setting targets for indicators

6 posts / 0 new
Last post
evamiph's picture
Joined: 05/30/2012 - 5:29pm (US)
setting targets for indicators

Hi everyone

does anyone have any suggestions on how to set targets for indicators? Is there a standard? For example, if a program wanted to reduce maternal morbidity by 30% compared to baseline, how to know that is enough or not...


Eva Sarr

ThomasWinderl's picture
Joined: 04/01/2013 - 7:19am (US)
setting targets for indicators

Hi Eva,
1. for outputs: relatively straight-forward if concrete planning for an intervention has been completed
2. for outcomes:
- more difficult; might require looking at past trends and additional calculations;
- it helps to imagine what you would consider a success of the intervention at the end of the programme; it typically hardly ever a scientific exercise but an educated guess by subject experts familiar with the area and the country;
- in my experience, even more important than setting a realistic yet ambitious target is the PROCESS of discussing and agreeing on targets; it often leads to discussions among stakeholders that reveal a very different understanding and level of expectation what success looks like;
best regards,
Thomas Winderl, Ph.D., MBA
Consultant for Planning, Monitoring and Evaluation

Dean Adam's picture
Joined: 09/26/2013 - 2:38pm (US)
setting targets...

My preference is to avoid setting targets initially until the project team has a reasonable understanding of what level of progress might be achievable or what a change (or not) in an indicator might mean.  I appreciate that stretch goals etc are all meant to be aspirational and motivating, but my experience is that they lead to a lot more work explaining why people and organisations might be doing well even though they aren't meeting targets - specially if you're working for organisations that have a strong accountability lens in how they consider targets and evaluation.
What I often feel is harder is how to address high levels of variation in what the indicators demonstrate between services - if you set a standard improvement rate (i.e. your 30%), going from 50-20% might be easy, going from 30-0% is probably difficult, considering that typically there will always be some level of uncontrollable adverse activity.   So in some ways, you're setting up the services that are already close to the target.  If you set a fixed target i.e. 10%, you may be missing the opportunity to find improvements in the services that are already performing close to the target, but also not appreciating the challenges facing the services further away i.e. are client groups/confounds/is everything else equal (and it seldom is). 
You can get most of the benefits of targets by using some form of comparitive reporting - but its important that this is in a context of service improvement, sharing learnings or opportunities etc rather than who is doing better or worse.  Which might mean that these are 'closed results' and not shared with funders etc.  at this level.  
Saying that, you may have a sense of what is achievable from other studies, or international standards.  
I guess I'm more a fan of indicators being a basis for discussion and don't believe that setting targets etc helps motivate or encourage innovation any more than people already are already inclined to do.  I suspect they result in perverse incentives as often as they encourage on task behavior.  Considering that most community and development literature talks about the importance of collective ownership for positive outcomes, I would suggest that any conversation about setting targets should involve those affected (both in delivery, recipients and funders of services).  
Its a little bit of a tangent, but I read this blog post the other week and think its an interesting challenge to how we socially consider motivation
Its HR focused, but I think it encourages some interesting thinking about why organisations do what they do, or work the way that they do.  
Having said all the above, I'm aware its not a particularly evidence based response and I would be interested in reading anything anyone can provide that agrees or disagrees with my suggestions.
PS, as another article, I find this is an interesting take on why you're measuring and what information you're using - again its a bit of a tangent to your main question but I think it raises important issues relevant for discussing targets.

evamiph's picture
Joined: 05/30/2012 - 5:29pm (US)
Hi Dean and Thomas

Hi Dean and Thomas

both of your responses are insightful; thank you. Dean- would the comparative reporting be quali and/or quanti? In my experience of International development, most of the donors want quantitative targets and indicators because often times the most reliable data that exists is statistics and is more readily available but the idea of complementing it with quali is great. I imagine an appropriate comparison group would need to be selected that is in a different region or division, for example.  

Thomas, I agree re scoping criteria and standards for success. I was caught out once when I mis-judge consensus on output and outcomes between stakeholders. 


Good day



bakhtzada's picture
Joined: 10/07/2013 - 11:48pm (US)
targets for indicators..

i think most briefly that if we convert the qualitative data into quantitative that would be very much better understand and would fulfill the donor requirements well. i think in Eva question, 30% decrease in morality or others could be simplified as an organization want to reduce that target from 2000 to 1000 in a year, so what would be the 30% after three months as indicators target, if 337 morality rate have decreased that would be better option, no standards would be there........i think your experience would provide you better guess....than others one's.....
Bakht zada

aliyuaminuahmed's picture
Joined: 12/23/2013 - 9:17am (US)
Setting Targets

Hi Eva,

Setting targets will depend naturally on the resources you have, the duration of your project and complexity of the intervention. Since you already have a baseline then you need to observe the trend and then try to consider the available resources and desired level of change. Here is a simple trick if you have a 5 year project then you split your targets "normally" that is if you plan to achieve 90% then in Year 1 your target will be 10%, Year 2 = 20%, Year 3 = 30, Year 4= 20% and final year = 10% reason is the the initial year you are setting up and starting your project (minimal staff and infrastructure) , by second year you must have finished hiring but is building capacity of stakeholders probably, in third year you would have all the available resources partners etc... definitely at that time you should be operating optimally. The fourth year you might begin to loose staff or probably higher demand from your sponsors and other programmatic issues by the 5th year you will be winding down so less staff are available etc...

However in targeting consider institutional capacity, environmental and political concerns in addition to the available resources. A three year previous patterns may be enough for you to take an annual average then add a bit on value through informed guess (projection) there you have an annual target. 

Be careful to set only one target per indicator and not to set too modest targets or too ambitious targets. "Targets should be like mini-skirts ; short enough to be interesting yet long enough to cover the subject"

MMR is a high level indicator which means that several projects contribute to achieving it. Unless it is a National Program it will be difficult for your organization to make such big difference. Also we need to consider if the MMR has stabilized over time say if for example in the past three NDHS the results were 95/100,000 or near similar results so its quite difficult to reduce it by 30% in a short time due to historical trends which may be attributable to the entire health system e.g. medical personnel, infrastructure, health seeking behaviors, the "three delays", commodities, etc... so its likely that one program cannot  address all these.

Add new comment