System review

You need to review your existing MIS – assess the current intake processes, information sources, technology, and reporting systems – and interview key staff from each department about the types of information they need. A ‘process mapping tool’ will help you assess your existing system. It not only reviews the actual technology platform but also the process for collecting and reporting the data, including who collects and uses the information. (See the resources section at the end of this guide for information on process mapping tools.)

There are nine questions you need to consider in order to make a thorough assessment of your information needs. Answering these questions before you choose tools or indicators, or make changes to the MIS, will save you both time and money.

Nine essential questions on your information system

1. What information is needed and by whom?

2. Who will use the information and how?

3. Who will the MFI collect data from?

4. How often will data be collected and analysed?

5. Who will collect the data and how?

6. Where/when should data be collected?

7. Who will process the data and how?

8. What data analysis will be done and who will do it?

9. How will the information be communicated and reported?

Question 1: What information is needed and by whom?

Since microfinance is all about serving clients, most of your information needs will relate to understanding your clients’ needs and capacities. For many MFIs, the mission implies targeting a specific type of client, and achieving certain outcomes for their lives or businesses. Therefore, SPM requires that information on client needs be analysed and understood from this mission perspective. MFIs with a mission to reduce poverty, for example, will need information on how to help clients move out of poverty, whereas an MFI with a mission to increase financial access for women micro-entrepreneurs will need information on the barriers that these women face. Regardless of their mission, all MFIs will want to know the following:

do staff members apply a consistent definition of the target group and related sub-groups?
is the current marketing and product delivery strategy reaching the target clients?
are our products helping clients to improve their lives, and if so, how?
how could the product and delivery mechanism be altered to provide the maximum benefit for the minimum cost?

It is also likely that you will be serving several different client segments, whether by age, gender, business type or geographic location. It may be expedient to treat all clients the same in terms of the information system; however, segmentation is essential so that you can understand the needs of different client groups, and therefore effectively achieve your social goals.

Question 2: Who will use the information and how?

Stakeholders at all levels of the organisation play an important role in organisational performance management, and they will need different information at different times. The IT Department, working closely with the SPM champion and Executive Director, should think about the specific needs of various stakeholders and invite each department to identify their SPM information and reporting needs.

The SPM champion and the Executive Director should act as go-betweens, ensuring that the IT department understands and buys in to the organisation’s social goals, and that other departments make reasonable information and reporting requests. It is crucial that the SPM team and IT department understand the purpose of the information need so that collection and reporting can be streamlined.

Broadly, you can use performance information to:

track performance against targets - as part of a performance management system or just for annual reporting
signal early warnings - identifying and responding to problems at an early stage
segment the portfolio - breaking down performance to examine differences by client group, loan officers or regions. Such segmentation improves understanding of different markets and allows for targeted responses, for example, it can help you understand different client exit rates
help clients reflect on changes in their lives over time
analyse clients’ use of services: are the services appropriate to clients’ needs? By monitoring client use and response to services, you can improve the quality of the services provided.

Performance information is useful to all stakeholders, but they might not need the same level of detail. For example:

clients could track their individual progress
field officers could assess whether they are reaching their own outreach goals
branch managers could look at branch outreach and exit rates to estimate whether they are on track for branch-wide bonuses
the Board and executive management might want to see detailed information at both the branch and organisational level.

Question 3: Who will the MFI collect data from?

Clients Many MFIs collect client feedback, but few consistently use it to make decisions and effect positive changes. You need to be strategic in determining what information and feedback you need to collect from clients.

Think about whether you need information from all clients, or if a sample will be sufficient. The box below describes situations where it is best to use sampling or census data collection techniques, when the daily monitoring system cannot provide enough information. When using samples, think about whether there are particular groups of clients you want to use and consider gathering information from non-clients as a comparison group.

