Nature Survey, Spring 2025 – Report 2: Those in Paid Roles

1. This document presents a second report from the Wales Nature Survey undertaken over the Spring of 2025. The first report provided summary data for all respondents. This second report focuses only on those who said they had paid role(s) i.e. those who answered that they only worked in paid roles or that they ‘had both a paid and an unpaid/voluntary role.’
2. Before the survey was developed, anecdotal evidence pointed to risks such as knowledge loss, skills gaps, and a lack of diversity in the current nature workforce.  The Nature Workforce Survey was developed to address the lack of formal data, to collate baseline data to inform policy and strategic workforce planning. The survey is a first step in understanding the nature workforce.
3. This summary provides a snapshot of the status of the nature workforce in Wales; the survey was open from 18 March to 18^th August 2025 and received 1,180 valid responses from both paid and unpaid individuals. Anyone who thought they ‘’contributed to nature recovery in Wales’’ could respond. Of the 1,180 responses, 745 people reported that they undertook paid work (with, or without an additional unpaid role(s).
4. However not all of them answered every subsequent question, so the number of respondents varies. The number of responses (out of a possible 745), to each question is given. In two questions, respondents could choose up to three statements, the total number of choices (out of a possible 2,340) is given for these questions.
5. The findings presented should be interpreted with caution, as they reflect only the views of individuals who voluntarily chose to participate. Participation was open to paid staff and volunteers across all roles and sectors in Wales, and responses to individual questions were optional. To maintain confidentiality, any data cells with fewer than five responses have been excluded. A small number of respondents from the education sector were omitted due to insufficient representation. The data are subject to response bias and should be viewed as indicative rather than representative of the entire sector. Any use or presentation of these results must be clearly prefaced with the phrase: “of those who responded.”
6. Data tables are in Annex A. 

Acknowledgements

7. Welsh Government thanks all those that helped design and pilot the survey and all those who responded. 

Next steps 

8. The next report will look at those who undertake voluntary work. The analysis will seek to identify commonality/difference in views or experiences by different demographics and characteristics. 

Paid roles: by work pattern (745 respondents)

Of the 745 people who reported that they undertook paid work, 336 (43%) said they work full-time with all their time dedicated to nature recovery. Another 214 (14%) indicated they work full-time, but only part of that time is spent on nature recovery. Meanwhile, 113 (27%) work part-time with full dedication to nature, and 82 (11%) work part-time only partially on nature.

Paid Roles and voluntary work (745 respondents)

Overall, 60% of all those with any paid role also undertook voluntary/unpaid work for nature recovery. Only 37% of those in paid roles didn’t do voluntary/unpaid work.   

Of the 336 people who reported that all their paid work was nature recovery 56% said they also undertook voluntary work. Of the 113 people who reported that all their part- paid work was nature recovery, 70% said they also undertook voluntary work. 

214 people reported that only some of their full-time paid work was on nature recovery and 60% of them also did some voluntary work. 82 people reported that only part of their part- paid work was on nature recovery, but 71% of them also did voluntary work for nature. 

Paid roles: by duration of funding (734)

Out of 734 respondents, 50% said their roles were not time limited. However, 13% reported that their funding was limited to 1–12 months, with another 9% under 2 years. So, 21% of the workforce were in roles with less than 2 years funding. A quarter (28%) have funding of less than 4 years. But another 17% described their funding as complex and did not fit any of the given categories.

Paid roles: by employer type (715 respondents)

Among various employer types, the proportion of paid staff also undertaking unpaid roles varied significantly. Of the paid employees in national government, local authorities and regulatory functions, about half also did unpaid roles (45%,53% and 52% respectively). But of the 262 in the third sector in paid roles with a nature focus,  65% also undertook unpaid roles.   

Paid roles: time contributed to nature recovery (736 respondents)

Of the 736 in all paid roles, 22% said they contributed more than 40 hours a week, but this varied from 11% of those with just a paid role and 28% who had undertaken a paid and a voluntary role.  31% said they contributed more than 33 and less than 40 hours a week, but this varied from 43% of those with just a paid role and 24% who had undertaken a paid and a voluntary role. 

