Are wages the fly in the Fairtrade ointment?

September 10, 2013

How are the emerging donors (from China to Azerbaijan) changing the aid business?

September 10, 2013

If high staff turnover is unavoidable, how should we redesign aid work to cope?

September 10, 2013
empty image
empty image

One of the implicit assumptions that often underlies programme design is that the people who initially come up with an idea and turn it into a project or programme then stick around and implement it.

The reality is often very different – high levels of staff turnover are almost universal in both NGOs and aid agencies, with serious consequences.groundhog day On the way over to Australia, I bumped into an Oxfam gender adviser in Afghanistan, who told me that typically expats work a couple of years, then leave, either through burn out or because they are on short term contracts. Then new arrivals start the learning process all over again (often including repeating the same approaches and discussions of the previous years – at worst, producing a Groundhog Day of ‘steep learning curves’, followed by loss of the accumulated knowledge. Then repeat.)

This problem doesn’t just apply to expats – a lot of local staff get a couple of years’ INGO experience under their belts, only to be lured away by better salaries working for the UN or government aid agencies, or the need to progress their careers elsewhere. Acting as a training institution for other, larger organizations may strengthen the system as a whole, but it doesn’t make it easy to run an effective programme.

So today’s essay question is, would our programmes look different if we assumed average staff turnover of, say, 2 years? Some thoughts:

Knowledge Management: nothing worse than coming into a project to find anything useful buried in a chaotic filing system, and much of the implicit knowledge (eg on how things actually work) not written down at all. How can we get better at capturing useful knowledge, and make it readily accessible? Should people’s departure include a week in a padded room without internet, to write up their experience? Make back to office reports after field trips more systematic and retrievable?

knowledgeOutsourcing the wisdom: an alternative is to find more stable sources of knowledge and insight – local academics, ex-aid workers who still live in the area, or (even better), make sure that ‘partnership’ includes a full briefing of incoming staff by communities or partner civil society organizations. In some cases it might even be worth setting up a network or organization to provide this kind of institutional/issue memory – imagine every incoming gender worker arriving in Kabul spent a week with half a dozen Afghans/expats with decades of combined experience in gender work. Even better if this was backed up with a serious bit of mentoring, with each new gender worker in Afghanistan able to pick the brains of someone who’d been through it before, or (as mentors may not be permanently on tap), set up a group of sages, who can receive and feed back on regular email updates from current staff. Think how much collective time that could save, allowing people to build on past knowledge, go further, stand on the shoulders of giants etc etc – wouldn’t that be worth funding?

Turnover makes it even more necessary to adopt working practices based on iteration and review, rather than designing the Big Plan and then implementing it – iteration (as in the power and change cycle) means each succeeding wave of workers helps design and improve the programme, and in the process understands it.

Any other suggestions?


  1. The primary issue that I identified during my work with INGOs is their lack of valuing “knowledge assets.” They have consistently failed to see knowledge as an asset like car, IT servers and so on. The current financial crisis in aid sector will be a good push factor to value knowledge as business asset and then protect and nurture it.

  2. great issue Duncan. We design things as if they are not operator dependent but they are! would be good to hunt around in the management literature for more on this issue – in medicine they are quite good. training practices and a high emphasis on formalising information (both about specific patients and procedures) is quite interesting, I think. The aim, of course, is that we don’t die when a new doctor comes on shift…

  3. Most expat contracts and many contracts for local staff are two year fixed term contracts, and it is not surprising if that sets a psychological expectation, and also if people start looking at other options before that two years expires. It would be interesting to know whether those organisations offering open-ended contracts experience a lower turnover.

  4. Hi Duncan,
    Thank you – I am so glad someone has written this piece. Much needed. 2 thoughts:

    (1) This is also the case for bilateral and multilateral aid agencies (beyond just the NGOs. Can we learn from DFID here / can they learn from the above? “2-year wonders” – as aid workers were called in East Africa when I was there – bedevil the development sector.

    (2) Beyond ‘better knowledge management’, is there also a more systemic management solution? EG can we ensure that projects have a life-span of two years, or at least a natural break-point after two years in sync with the staff turn-over? Can we have an ‘agile’ process for development projects where incoming project officers have the opportunity to influence project design without going back to the drawing board?

    best, Martin

  5. Hi Duncan

    One of the things I have tried (but failed!) to set up is using a case study template to capture project learning from the initial design stages.

    All too often a case study is an after thought at the end of a project – where the project manager has already left for the next job, and when the reasons why they set it up in the way they did is a dim and distant memory.

    If we are serious about our learning, we should be starting to build our case study while the actual decisions are being made – so we can document the factors influencing those decisions, and how decisions were taken. And importantly down the line why and how things changed (rather than pretending it was always set up that way!)

    By incorporating case study writing (through a simple template) into the project right from the start the case study essentially ‘writes itself’ over time, but could also be a useful tool for checking thinking and logic of the design outside of cumbersome proposals and log frames – in a few paragraphs does the logic make sense? This also leaves you with a useful document of what’s happened so far for anyone come into the project later on.