Census or sample for data collection? It depends…
A representative sampling of clients may be most appropriate when:
an issue arises from the ongoing monitoring system
the budget does not allow for a complex study
the issue needs immediate attention.
Typical methods would include:
focus groups
client interviews
client satisfaction surveys.
A census, collecting information from most, if not all, clients, is often used when the information is readily available (from client files and loan application forms, for example).

Staff can also provide valuable insights into emerging issues, concerns and opportunities, especially loan officers and other field staff who regularly deal with clients. Field staff can help you fully understand and interpret client feedback. Listening to field staff has other benefits too, as they are likely to feel more involved and fulfilled, and may stay in their post longer.

Management learns to listen to staff – the hard way…

An MFI in West Africa noticed that its clients were not graduating to the next loan size. This was because of the increased risk they would have to assume for themselves as well as their other group members, who were co-guarantors. After conducting research, the MFI found that many female clients were not able to accumulate assets, particularly market stalls, for their businesses, which would allow them to qualify for standard individual loans requiring guarantees.

Loan officers and all department heads were involved in designing an individual loan product so that clients could purchase fixed assets instead. However, management did not heed staff warnings that the clients would not accept a guarantor requirement for this larger loan product. Needless to say, the initial results were disappointing, and despite significant marketing efforts, few clients were interested. Eventually, management dropped the guarantor requirement and product sales improved.

Question 4: How often will data be collected and analysed?

This depends on the kind of analysis you undertake, and how often your stakeholders want to use or report on the data. Practical considerations, such as existing operational systems, may also affect this decision. Remember: how you use information, and how often, will drive how often you need to collect it. For baseline information, consider whether you want to collect it daily as clients enter, or on a sample basis (either annually or more frequently). For follow-up information, think about how often you need it: annually? After every loan cycle? Every five years? Be careful not to collect information more often than it is needed.

Tip: Look at how financial information is collected as a guide to how to collect social performance information

Question 5: Who will collect the data and how?

Your SPM team (or the department in charge of SPM) could undertake periodic or annual surveys. Or you could outsource to a local research firm. Day-to-day data collection is likely to be carried out by branch staff, with the data then fed systematically into the MIS. But you will need to consider the effect of having loan officers collect non-loan related data.

Having a separate person collect the social performance information could help avoid conflicts of interest or bias, by allowing clients to answer frankly without fear of negatively impacting credit decisions. But it might also be that clients are less likely to talk openly to a stranger. If you do choose this method, be aware that it tends to cost more and involves some logistical planning.

Some other issues you need to address when considering who should collect the data:

Is it a priority that field staff be involved and learn from the process?
How much time and disruption would it take to complete the task?
Do staff members have the ability to collect the data correctly? Are they motivated to do so? (Will they be penalised for shifting away from their normal duties?)
Is there a local research organisation or university that can commit to regular, inexpensive data collection?

Why should staff be trained on data collection?

One of the most critical steps in adapting your MIS for social performance involves training staff. If you don’t provide good training and follow-up, staff might interpret the same data collection questions differently (such as for client income and expense data), which would limit its validity. Training should not only enable staff to properly collect and analyse the data, but can be used as an opportunity to build staff commitment and buy-in, leading to improved consistency and quality of data collection.

You can collect client information as part of the loan application process, as part of other existing forms, or with a new form. There are pros and cons of every method (see Table 6.1). But remember that you should also collect information from other groups, such as competitors and non-clients.