Paid roles: by which Local Authority they work in (625 respondents)

Of the 625 people that responded, Pembrokeshire had the highest number of responses from those in paid roles with 113 responses or 18% of the total. 80 in paid roles said they worked in Gwynedd (13%), with 71 (11%) working in Powys.  55 staff (9%) said they worked in Ceredigion. No one said they had paid work in Merthyr Tydfil and just 26 (4%) paid staff said they worked in Cardiff, which is lower than expected given the size of the local authority and the number of head offices located in the city. 

Across local authority areas, the balance between those who only did a paid role and those that did paid and unpaid roles varied. But care should be taken when only small numbers of people responded. When looking at those areas with the highest numbers of responses the split is approximately 60% paid only and about 40% paid and unpaid (Carmarthen, Ceredigion, Gwynedd, Pembrokeshire), this is reflected in Wales figures of 61% and 39%.

Paid roles: by local and national scope (744) 

Of the 744 in all paid roles, 47% said they spent most of their time working across multiple local authority areas, or pan-Wales or also outside Wales.  By contrast, of those that responded another 30% said they worked in one local authority area in Wales. Only 3%, just 20 people, spend most of their time working outside Wales. There was no real difference between just paid or those with paid and voluntary roles in these aspects.

However, whilst 21% said they mostly worked in one local authority area, but also did some work in multi area/pan Wales/outside Wales; this varied from 15% of those with just a paid role and 24% who had undertaken a paid and a voluntary role.

Paid roles only: training in the last 12 months (745 respondents)

Of the 745 paid staff who responded, 426 (57%) have undertaken training in the past 12 months. 20% had training between 13 months and 5 years ago. Only 9% of respondents did not undertake any training. 

Of those who had both paid and unpaid roles, 7% had no training, whilst 57% had training within the last 12 months. For those with only paid roles 11% had no training, but 58% had training within the last 12 months.

Paid roles: training hours in the last 12 months (746 respondents)

When looking at training in the last twelve months for those in paid roles, Of the 746 people in paid roles who responded, 20% had not undertaken any training. Meanwhile, 43% completed one day of training, and another 25% up to two days. A further 24% had up to five days, and 14% had more than five days of training. A slightly higher proportion of people in paid and unpaid roles longer training hours, compared with those in paid roles only who were in slightly shorter hours. 

Paid roles: current studies/training (399 respondents)

399 people in paid roles said they were engaged in studies related to nature. Of those studying or training, 70% were in work-based learning, higher education was pursued by 16% and further education by 12%.  Only 1% of all respondents, just five people, were in apprenticeships. 

Paid roles: by highest qualification level (745 respondents)

Caution should be taken when interpreting the response to the question about the highest level of qualification. Respondents were asked to identify their NVQ level and were provided a ‘look up table’ to equate their qualifications and NVQ levels. 

Of those that responded 49% identified NVQ Levels 6,7 or 8 (degree, postgraduate or above). Other than that, only 10% of respondents specifically identified any other NVQ level whereas 34% said ‘none of the above’ or ‘other’.  The results indicate that about half the workforce has a degree or above, but below degree level,  respondents were probably not able to identify their NVQ level. 

Paid roles and work-based specialisms (1,829 choices)

Of the paid staff who answered this question, each chose up to three specialisms or work areas resulting in 1,829 separate choices.  The most popular choice was ecologist, with 225 choices or 12%. Of these 74% did both paid and unpaid work.   The next popular choice was ‘surveys and fieldwork’, with 175 choices or 10%. Of these 73% did both paid and unpaid work.   Countryside ranger/biodiversity/ conservation officer/ was chosen 163 times and practical land-based work with 149 choices; for both over 60% did both paid and unpaid work.

Paid roles and perception of challenges facing the workforce (2,133 choices)

Of the paid staff who answered, each choose up to three challenges facing the nature workforce resulting in 2,133 separate choices. 

The proportion of selections for each choice were broadly similar for those in paid roles only and those in paid and unpaid roles, i.e. views were similar across the two groups. 

