About

What is KivaLens?

KivaLens is an alternative way to search for Kiva loans offering a lot of options that Kiva does not. After you've found your loans and put them in your KivaLens basket, you just click a button and the loans will be transferred to Kiva where you'll complete your purchase.

Rebuilt and Rethought Again

KivaLens has been rebuilt. I decided to re-invent it using different tools that weren't available when KivaLens was first written with Silverlight. Please let me know what else you'd like to be able to do. You can also join the KivaLens Lending Team to discuss and get announcements.

Questions/Problems?

All of the data that KivaLens displays is either a direct presentation of or computation from data that is pulled from Kiva.org's Public API with the exception of data from the Atheist Team's research. If you have any questions about why a loan is showing certain data (where you have confirmed that it's the same on Kiva's own site) or if you have questions about what that data means, please check out Kiva's Help Center or contact Kiva Customer Service.

If you find a discrepancy between Kiva's site and what KivaLens displays for loans or partners or if you feel that a computation or filter is incorrect, or you find something unusual about how KivaLens is working then let me know and include as many details about what happened so I can reproduce the problem.

Any questions about the Atheist Team's MFI research should be directed to that team. I only auto-pull, display and filter using the data, I do not produce or verify it.

Features

• Sorting/Filtering by Final Repayment Date vs Repayment Terms While Kiva allows you to sort by repayment terms, there are a few issues with this method from the perspective of the lender. Since different loans have different dates on which they posted to Kiva and the disbursal dates can have wide-ranges as well, an "8 month" repayment term on a pre-disbursed loan that posted almost a month ago could already have repayments underway. While the same repayment terms on a post-disbursed loan that just posted today could end up with a final repayment date months different from the first loan mentioned. Yet Kiva will sort them side-by-side because they have the same "8 month" repayment terms. Since Kiva only allows you to sort by repayment terms but not filter by them, it still shows you all loans. KivaLens lets you remove loans that don't match your selection for final repayment. Also, KivaLens doesn't sort by the term for the borrower, but the final repayment date relative to today, erasing the issues created by pre/post disbursed loans and different post dates. In Kiva's system (and the API, so also KivaLens), all repayment dates are shown at the first of the month, while many/most partners will actually settle the payments on the 17th of the prior month. So, many times you'll actually get your repayments about 2 weeks sooner than what KivaLens displays, assuming the partners settle early (as is common) and the borrower and partner are not delinquent.
• Better Sorts for Repayments! You have the ability to sort by final repayment date (which is better than "repayment terms" sort on Kiva since it's from the lender perspective), but there's a new sort by when the loan is scheduled to repay -- by percentages. Let me explain. Let's say you have 3 loans, all with a final repayment date 3 months away. The new sort will prioritize which pays more sooner. So, let's say loan #1 has a equal amount you get back each month. #2 has a larger amount on the first repayment then by the final it's very little and #3 only has one single repayment at the very end. The new sorts will prioritize by how quickly you get 50%, then 75% then 100% back. So for the loans above, #2 would be the first, then #1, then #3. With pre-disbursed loans that start paying back right away, the first repayment to the lender can actually be a combination of multiple repayments made by the borrower. Using this sort will help find the loans that will allow the quickest turnaround for your money, if that's your goal.
• Any/All/None Filtering What is the difference between "Any" and "All"? "All" is only available for properties of a loan where it can have multiple. So, a loan can only have one country, but it can have multiple tags so there won't be an "All" option for Country since a loan couldn't be in multiple countries and you'd get no results. If you select "All" for tags, then the loan must have  all listed tags. If you select "Any" for Tags, then the loan has to match at least one  of your selection. "None" allows you to exclude results, so if you are opposed to "Liquor Store / Off-License," "Balut-Making" and other controversial activities, you can exclude them.
• Auto-complete drop-downs for Sector, Activity, Country, Region, Social Performance Badges, Themes, Tags, Partners, etc. KivaLens will also show you graphs along the side (when on a large enough device) indicating what options you have for that criteria so you don't have to guess at what will narrow your options for you.
• A Ton of Filters Kiva doesn't have including "Loans at Risk %" and "Currency Loss %" for partners to further reduce your risk. Repayment interval filtering so you can find only loans paying back monthly, granular currency loss coverage options, mentioned ages, specific percentage of genders in a group, $/hour, expiring in days, disbursal days, average loan per capita income, years the partner has been on kiva and how many loans they've posted.
• Bulk Add! This is a favorite feature for previous KivaLens users and what I heard about the most from mega-lenders that they missed after the previous version of KivaLens bit the dust in July of 2015. Add tons of loans that match your criteria to your basket all at once!
• See graphs! View graphs that show the distribution of the current filter. Ex: Select "Retail" as your Sector filter, and by clicking into the Country box, you can see how Retail loans are spread across Countries and what Activities are available by clicking into those drop-downs to make the graphs down the side appear (only on large displays).
• Live Filtering = Speed! Since it downloads all of the loans at the start and then filters them in your browser, you get "as-you-type" filtering. Drag a slider around, pause for a second and watch your results change right then, it doesn't have to talk to the server.
• Filter on the Atheist Team's MFI research with sliders for Secular and Social scoring. When you click on a loan, the Partner tab will show a new section displaying data pulled from that team's spreadsheet. This feature is on by default, simply turn it off in the Options tab.