Table 6.1: Different methods of collecting data from clients: pros and cons
Method	Pros	Cons	Cost*
Routine / Regular	Comprehensive and covers all time periods	Data needs to be entered into database to be useful	$$
Loan applications	If trained and supervised properly, loan officers can collect important baseline/market profile data at little cost additional cost to MFI. This builds on existing processes and is easy to conduct quality control.	Adds to time taken to complete loan application and may have data quality issues without proper field staff training.	$
Client visits	Visits by staff, such as internal auditors, can collect information on staff performance and satisfaction with the MFI’s services that can be used to improve policies, products and procedures.	Clients can feel uncomfortable giving critical, yet constructive, feedback to MFI staff.	$$
Suggestion boxes	Creates a format for clients to share new and innovative ideas, as well as how to improve in ways that can save money and increase outreach for the MFI.	A lot of information goes unused, as the open-ended format can be hard to make comparisons, and it is difficult to know when a recurring complaint is a big issue. Many clients may not be able to read; or putting a suggestion in the box involves visiting the branch, which may be far away.	$
Exit interviews	Good way to understand reasons for exit, and serve as a last chance to retain client.	Difficult to get disgruntled client to take time to give feedback.	$$
Informal feedback	Through field staff interactions, and discussions during monthly/quarterly staff meetings.	Need to structure and report information systematically and loan officers may simply not have the time.	$
Occasional	Can be adjusted for new and specific inquiries	Can be expensive and time-consuming	$$$
Surveys	Can be used to monitor client satisfaction as well as to conduct market research for new potential markets and products	Results depend heavily on whether the right questions were asked, in the right way, to a representative sample.	$$$
Focus groups	Helps to better understand survey responses, capture concepts in clients’ words/perspectives.	Requires clear guidelines, good coordination and strong facilitator and rapporteur to yield usable information. Can be expensive if try to have a representative sample. The tool overall is qualitative in nature, so the results cannot be validated in a statistically significant way. Infrequent application means not useful for day-to-day management decisions.	$$$
Impact studies	Good for identifying broad range of impacts on client’s family.	Can be expensive if done in statistically significant manner (with a non-client comparison group). Important to account for drop-outs. Prone to criticism due to difficulty with attribution.	$$$$

*The more $, the more expensive in terms of direct costs and staff time.

If you provide microenterprise loans, it’s essential to collect information at the client’s business or home.

Question 6: Where/when should data be collected?

The three most common places to collect information are:

at the client’s home
at centre meetings
at the client’s business premises.

Visiting the client’s home or business is imperative as answers can be visually confirmed, which is important for protecting data quality. But such visits can be costly and time-consuming, which may limit the sample size.

There are certain times when it is important to collect data:

during the loan application process, loan disbursement or loan repayment
after a client has exited the programme
during loan supervision visits
during group meetings
during special group meetings or through visits (e.g. focus groups).

The most common and probably most efficient way is to collect data during normal operations (such as the loan application process). But it’s important to bear in mind that the timing of data collection can influence the client’s answer. For instance, during the loan application process, the client may be short of cash and may give a different answer in order to get a larger loan.

Fonkoze (Haiti) has employed social impact monitors located in branches. Their role is to analyse the poverty status of incoming clients and track changes over time, and to research client needs and experiences (focusing on client satisfaction and retention). Based on their findings, they recommend ways for Fonkoze to improve existing services and introduce new products. The social impact monitors use a number of tools, including:

an Evaluation card, which has an embedded Progress out of Poverty Index (PPI) set of questions. They also use their own tailored questions, allowing them to place clients in the right programme (based on their poverty level), and monitor their progress over time
regular focus group discussions, which explore issues relating to the design and delivery of products and services
exit interviews, which find out why individuals leave and what changes might have prevented this.

Question 7: Who will process the data and how?

Whoever has the responsibility for data collection will usually also process it. How to process the data will largely depend on the resources you have available. Ideally, field staff will have electronically entered the data, in real time, while talking to clients. However, in reality, very few MFIs have this capacity; often, they use a two-step process, whereby field or administrative staff write clients’ responses on the paper form, and then enter the information from the form into the database. The data can be entered into an integrated database (preferred method) or a separate database set up for the sole purpose of social monitoring. (Alternatively, the data may be kept in manual form.)

Three tips to consider:

make sure that procedures to ensure checks and balances are in place so that responsibilities and lines of supervision are clear
manual systems are simple and less costly, but are time-consuming. They may be most appropriate for an annual survey.
some MIS systems only process and analyse group data versus individual client data, causing some group lenders to begin tracking all individuals within the groups.