    By the end of the project all you need to do is fill in the results and you’ve got yourself a case study!

    Maybe one day I will get something like this set up – in the meantime it would be great to hear if others have been successful in anything similar.

  6. We have done two things which have been of great help to us. During the earthquake of 2005 we could see that our programme which had been running for over ten years in the Mansehra region of Pakistan would lose all its staff to the hundreds of international organisations that suddenly arrived in the area with three times the salary that we offered to their staff. We adopted a sliding scale of salary which would be increased during the period when the organisation would come under such stress and be comparable in some respects to the salaries that were being offered. This helped us because we hardly lost any staff and could change the salary structure according to the market situation. This was repeated in the floods of 2010.
    Secondly, the organisation has also invested in what we call core staff. These are important staff who have to be retained over a long period of time and are offered terms which encourage them to stay. This has helped us retain staff for over a decade in important positions.
    Thirdly, lets acknowledge that this problem can only be overcome by building local capacities. In this connection we also need to recognise that one of the most important capacities that need to be built in local organisation is the what is called the capacity for survival and sustainability. This means investing in the institution. There have been some very interesting example of this in Pakistan which has paid rich dividends in helping retaining staff and institutional memory

  7. High turnover is a serious challenge, and I agree with many of the suggestions here.

    But I don’t think we should be so quick to assume that high turnover is “unavoidable”. Shouldn’t we instead give more attention to finding ways of retaining staff for longer than two years?

  8. Very recognizable! I do not believe improving writing down ‘your lessons learnt’or describing cases will be the magic bullet. As the nice picture in your blog shows, a lot of our knowledge is pretty hidden; implicit, we sometimes don’t consider it important, or we think it is common sense. Ideally ofcourse the leaving sraff memebre has a week close working with the new staff member in which picking the brain of the leaving staff is crucial. Where this is not possible; let’s ensure conversation between leaving/new staff can be facilitated, leavers might have 2 days left of their contract to spend afterwards on Skypes/calls with their replacement. Or.. a suggestion would be to interview and tape (audio/video) every staff that leaves. OF course they had to make their archives and filing of reports in order, but now we try to get all information out by asking them anything else related to what they did (squeeze them out 😉 )

    Nancy Dixon posted some interesting stuff on this as well on her blog Conversation Matters

  9. These all see like ban-aids. Instead, let’s focus on the actual problem: staff retention.

    I, like many other NGO workers, work on a series of short-term contracts. In my case, I’m a policy advisor now on my 7th Oxfam contract in under two years! Oxfam and others should do a better job with long-term planning and resourcing. It’s vital for continuity and relationship building.

    In my opinion, the minimum policy contract should be 2 years.

  10. On this issue, I would ask where is the commitment from upper management to the human resource and program managers to implement policies and practices to keep good staff? I’ve seen harassed middle management and long-suffering HR “plugging holes” rather than being supported to strategically assess the skills of their staff to retain, develop and (re)deploy them.

    These organizations lacked management support to retain and develop staff. People were seen as expendable. In the bureaucracy, they were budget line items on spreadsheets. Field and technical staff were seen as only being competent in one skill set. Once their program ran its course, then they were let go along with their experience. When new programs needed staff, then the hamster hiring wheel would spin once again. Exigency spun the wheel rather than considered management planning.

    Field people were particularly vulnerable to feeling like they were floating from contract to contract. There was no sense of having a career, unless you were lucky enough to have a sinecure at the HQ level. Unsurprisingly, no commitment from the organization corresponded to an equal lack of loyalty to it. I’ve seen many people leave an organization because they felt they weren’t being valued for what they have done and, importantly, what they could do.

    This utilitarian approach to “human resources” is ultimately self-defeating. Making a strategic commitment (backed by resources) to keep good people would send a strong message across the organization and boost morale. This commitment is a recognition that the long term integrity of an organization is just as important as filling spaces. Without this, then everything else to address burn-out and turnover tend to be provisional work arounds.

  11. Encourage Oxfam to lead the way with contracts of a meaningful length – enabling the same type of role security that exists in head offices – so that staff and partners experience better continuity in programming

  12. Hi Duncan,
    Important issue, though I would like to highlight a silver lining.

    Perhaps one of the main role of NGOs is not delivering services themselves, but creating a consensus on what is important in development.

    The NGO as a breeding ground for development talent plays an important role in this.

    If I work with a government official with a MSF background, I normally expect a hands on approach, directed to the poor, with a soft spot on Health.

    So perhaps staff outplacement is an important way to grow influence with your principles and values, and even with your network. I used to work for a union who actively pursued such a policy (and explained the low pay to their staff).

    On the other hand, in the real world, only systems with enough redundancy work. In development we tend to believe in eliminating overlap or working as if there is no tomorrow, but this does not mean it is a good way to work.

    Perhaps, indeed, if we want to be professional and have impact, we should limit the scope and build in enough training and redundancy taking the staff mobility into account.

  13. Rings many bells.Seems to me:

    1 Nature of much development work is based on good networking and relationships. When staff move a lot of good will/informal contacts/Networks can go with them.However much you handover, some staff click with local networks others do not?