Paid staff by age (721) 

Overall, 66% of those who have a paid role are over 45 years old (with only 34% under 45 years old). Of these, 22% are already over 65 years old, whilst another 21% are between 56 and 65 years old.   

Paid staff: by gender identity (741 respondents)

Of the 741 paid staff who answered, 56% identified as female, and 41% as male, 3% identified as other or preferred not to say. 

Paid staff with physical or mental health conditions or illnesses lasting or expected to last for 12 months or more (736 respondents)

Among paid respondents, 567 (78%) reported no physical or mental health conditions or illnesses lasting or expected to last for 12 months or more. 153 (19%) said they had such a condition.  The proportions were very similar for those with only paid roles and those with both paid and unpaid roles. 

Paid staff: by Welsh language capability (742 respondents)

Of the 742 respondents in paid roles, 97 (13%) said they were fluent and another 11% said they could speak a fair amount.  50 (7%) said they could not speak Welsh.  The remaining 69% could either speak a little Welsh or could just say a few words. There was no real difference between those only in paid roles and those that had an unpaid role as well. 

Paid roles: by ability to perform role in Welsh (749 respondents)

Of the 745 in paid roles who responded, 171 respondents (23%) felt capable of performing their role in Welsh, while 571 (76%) said they could not.

Paid roles: by ethnicity (734 respondents)

Of the 734 respondents in paid roles who replied, 719 (98%) described their ethnicity as White. 15 (2%) respondents identified as Black, Asian or Minority Ethnic. Care should always be taken when considering small numbers, but there was a slightly higher proportion of people identifying as a Black Asian and Minority Ethnic person undertaking both paid and unpaid roles (2%), compared with those just doing paid roles (1%), but the difference in numbers is very small. 

Respondents with paid roles: by roles type (745 respondents)

Respondents with paid roles: percentage undertaking paid only, or paid and unpaid roles

Respondents with paid roles: percentage undertaking paid only, or paid and unpaid roles

Paid role(s) in nature recovery by duration of funding (734 respondents)

Percentage of staff in paid roles: by employer type (726 respondents)

Taken together, both your paid and unpaid role(s), how much time do you estimate that you contribute to restoring or enhancing nature in Wales? (736 respondents)

Paid staff: by local authority (625 respondents)

Paid roles: by local authority (625 respondents)

Paid roles: by local and national scope (749 respondents) 

Have you undertaken training relevant to your role (s) in nature (paid and/ or unpaid)? (745 respondents)

In total, how much training have you undertaken in the last 12 months, relevant to your role(s) in nature (paid and/or unpaid)? (746 respondents)

What best describes your current studies/training, related to nature recovery? (399 respondents)

What is your current highest level of qualification? (NVQ levels) 
Find out more about NVQ levels on GOV.UK. (1)  (745 respondents).

Paid roles, respondents, by specialism (each respondent could select up to three areas of work, 1829 choices)

Paid roles: the most significant barriers facing the nature workforce in Wales. Each respondent chose up to three statements (2,133 choices)

Paid roles by gender (741 respondents) 

Paid roles; do you have any physical or mental health conditions or illnesses lasting or expected to last for 12 months or more? (736 respondents)

Paid roles: which of the following best describes your ability to speak Welsh? (742 respondents)

Paid roles: I am capable of performing my role in nature recovery, in Welsh (749 respondents)

Paid roles by age (721 respondents) 

Paid roles and ethnic identity (734 respondents)

Survey Approach and Design

Introduction

1. The survey was voluntary and self selecting. No definition of the "nature workforce" was applied, due to the diversity of roles and lack of comprehensive data. The survey was designed to be inclusive and self-selecting, including anyone who believed they contributed to restoring nature. The approach acknowledged the complexity of the sector, including full-time, part-time, and project-based roles, as well as paid unpaid contributions. The survey was open to individuals living or working in Wales, including those in cross-border or remote roles. It also included those in policy, regulation, enforcement, and operational delivery roles.
2. The survey was developed electronically using Smartsurvey, with input from stakeholders and tested through two small pilots. It was bilingual (Welsh and English), with respondents choosing their preferred language. Respondents were invited to join a stakeholder list for future engagement. Participation was voluntary, and data were stored separately to maintain confidentiality.