Reducing Risk

None of this guarantees you anything. Use your own discretion, research, thoughts, team message boards, exchanges with Kiva's Customer Service, etc to help inform your lending habits.

• Institutional Risk: The "Risk Rating" for a partner is based on Kiva's assessment on how likely it is that a partner will fail (institutional default) based on a huge formula as well as some good old-fashioned gut feelings. Example: Some MFIs are running their business from a spreadsheet while others have a comprehensive software package (MIS). There are many things that together give an indication of how sustainable they are over time. The higher the star rating, the less likely it is for the partner to fail. This is not any indication on whether or not the borrower is risky and is no guarantee.
• Currency Exchange Risk: Even if a borrower pays back in full, if they are paying back in a currency other than USD, there is a risk that you can lose some money due to the exchange rate. Using the Currency Exchange Loss % Partner slider, you can use history as an indicator of the future for how much you may lose (but there's no guarantee). Some partners cover more of the currency loss than others. There's also a drop-down for "Currency Loss" that allows you to exclude loans where loss is possible.
• Default Rates: All partners will have defaults, some partners choose to cover the losses of defaulted borrowers in order to pay back Kiva lenders as a part of doing business. Choosing partners with 0% default means you are most likely choosing a partner that is covering losses, which if you have large amounts of money in, may be exactly what you're looking for.
• Portfolio Yield: Reducing organizations down to a number and then making assumptions based off that number can actually unintentionally exclude the very partners you would actually most like if you knew more about them. One of the most significant learnings I had when I worked at Kiva was regarding Portfolio Yield. PY values are based on costs charged to the borrower over the amount of the loan. So, consider this... a partner that specializes in rural loans to agricultural borrowers where the expense to the partner to service the loan (travelling hours to collect weekly/monthly from one or only a few borrowers) is actually quite high and the loan amount is low, it will have a higher PY% relative to a more urban partner that only does high value loans where their cost to service the loan is very low since the borrowers come in to their office. So, don't judge high PY values too harshly or you may just be excluding partners that are servicing the most needy borrowers.
• Profitability: When I initially wrote KivaLens (prior to working at Kiva), I incorrectly assumed that a partner with a negative profitability naturally meant they were at greater risk for institutional collapse and thus cause all outstanding loans to fail. While this is not only a logical conclusion, it is also stated as a risk for negative profitability on Kiva's site. However, this is not always the case. There are a number of partners that receive money from outside sources where this money is not factored in when calculating their profitability. So, just because your borrower is getting a loan from a partner that has a large negative profit, they're not necessarily more likely to collapse; it's not that simple. However, if the partner is not receiving money from outside sources and are operating a loss for extended periods, they may be at higher risk of institutional default. Since there is no way to tell which MFIs are or are not receiving money from outside sources, the "safest" bet is to choose partners with positive profitability. But hopefully this answers any questions where you watch a partner operate at large losses for years but remain afloat.