    2 Sadly often new staff, not so keen to learn from predecessors, often a “new broom” syndrome?

    3 Why not seek to appoint country Directors at least on 5 year contracts, and build a completion bonus into it.

    4 INGO’s are pretty bad in my experience at poaching capable local staff. So some self management/good practice amongst INGO’s required.

    5 In my experience donor staff often do lack grounded country based knowledge and Govt officials are good at playing them whilst their time limited contracts run down.

    6 Lastly, this merry go round perhaps reinforces fact that Outside agencies can never be that central to lasting development needs to be locally owned.

  14. This is a huge issue, even if turnover is not that high. My team (15 ppl) uses a private (invite only) google blog for most knowledge management. The ability to use tags and search is hugely helpful for finding things later. Everyone gets the posts delivered to them by email, so it functions as a list-serv with central, searchable storage. More on that here:
    Tools like dropbox and google drive also help a lot. We keep a google doc of potential ideas, contact info for who recommended them, etc.
    We are working on creating light-weight knowledge management tools for visits, meetings, etc. at BRAC and would be happy to discuss. Realistically no one will get the week that they need to sit in a padded room and write, but maybe we can get them to talk to a junior staff for 15 minutes a month, and that gets written up and stored thoughtfully.
    This project may interest you–documenting scale up real-time to develop these types of tools:

  15. Even if we improve the HR side of talent retention, we will always face the issue of staff leaving at one point and therefore the risk of losing institutional memory with them. So knowledge management is not a side issue here, but at the heart of the problem.

    In my view, it is a mistake to think of “capturing knowledge” as something you do at the end of an assignment (the tongue-in-cheek “padded room” you suggest ;). Most of this capturing will come too late, is out of date, not context-sensitive enough, and doesn’t allow for course correction of the project reflected on.

    Instead of envisioning a huge peak of knowledge management activity at the end of a project, we should strife for a moderate, but continuous flow of information throughout the project period, from start to finish. This could be done through open sharing of regular project logs, regular online discussions, or frequent personalized blog posts with impressions, reflections and learning points from different project stakeholders. It would be more a working mode of “working out load”, sharing what you are doing while you are doing it, rather than frantically trying to write down everything you know two years after the project started.

    The advantages of working out loud are many:

    – Experiences are still fresh, and therefore likely to be more accurate and relevant
    – By sharing learning points and reflections just-in-time we actually have the chance that they directly influence ongoing execution, therefore improving the project as we go
    – As Maria above points out, whatever is shared openly (e.g. through blogs, external or corporate social media or discussion forums) will be available as a record for whoever takes over the work

  16. Very interesting post, thanks for raising the issue.

    Some people already suggested investing on staff retention, particularly by extending contract length. However, these measures would probably reduce turnover, but not eliminate it altogether.

    People leave for a variety of reasons, not all of them related with contract duration. There is a high rate of staff turnover even among well paid, long-term UN staff (they do not quit their jobs, usually, but move to another assignment). It seems to me that, particularly within humanitarian agencies, there is an organizational culture that rewards people who have “been to places”, thus pushing staff to move from country to country.

    Moreover, extending contract duration may not always be possible: most aid agencies live on project funding: staff is recruited within a given project, and contacts expire when the project ends, unless a new project gets approved (by that time, however, some people might have left already).

    Therefore, finding ways to effectively sharing knowledge is absolutely necessary. We would all benefit from it, even if staff turnover was substantially reduced. I totally liked the example made by Johanna Macrae, doctors routinely pass information on their patients to the next colleague. Maybe they won’t need to intervene during their shift – but if they do, the system ensures they know what they need to know.

  17. So true.

    Some of the things we are trying at DFID in Kinshasa (with some – though not universal – success):

    1. We have set up an independent experts panel to whom we will give complete access to programmes and open ourselves to challenge and critique. If we get this right, they will be there longer than many of our staff and provide some continuity and institutional memory

    2. Trying to be really honest about learning (and failure) keeping a record of learning that can be passed from staff member to staff member

    3. We would like to encourage arriving staff to spend 3-4 weeks out and about (with no email) learning about the country and programmes. (In the NHS I spend 6 months doing this to learn about the NHS and hospitals are run)

    4. We have run a annual partner survey for 2 consequetive years to try to get partners to tell us what it is like to work with us, encouraging critique and challenge. We have learnt that we often change staff members’ jobs, during postings. I hope we can keep doing these surveys and learning from them.

  18. Great post and the replies all add insights. Staff churn is an issue for donors too so we are developing shared resources to help people manage their transitions well – both leaving and arriving. We do capture lessons but our challenge is to continue improving the sharing and application of our huge fund of knowledge. Writing things down isn’t always a very effective way of sharing! And do we value the wisdom of long serving local staff as much as we should?

  19. The establishment of talent pool may offer solution to the problem. People are recruited in the talent pool at the junior level where education is the only pre requisite and suitable candidates are sought in the university campus. The junior staff is deployed and gains experience in various projects. In case there is no place for deployment, the staff both junior and senior are with talent pool drawing minimal salary.

Leave a comment