Information on those targeted: the nature workforce

3. The survey intentionally avoided any rigid definition of the nature workforce, the inclusive, self-selecting approach reflected the complexity and diversity of roles across sectors. The survey was completed by anyone who thought they were contributing to restoring nature. It was a self-selection process. There was no attempt to define the nature workforce. There were known to be a myriad of roles, with some working full time on nature, others part-time, funded by a range of projects. It was a very complex picture and as a result it was unlikely that all organisations/businesses could be identified. It was considered that larger organisations would find it difficult to accurately report on the nature workforce. Respondents were asked to complete the survey only once, regardless of how many roles they held to avoid duplication.
   1. Scope The survey tried to reach practitioners doing nature recovery on the ground, but also those in policy, regulation or enforcement. Anyone who lives, or works in Wales, acknowledges the realities of cross-border working and remote roles. This flexible definition was appropriate and helped ensure that contributors to Welsh nature recovery were not excluded.
      1. Unpaid/volunteering and paid roles, the survey design allowed individuals to report on both paid and unpaid contributions, recognising the importance of volunteering in the sector.
      2. Job roles: survey includes full-time, part-time, and project-based roles. The survey was intended to reach a wide range of contributors, including land-based roles, practitioners and volunteer management. But also, decision makers (e.g. elected members, CEOs), those who contributed to delivering nature projects; people working in communications, volunteer management, finance, fundraising, project management. This breadth reflected the real-world complexity of nature recovery work. Although the survey was not aimed at internal Welsh Government or Senedd staff, the self-selecting nature meant that some may have participated.
      3. Volunteering: The data collection included unpaid workers. Volunteering forms a key aspect of delivery for nature. But each person only filled it in once, irrespective of the number of roles. So, an individual who worked for a private contractor doing peat bog restoration, volunteered for their local nature group and was also a community councillor overseeing nature projects locally, filled it in once.
4. Workforce Information collected: 21 questions were developed to cover a range of areas of interest to future workforce development and planning.
   1. Demographics; age, gender, race, disability, Welsh language ability.
      1. Workforce information; full/parttime, funding source, tenure of post, category of employer, location, qualifications and training.
      2. Nature knowledge; broad role, areas of expertise. Highest qualifications, recent training etc.
      3. Perception of current workforce planning, issues and priorities.
5. Stakeholder list. Respondents were also invited to join a stakeholder list for future engagement, to build a direct communication channel with the workforce and reduce the reliance on indirect distribution methods. Participation was voluntary. Recipients were asked to agree to be on the stakeholder list and agree that their email address be held for 5 years. The intention was to create a contact list to contact the nature workforce directly, e.g. more detailed surveys on, for example, qualifications and skills. The data were held in a separate survey and in a separate database. 

Survey design process

6. Smartsurvey: Not least due to resource constraints, but mainly to try and reach as many people as possible Smartsurvey was used. It is an electronic survey tool used by the Welsh Government. It was relatively easy to use to create the survey. The routing function ensured respondents were only asked questions that were relevant to them. The entirely electronic survey was developed with stakeholder input and two small pilots were undertaken to test methodology and questions.
7. On-line only: The survey was available to complete online, including on a suitable mobile phone. It was sent out via email and crucially, people could forward it on to their contacts, extending the reach. It was assumed that anyone in paid work had access to ICT. It was possible that some volunteers did not have independent access to ICT, but they would volunteer for an organisation with ICT. Online was a significant benefit but also had possible limitations due to ICT coverage and literacy.
8. Bilingual: The survey was bilingual. Each person chose the language of response. However, regarding some questions, if the user required further information, they were directed away from the survey, to the .gov website for more information. The .gov website was in English only in compliance with Welsh Government guidance.