• Group Loans: Due to the high variability with how group loans are organized by partner, you cannot necessarily make assumptions about group loans versus individual. Some partners group riskier borrowers together so they can prove themselves and graduate to individual loans and the grouping is arbitrary just to make repayment collections simpler to lower the cost of servicing the loan. Some groups are "solidarity" groups where group members have committed to pay for individuals who default. While you can search for "solidarity" in the name field to find some "safer" loans, most solidarity loans do not have that in their name and there is no way to tell from the loan data to help you find more.
• Repeat Borrowers: A borrower coming back for a second or third loan does NOT mean they are more risky. In fact, it usually means the opposite. If a borrower has completed a loan successfully and the partner has chosen to give them another loan, then many times that means they are more likely to repay. Repeat borrowing does not necessarily mean they are in a cycle of debt, but instead that they've shown that their previous loan was successful in improving their business that another (usually larger) loan can help them to continue to grow their business. Use the #RepeatBorrower tag search to help find them. KivaLens user (Thomas) used Kiva's data snapshots and compared repayment rates for ended loans tagged #FirstLoan vs #RepeatBorrower and found loans tagged with #RepeatBorrower had a repayment rate of 99.16% vs loans tagged with #FirstLoan of 98.55% (using data current as of Dec 8, 2015). To get more information and see the code used to calculate, check out this message.
• Diversify! When searching for loans, your criteria may keep bringing up loans from the same partners over and over again. Having too much money with only a handful of partners means that if those partners have institutional default, you could lose all of your outstanding balance. Diversity among countries can also be beneficial because it protects you from your portfolio being as impacted by natural disasters, wars, etc. Use KivaLens' Portfolio Balancing Criteria options (on Portfolio tab) to do this for you. After you set your Lender ID on the Options page, go back to Search and there's a pre-set Saved Search of "Balance Partner Risk" which sets the Portfolio Balancing options to hide loans from any partner where they have > 0% of your active portfolio. If you have a lot of active loans, you may need to increase the percentage to allow for more per partner. Also use the "Limit to top" option to only show 1 loan (or a couple if you're doing large volume) per Partner.
• Something to consider: Ultimately, you shouldn't be considering Kiva as a "bank" (which is why I changed the name from kivabank to kivalens years ago). Over time, even the most cautious lenders are very likely to lose money to default or currency exchange unless you are extremely lucky. You should lend with the expectation of some loss and be happy when everything goes perfectly. You are lending to help people, not for any return on your money (obviously) or even with any assumptions of total repayment. Happy lending!

History

KivaLens was initially created in 2009 (as a Silverlight app) because I wanted to find loans in ways and Kiva did not offer the filter/sorts I wanted. I ended up working at Kiva for a few years and helped incorporate a lot of the search features directly into Kiva's site which also then made their way into the API as well. There are still some great things KivaLens does that Kiva does not do that are handy and fun! Some projects I worked on at Kiva: Super Graphs for lenders and teams, /Live, Zip, Leader boards, Message board search with time-line, the Home page (at the time), and Estimated Repayments are some of the ones I enjoyed the most, many of which were "Innovation Iteration" projects where engineers implement their own ideas.

Kiva Lender Assistant (KLA) Chrome Extension

There's also a Google Chrome browser extension I wrote that inserts repayment graph (sparklines) on the Lend tab along with other repayment info, talks to you about things it notices about the loan, details about lenders and teams you hover over, etc. There are also a number of features that KivaLens has that require KLA to work. Check out screenshots, a detailed description and install Kiva Lender Assistant here or checkout the source code on github (developers wanted!).