Question 8: What data analysis will be done and who will do it?

When deciding how to analyse social performance data, it will be helpful to consider how your financial performance is monitored and analysed. Data on financial performance is used by staff at all levels of the organisation, and staff are generally able to analyse information relevant to their job. Rather than centralising information and communicating results to field staff and branch managers, they are able to see information themselves and carry out their own analysis.

Your MFI needs to monitor data and trends, use trigger points to flag up concerns, and conduct follow-up investigation. This can mean using more qualitative techniques, such as staff and client interviews, and focus groups, but it doesn’t have to be a complex process.

Unitus helps MFI partners get to the root of the problem

Unitus advised its MFI partners to construct ‘use cases’ – best estimates of causes, and how they may affect certain observed results, much like a process mapping exercise. Unitus recommends that the MFI starts with: ‘What do we see that is troubling or at least worth investigating further?’ (e.g. the MFI finds out that new clients are above the poverty line, contrary to the desired result.) Using the best means available, the MFI then traces back to a probable underlying cause, often one aspect of its operations, such as incentives that encourage outreach to better-off clients. If an operational procedure is linked to a gap between actual results and desired results, the MFI can then realign the activity to its mission and strategy (e.g. ‘Our mission is to provide the poor with financial services, therefore we will redesign the incentive system to reward loan officers for their success in reaching out to poorer clients.’)

Some types of analysis can be conducted by both field staff and management, such as:

tracking performance against targets
identification of patterns and trends
understanding of issues related to clients
simple portfolio segmentation (identifying clients with big, small, or no changes)

Different levels of staff manage performance in different ways. Make sure they have the data they need.

However, senior management (or external specialists) should take responsibility for more detailed analysis, such as:

analysing the percentage of clients experiencing a positive or negative change
time series analysis which tracks client progress over significant periods of time
complex portfolio segmentation looking at change related to certain client variables, or identifying characteristics of clients with greatest or least change or exit.

These types of analysis may be needed to correlate certain results. Find out if local consulting organisations or your regional microfinance network can provide help in this area.

Remember, you need to implement a formal process for analysing data (including deadlines and specific responsibilities for staff) and a detailed policy on how the organisation will use the data and related recommendations.

Question 9: How will the information be communicated and reported?

When staff understand that information is used to improve performance, and see the practical results of data collection, the quality of the data they collect will improve. You can facilitate this by designing effective mechanisms to report social performance information. Start by thinking about how you communicate and report on financial information, and look for ways to build on this. Make sure you create feedback loops for conveying field-level information between management, staff and clients. Good reporting formats should facilitate analysis and understanding by those who use the information and make decisions based upon it.

For example, Table 6.2 presents a sample report from NWTF in the Philippines. The table presents poverty status (collected through the PPI) by loan cycle. They are also able to disaggregate poverty status by loan size, type and age of business, number of businesses, poverty level on entry, repayment rates and movements out of poverty. By having timely and clear information on which types of clients are making progress and when, management can make more effective decisions about how to adapt their services appropriately.

Table 6.2: Sample PPI report, NWTF
Loan Cycle	No. of Clients	Bottom 50% below National poverty line	Top 50% below national poverty line	Between national poverty line and $2/Day	Above $2/Day
1	13,730	35.90%	23.34%	8.26%	32.48%
2	5,625	36.35%	23.59%	8.35%	31.70%
3	3,306	35.85%	23.76%	8.36%	32.02%
4	3,002	34.68%	23.08%	8.42%	33.81%
5	2,477	31.40%	22.96%	8.75%	36.88%
6	2,133	31.27%	22.57%	8.68%	37.47%
7	1,816	30.00%	22.28%	8.72%	38.99%
8	1,883	29.10%	22.42%	8.84%	39.63%
9	1,735	27.98%	21.68%	8.94%	41.40%
10	5,481	24.52%	20.91%	9.18%	45.38%
	41,188

Social Performance Management Resource Centre