Method of distribution

9. Timeline: the survey was launched on 18 March. Most responses were received in the first 8 weeks. Following the final communication about the survey closing, it was formally closed on 18th August.
10. Distribution: The survey was distributed via email to known contacts with a request to cascade it further i.e those with an email address. This was intended to reach a wide audience of individuals. This was a known trade-off of the distribution method. This snowball sampling approach relied on indirect distribution and was intended to reach individuals beyond the Welsh Government’s direct contact list. However, self-selection meant that some employees or volunteers did not complete it.
11. The contacts included ENGOS, umbrella organisations e.g WCVA etc local government and statutory organisations e.g NRW. These organisations included those engaged in delivery, grant recipients, those engaged in policy development etc. Recipients were then asked to forward it to anyone they think might consider themselves part of the nature workforce. This was particularly important for contractors working on nature projects, for whom contact details were not known. The methodology was reliant on those who procure services to send the survey to the contractors they use.
12. A survey of employers was not used due to the dispersed nature of the workforce and the potential burden on HR departments, especially in large organisations (who might not have had information about funding routes, or the latest qualifications and training). Also because of the potential burden of completing the survey, again for large organisations with many employees.
13. Duplication: The method acknowledged the risk of duplication, because individuals may have received the survey multiple times. No complaints were received.
14. Publicity: staff in CCRA publicised the survey at stakeholder meetings and conferences and included the link in the email signature. Deputy First Minister issued a press notice announcing the data collection. Given the true population was not known it was unclear how successful it was, save to say, over 1,000 received and completed it.

Sample/response rate

15. Response rate. The response rate could not be predicted or calculated after the survey due to a lack of knowledge about the number of people in paid or voluntary work for nature. The final dataset included 1,180 responses.
16. Participant drop-out: (i.e. attrition) was not known. It was not known if overall response and data quality would have been better with an electronic, or a manual survey, or using HR Departments. The nature workforce was a highly connected one and there were several groups and forums which meant it could be circulated by email, rather than knowing a contact address or organization. But the electronic survey relied on access to ICT via computer or mobile phone. This is a weakness and potential limitation, although for the paid workforce it is unlikely that many would be in employment and be without access to ICT.
17. Manual surveys would have needed to be distributed to an unknown and highly dispersed group (without contact details) and then returned by post. Both provide a significant hurdle to participation. HR Departments may not have known who was in the nature workforce and/or may not have had up to date information.  The methodology appears to have been successful with over 1,000 responses.
18. Not everyone answered every question. There were 1,180 respondents. But for example, only 780 respondents indicated they were in paid work. Whilst all of them answered the question about full-time or part-time status. But only 740 of them answered the question about their paid role.
19. Valid Responses/Uncompleted returns: Only responses deemed valid by SMARTSURVEY were included; 649 incomplete responses were excluded, primarily because respondents did not reach the end of the survey and therefore did not consent to data use. Of the uncompleted returns it was not known how many of those people ‘restarted’ and went on to complete a return. It is also possible that some opened the survey and realised it was ‘Wales only’ and that it was not applicable to them. Others may not have been able to use the technology for other reasons.

Limitations of the approach used

20. To summarize the above, several limitations were identified:
21. Integrity and Trust: Some engaged with the survey design and pilot expressed skepticism about respondents’ confidence in providing information to the Welsh Government and data confidentiality. The survey underwent extensive survey control processes and review and was approved by Welsh Government Knowledge and Analytical Services (KAS). Over 1,000 people did respond.
22. Low or Patchy Response: The self-selecting nature and indirect distribution would seem to have led to uneven participation by geographic area e.g Pembrokeshire.
23. Use of an on-line survey and Smartsurvey: an on-line survey may have reduced participation. A small pilot was conducted to test the survey and technology. It had been used by other departments, but possibly not using the snowball methodology. Communications colleagues were involved in ensuring the survey met guidelines. It was a survey of the workforce, and it would be expected that respondents would have access to computers and mobile phones. But the survey’s digital format might have excluded individuals without internet access, or possibly more likely, digital literacy. It might have increased the likelihood of partial completions. But it was not known how responses rates or quality would have been affected by being collated manually and/or through HR departments.
24. Duplication: Multiple survey invitations could not be prevented, and respondents were asked to complete the survey only once. The survey was intended to be circulated as widely as possible, to capture as many people as possible. In the introduction to the survey, the issue was highlighted and respondents asked to only complete it once.
25. Snowball Sampling: This method introduced some uncertainty about who received the survey and potential bias in responses.
26. Data checking: Smartsurvey results were checked manually. In addition, all information produced by Co-Pilot was double checked. 

