Monthly Archives: November 2013

Do For-Profit Schools Give Low-Income Children A Real Choice?

At the opening of our school in Embu in May 2012

At the opening of our school in Embu in May 2012

Before reading this, please take a minute to donate to the Ravindra Ramrattan Memorial Fund.  Ravi was killed in the terrorist attack at Westgate Mall in Nairobi on September 21st.  You can read about his story on this blog.  

Bridge International Academies, a chain of low-cost private primary schools based in Nairobi, Kenya, was the subject of a recent episode of All Things Considered on NPR.  I would encourage you to listen to it in full, as the cofounder and my old boss, Shannon May, does a great job of explaining the philosophy of Bridge. The radio segment and accompanying blog post discuss the Bridge model, and highlight a few common criticisms.  Here is one:

“If somebody suggested that kind of an educational model, in this country they would be laughed out of the educational community,” says Ed Gragert, the U.S. director of the Global Campaign for Education, which advocates for increased access to education in the developing world.

“That’s not how kids learn best,” he says. “Kids learn by interacting with each other. It seems like we are going back for the sake of somebody making a profit to where a robot could teach that class.”

I worked as a business analyst with Bridge for a year, where I was responsible for all things data, including longitudinal student performance testing, marketing analytics, and a whole lot of one-off projects.  When I left the company in May 2012, I wrote a five-part post discussing my thoughts on Bridge and its model. Given that I wrote the post 18 months ago, some of the processes, data, and technologies I describe may be a bit dated.  For example, Bridge hadn’t yet rolled out e-readers and was just starting to utilize the smartphone application.

I am reproducing the unedited post here in full.  At the end, I’ve added an updated conclusion that gives my thoughts on the criticisms raised in the NPR segment, and my thoughts on Bridge and the state of education today.


301812_10100100143613204_2077715292_n

I. The model

After six months learning about agriculture in West Africa and working on a project whose objective was to improve the private sector, I decided to return to the private sector, since the public sector was not very good at making it any better.  I had interviewed with the Acumen Fund for its global fellowship in Nairobi six months prior.  I knew a few folks through my Kiva connections, and began networking for jobs there.  I cold-emailed a few, hit up friends for introductions, and stumbled into an informational interview with Jay, the founder of Bridge International Academies, through a former Kiva Fellow in Benin.  When I told him I was trying to move to Kenya four months later, he said “let’s meet when you get here.”

So, when I got there – actually, 10 hours after I got there – I met with him to discuss the prospect of me working with the company.  The company, as a background, is a chain of low-cost private primary schools serving the slums and low-income communities in Kenya.  When I met him in January, they had just opened their 20th school.  When I met him again in May, they were at 25 schools.  When I left two weeks ago, Bridge had 75 schools throughout Kenya, and is planning on opening another 200 by the end of 2013.

The model is as simple as it is elegant.  Bridge is creating a “school in a box” – a highly standardized, systematized, and replicable model for an individual school, where everything, from the curriculum to the training to the school operations, is designed for scale.  The process begins with market due diligence, where a team of research associates interview 40 households in each community where we are considering opening a school.  We survey the parents to determine whether there is a market for one of our schools.  For the last four months, I worked with the research team, developing an algorithm to predict the size and profitability of each new school, allowing us to determine how much we need to pay for land, how many classrooms we need to build, etc.   I redesigned our research report and, using our enrollment figures from this year, created a rubric based on population density, market size, cost-competitiveness, and the number of competing schools, which gives us a fairly accurate projection of how a school in that community will do.

Once we have approved a community, we scout for suitable plots and negotiate for the land.  Our construction team builds the school, and our training department ensures that teachers are trained and ready to teach by the time our school opens.  The curriculum team – which consists of 40 Kenyan and American educators – script every minute of every lesson, from math to English to science to Kiswahili, which is then delivered to the students by the teachers.  The incentive structure for school managers is based on the number of students they attract to their school, while the teachers are given bonuses based on performance.  All problems at the school – from teacher complaints to requests for water or desks – are routed through an in-house call center, which also makes outgoing calls to schools and teachers to ensure that the ship is running smoothly.  Lastly, the IT department has designed a billing system that allows parents to pay with M-PESA, and a school-management Android smart-phone application that automates much of the payment and performance monitoring at the school level.  All in all, it is a remarkable model.

