outcomes and demonstrating them without ambiguity.
17. A pilot design for potential SamfunnsBro partners
A sensible first experiment would be a small, non-exclusive, non-procurement-interfering pilot focused on information and pathways rather than guaranteed placements.
First, identify a suitable educational partner, willing employers and a genuine community need. Define the purpose, responsibilities, accessibility measures and data practices. No construction-site activity should begin without appropriate employer and safety arrangements.
Second, prepare verified resources explaining apprenticeships, approved training enterprises, the difference between paid work and volunteering, and where authoritative information can be found. Relevant experts should review and date the material.
Third, consider a public information session and a limited number of career conversations. Any introductions to employers should be transparent and voluntary. Participants must understand that an introduction is not a job offer or training contract.
Fourth, evaluate what happened. Were resources understandable? Did participants receive useful next steps? Were referrals completed? Did privacy, safety or fairness concerns arise?
Only after a documented pilot should partners consider more complex integrations with actual construction projects. If public procurement is involved, the authority must independently approve the lawful role of intermediaries.
Numerical promises about jobs or apprenticeships would be premature. Targets should follow an evidence-based assessment of capacity and demand.
18. AI may assist, but it must not approve people or projects
AI can help draft accessible explanations, translate public information, organise resource libraries and suggest questions for authorised training providers. It may also flag missing documentation for human review. These are possible uses, not claims that Jarlhalla has deployed them.
An assistant should not invent qualifications, disclose personal information, make unsupported eligibility decisions or claim someone has been approved for a regulated role. It should not silently convert a general interest in construction into a profile shared with employers.
Participants should understand the role of AI and have access to a human contact. Outputs affecting employment, education, safety or contractual rights require appropriate professional review.
A narrow assistant that points to authoritative sources and acknowledges uncertainty is a better starting point than unrestricted automation.
19. Questions to ask before declaring success
Before presenting results publicly, the client should ask whether the activity was actually delivered, whether the participants understood their roles and whether the evidence can be independently checked. An attractive photograph, testimonial or project announcement cannot substitute for documented performance.
The project team should also examine who did not participate. Were the information channels accessible to people outside existing professional networks? Were opportunities described clearly enough for newcomers? Did scheduling, travel or language requirements exclude potential participants? These questions should be asked without collecting unnecessary personal information.
Another test concerns durability. A single workshop may be valuable as an introduction, but its benefit is different from completing a training module or progressing toward a recognised qualification. The report should not collapse those outcomes into one measure.
Finally, the team should publish limitations alongside successes. If a target was missed because an appropriate supervisor was unavailable, the reason should be recorded. If an apprenticeship requirement did not apply, the project should not imply that it did. If a referral produced no confirmed next step, the report should say so.
The aim is not to create a perfect story. It is to establish a repeatable process that earns trust through accuracy, accountability and learning.
Conclusion: the building is finished, but learning can remain
The social legacy of a construction project need not be accidental. Public procurement provides ways to consider legitimate social outcomes, while Norway's apprenticeship rules show how training obligations can form part of contractual delivery when the relevant conditions apply.
Yet the strongest lesson is not about a percentage. It is about the difference between announcing an opportunity and making it real. Quality training requires supervision, fair terms and evidence. Inclusion requires accessible pathways. Public procurement requires transparent, proportionate and defensible decisions.
A future SamfunnsBro partnership could connect these elements through mentoring, knowledge-sharing and responsible introductions. That possibility deserves a governed pilot, not a premature success story.
A completed building is visible. A person leaving a project with stronger, recognised skills may be part of its enduring human legacy. Both deserve to be designed, protected and evaluated with the same seriousness.
References
- European Commission, Buying Social (2021): https://op.europa.eu/en/publication-detail/-/publication/47c69b3a-cfcf-11eb-ac72-01aa75ed71a1
- European Commission, Social procurement, accessed 23 October 2026: https://commission.europa.eu/funding-and-tenders/tools-public-buyers/social-procurement_en
- European Union, Directive 2014/24/EU, Article 70 (2014): https://eur-lex.europa.eu/legal-content/EN/ALL/?uri=celex%3A32014L0024
- DFØ, Veileder til anskaffelsesforskriften: Tildelingskriterier, accessed 23 October 2026: https://www.anskaffelser.no/veileder-til-anskaffelsesforskriften/26-tildelingskriterier
- DFØ, Veileder om krav om bruk av lærlinger i offentlige kontrakter, accessed 23 October 2026: https://www.anskaffelser.no/maler-veiledere-eksempler-og-andre-verktoy/veiledere/veileder-om-krav-om-bruk-av-laerlinger-i-offentlige-kontrakter
- Utdanningsdirektoratet, Lærling, updated 5 March 2026: https://www.udir.no/regelverk-og-tilsyn/skole-og-opplaring/retten-til-videregaende-opplaring/larling/
- International Labour Organization, Quality Apprenticeships Recommendation, 2023 (No. 208): https://www.ilo.org/resource/other/r208-quality-apprenticeships-recommendation-2023
- Jarlhalla Group, SamfunnsBro, official concept page, accessed 23 October 2026: https://jarlhalla.com/samfunnsbro/
- European Data Protection Board, Basic principles, accessed 23 October 2026: https://www.edpb.europa.eu/topics/key-gdpr-concepts/basic-principles_en
- DFØ, Sosial bærekraft innen bygg og anlegg, accessed 23 October 2026: https://www.anskaffelser.no/kategorispesifik-veiledning/bygg-anlegg-og-eiendom-bae/sosial-baerekraft-innen-bygg-og-anlegg
- DFØ, Veileder til anskaffelsesforskriften: Kontraktsvilkår, accessed 23 October 2026: https://www.anskaffelser.no/veileder-til-anskaffelsesforskriften/27-kontraktsvilkar