Analysis approaches (including AI tools and quality checks)

27. The analysis was conducted using Smartsurvey. Data were analysed and reported using Excel, Smartsurvey and Microsoft Co-Pilot. The results were only about those who responded, nothing was recorded about those that did not respond. 

Quality assurance process

28. Question by question analysis by work type, was initially produced using Smartsurvey pivot tables using question 1 to identify work type. This was a simple process, with automatically generated tables. Each report was put into excel and modified to only report on those in paid roles. Excel was used to create a very simple summary, identifying totals for each question. Often tables were transposed to create more user-friendly output.
29. Where the number of responses was less than 5 people in any one ‘cell’ of a table, the results are taken out of the table entirely to protect confidentiality. This meant the ‘control total’ was often slightly lower than 740.
30. Quality assurance was a key aspect of the analysis process.
    1. Figures were rounded to the nearest whole number to avoid a false perception of precision.
    2. It was noted that in questions where the respondent could only provide one response, answers might not add to 100% (i.e. might be slightly more or less)
    3. A download of the base data was generated by Smartsurvey and put into Excel for sense checking. As expected, the number of rows was 1,180. Sense checking, including using estimation were very important tools. The data produced could not just be assumed to be correct, spotting any errors and anomalies was important.
    4. Anomalies were investigated and confirmed to reflect actual responses. The responses were not ‘incorrect’ they reflect what respondents said, but they were of note in terms of interpretation and use of the data.
       1. Geographic distribution, by local authorities, did not follow population in Wales.
       2. Qualifications data: 40% reported an NVQ level 7 but most other respondents did not choose a specific qualification level. It is likely that NVQs levels were not understood.
    5. Respondents could skip questions, and routing logic, meant the base number was different for most questions. Data were checked to ensure they reflected those that responded as shown in the table below.

KAS assurance process 

31. Once pivot tables were produced.  An Excel original spreadsheet was created.
    1. Check the number of ‘paid’ responses was 740.
    2. Only show cells with more than 5 respondents.  
    3. Convert results to a percentage based on the new control totals.

Report writing

32. A summary report of the data was written. A methodology and lesson learn section were included. 

Data publication

33. The survey was anonymous. Confidentiality was always a concern to recipients, and the stakeholder list information was kept separately from the survey responses. The survey itself and privacy notice were discussed at length with KAS colleagues. Figures were rounded to the nearest whole number, because including decimal places might provide a false sense of accuracy or precision, the data were only of those who responded. Whilst very few people from black and ethnic minority communities had responded it was, however, crucial to begin to tackle the issues of inclusion to be able identify this issue, albeit with small numbers.
34. The report, this detailed methodology and lesson learnt will be published bilingually on the internet. All those that provide a contact email address would be sent the report.
35. This initial report is intended to be the first of several reports on the data, more would follow on different aspects of the survey, such as cross tabulation, reports the paid   and unpaid workforce etc. Each will be made public.

Early reflections on the methodology, risks and lessons learnt

1. Direct-to-Individual Approach
   The decision to bypass traditional employer-based/organisation based surveys and engage individuals instead directly was a success.
2. There was a strong response of 1,180participants, paid and unpaid; a far higher response than anticipated.
   1. for the first time data reflects individuals. Each person was only counted once, even if they worked/volunteered for several organisations.
   2. HR departments did not have to compiledata.
   3. Contact details from participants have been collated.
3. However, some reflections are.
   1. a further 647 responses classified as invalid. They require further analysis; some were entirely empty/bank others have one or two questions answered.
   2. response was uneven. Some geographic areas had very few responses, while others—particularly Pembrokeshire—seemingly overrepresented. Further detailed analysis may reveal greater insight.
   3. Participation was voluntary and some employers seem low as a whole; e.g despite Natural Resources Wales (NRW) employing approximately 2,000 staff and having a focus on nature, very few responses were received from this organisation.
   4. Some individuals received the survey multiple times, while others—particularly contractors and those not on known mailing lists—may not have received it at all. Still, the strategy relied heavily on goodwill and informal networks. As a result, some key groups were missed, and response levels varied widely between organisations and regions.
4. That said, overall, this confirmed the approach was effective in reaching a wide range of respondents. But it did not reach everyone (with gaps in reach to some organisations, employees, and geographic areas).