156194_10100100143663104_740939801_n

II. Why is Bridge successful?

I think that one of the reasons that Bridge has been so successful at innovating has been its willingness to bring in a multidisciplinary team to run the show.  People like me, who have no background in education, but a good deal of experience in other areas, bring fresh ideas to an industry that, apart from certain ed-tech companies and charter schools like KIPP, is not known for innovation.  Our head of operations was the former director of business development for Dominoes Pizza in Asia, responsible for introducing the retail chain to a completely new market.  The founders include one successful tech entrepreneur, a division lead at IDEO, the design firm, and an anthropology PhD.  It doesn’t get much more interdisciplinary than that.

Another reason I think it has seen success where others like it have seen failure is that it refuses to accept excuses for poor performance on the part of its employees and vendors, and sees itself first and foremost as service provider that puts its customer first.  As a generality, business in Africa moves more slowly and expectations are sometimes different.  The business culture – with many notable exceptions – is such that the customer is never at the top of the priority list.  The other day, a coworker told me that a vendor to whom we paid a lot of money to perform a service which he performed poorly – was upset that the management was too busy to see him.  “This is an outrage – I am a client,” he said.  The fact that we were, in fact, the client and he the service provider undoubtedly never crossed his mind.

Bridge demands quality from it suppliers, timeliness from its vendors, and results from its employees.  This mentality ensures that the best people for the job are in place, and they can perform their jobs efficiently.  In the NGO world (or at least the ones I have seen), this approach to doing business is hardly, if ever, the norm.   If a deadline is missed, people say “what can we do – it’s Africa time.” Or an NGO holds a conference that only attracts participants for the per diems and lunch provided.  Accountability is not part of the lexicon.  At Bridge, on the other hand, if a vendor screws up the logo on the 1,000 shirts it ordered, it refuses to pay until the mistake is corrected.  It’s only business.

That is because Bridge, above all else, is a business.  It happens to be building low-cost primary schools in slums, but it is first and foremost a profit-oriented enterprise.  If Bridge is going to reach 1,000 schools and one million children in countries around the world, it has to be laser-focused on the bottom-line to succeed.  Drawing from a talented pool of private sector veterans and a founder who started and sold a company in his twenties, Bridge understands this well.

This kind of truly business-minded approach to development is rare, even within the relatively new and trendy industry calling itself “social enterprise.” While many social enterprises – companies that try to turn a profit while doing good – struggle to balance the demands of a double bottom-line, Bridge has create a model where the profit motive is inseparable from the social mission – one cannot exist without the other.  If Bridge students perform poorly on the KCPE exam – the test culminating primary education in Kenya – parents will pull their kids from Bridge schools en masse.  On the other hand, if they perform better than students at other schools, Bridge schools will double in size in a day.

That is because poor parents, just like middle- and upper-income parents, are discerning consumers when it comes to education.  They look for quality and, more importantly, value for the little money they have.  This is generally true for most products and services, but particularly so for education, which parents see as a means of getting out of the slums.  For Bridge to grow, it must be educating its students better than the alternatives in the community – either government schools or other non-formal schools.  In this case, it means ensuring that we outperform other schools – both government and other non-formal schools.

562998_10100100143847734_1423808045_n

III. The academic foundation of the concept

The Bridge model is a fundamentally libertarian idea.  It is premised on the belief that school choice is a good thing.  Many organizations, including many establishment development titans, believe that education should be a public good, provided free by the government.  This may be true in theory, but, like most development theories, it is rarely true in practice.

For example, Kenya already technically has a free primary education system, where all students, regardless of socioeconomic status, are guaranteed the right to an education at no cost.  Yet, in the slums, where the houses are built illegally, few government schools exist to serve the communities.  And even in areas where public schools do exist, additional fees payable to the head teacher and others mean that parents are paying almost as much for school as they do at Bridge.  Outside of Nairobi, many of the government schools are as close to free as a school can get in Kenya, but they are often overcrowded, far away, and staffed by complacent teachers who are either overburdened with too many students or complacent when it comes to teaching.