Survey design 

5. The bilingual electronic survey, developed with stakeholder input and distributed via email with a request to cascade further, enabled broad participation. The open invitation to anyone who considered themselves part of the nature recovery effort in Wales allowed for inclusivity across roles paid and unpaid.
6. But around 647 responses were incomplete/empty. This suggests that while the design was accessible, some respondents may have encountered barriers to completion—it is not known if they realised, they were not eligible or due to survey length, clarity, or other factors.
7. 100% response may never be achieved, but potentially future surveys may benefit from more targeted outreach and possibly improved survey design, for example for the qualifications data, explaining NVQ levels to encourage more balanced and complete participation.
8. Responses to the question on NVQ level appear to indicate the question was not well understood. The geographic distribution does appear to not reflect population.

Distribution and duplicate emails 

9. The method of allowing the survey to be forwarded and asking individuals to complete it only once proved to be a resource-efficient way to gather data. The simplicity of the approach contributed to the strong response rate.
   But the lack of control over who received the survey may have contributed to overrepresentation in some areas and underrepresentation in others.
10. While it was acknowledged that some individuals might receive the survey link multiple times, this was anticipated and addressed in the survey introduction. There is no indication that this caused confusion or deterred participation. 

Response, inclusivity and reach

11. This risk of low response did not materialise. The volume of responses exceeded expectations for a voluntary, self-selecting survey. While the data is not representative of the entire workforce, it provides a robust baseline and valuable insights into the nature recovery sector.
12. The open, self-selecting nature of the survey encouraged participation from a wide range of individuals, including those in non-traditional or multiple roles. This was a strength of the approach. It is not known if some perspectives were underrepresented. This limits the ability to generalise findings across the entire nature workforce. However, the low responses from some areas and incomplete responses reduced the usable dataset. Clearer guidance and/or improved survey design in future iterations to reduce drop-off may be needed.
13. The survey successfully gathered data on demographics, employment, volunteering, qualifications, training, and workforce perceptions. The inclusion of an agreement mechanism for future contact was also effective, laying the groundwork for ongoing engagement.
14. The volume and quality of completed responses provide a solid foundation for analysis; the data allow for basic tabulation and, in many cases, cross-tabulation by role, geography, demographics etc. But the patchy coverage means that some analytical opportunities were constrained. These limitations will be clearly acknowledged in any interpretation or presentation of the findings.

Use of technology 

15. The survey was a new initiative for the Climate Change and Rural Affairs (CCRA) portfolio. Despite initial uncertainty, the response rate was strong, with 1,180 individuals participating from across Wales. This suggests that the approach was effective in reaching a broad audience and generating meaningful engagement.
16. No significant issues were reported regarding the survey platform (SMARTSurvey) or question clarity. The pilot testing and involvement of communications colleagues appear to have been effective in ensuring accessibility and usability, but the number of incomplete/unusable responses warrants further investigation.

Integrity and trust

17. There were concerns that individuals might be sceptical about sharing information with Welsh Government. However, the high number of responses and the breadth of participation across sectors and roles indicate that these concerns were largely mitigated. The approval by survey control and proactive communication by staff helped build trust.

Conclusion

18. The methodology adopted for the Nature Workforce Wales Survey 2025 was broadly successful. It overcame many of the challenges posed by a dispersed and complex workforce and delivered a valuable dataset. However, the uneven response distribution, underrepresentation of key organisations, and the volume of incomplete responses highlight areas for improvement. Future surveys may benefit from targeted outreach, improved survey design, and additional support to ensure more balanced and complete participation.
19. Overall, the risks identified prior to the survey were well-managed, and most did not materialise to a degree that undermined the survey’s success. The approach proved to be a resource-efficient and inclusive method for gathering baseline data on the nature workforce in Wales. The results now provide a foundation for future workforce planning and policy development.