Much of the free primary education system in Kenya was subsidized by foreign donors.  But these donors eventually decided to scale back funding after massive corruption scandals were exposed.  In the main Kenyan daily newspaper, the Daily Nation, every day exposes a new corruption scandal.  Needless to say, the state of the government education system is underwhelming at best.

Because of the failures of the government system, thousands of non-formal schools have sprung up throughout the slums to serve these communities.  Education entrepreneurs, churches, NGOs, and other groups build and operate schools to fill the void left by poor state-run education.  While the Global Campaign for Education and others would like to believe that these schools do not exist, they are everywhere.

When people speak and write about Bridge, they credit the company with a radical new approach to education, offering private education as a means of providing quality education.  But Bridge is hardly innovative in this respect.  Non-formal schools like Bridge have existed for decades.  It was not until the early 2000’s that a British academic named James Tooley began seriously researching education in the slums of India, Nigeria, Ghana, Kenya, and other countries, and finding bustling schools with hard-working teachers.  He published a book called The Beautiful Tree and several articles for the Cato Institute detailing his findings, which were influential in seeding the idea for Bridge.

Rather, the real innovation Bridge brings to this sector is its relentless pursuit of efficiency gains and systematization of the day-to-day running of a school.  Technology as a means of creating scalable payment and performance monitoring systems, a scripted curriculum written by subject-matter experts, modular school construction using low-cost materials – these are all key innovations that have the potential to revolutionize this sector.  But the concept of a low-cost private primary school for the poor is nothing new.

380295_10100100146347724_1437534620_n

IV. The criticisms of Bridge

The first and most obvious criticism of the Bridge model of education is that a scripted curriculum creates a non-dynamic learning environment for children.  The western model of education is premised on the idea that critical thinking is essential to success.  The very idea of a liberal arts education is a distinctly Western concept.   So, naturally, when people hear that our teachers are high school graduates who are taught to teach by reading to children from a script, they automatically assume that the quality of the education is poor.

It is true that the standard of education at a Bridge school is going to be far below that of more expensive schools.  But when compared with the alternatives – which include government schools staffed by often unmotivated teachers and other non-formal schools offering little in-house teacher training – Bridge offers an education that is subject to rigorous testing and review.  For example, our curriculum is written by a team of Kenyan education professionals and Teach for America alumni, many of whom have Masters Degrees in education.  It has a video team that films lessons to be reviewed by the curriculum writers.  They look for level of engagement among the students and adjust the approach to maximize comprehension and retention.  Student exams are digitized and reviewed both to identify weaknesses in the curriculum, but also review teacher performance.  Lastly, the school managers audit the teachers on a regular basis to ensuring that they are performing adequately.

Lastly, and most importantly, Bridge undertakes a rigorous longitudinal testing study of 5,000 students every six months to monitor improvements in reading and math.  Using a test developed by the Research Triangle Institute and USAID called the Early Grade Reading Assessment and Early Grade Math Assessment (EGRA / EGMA), Bridge compares the performance of 3,000 of its own students with that of 1,000 students at government and other non-formal schools.  The results, which are shown on the website, show strong performance gains in basic reading skills, compared with its peers and less strong, but still measurable, gains in math.  I know this because I was responsible for leading this student testing and performing the analysis.  Using this data we can then tailor our curriculum to address our problem areas and improve the curriculum.

This level of analytical rigor is simply not possible at other non-formal schools.  Why?  Because Bridge is able to leverage economies of scale.  It can invest huge amounts of resources into improving its model because it knows that all changes can easily be rolled out across every single school in a day.  When I first met Jay in January 2011, Bridge had just broken 10 schools and opened its first school outside Nairobi.  By the time I left, the company had 73 schools across Kenya.

This level of growth means two things.  First, since the unit economics are such that each individual school is profitable at a relatively small size, more schools mean additional revenue that can be poured back into the company.  And second, any major policy changes can be backed with incredibly rich data sets.  As the company’s business analyst, I was working with datasets with sample sizes in the tens of thousands.  For someone trying to use data to better understand how our parents think, pay, and act, and understand what makes a good school, I was in heaven.

524471_10100100144406614_1233840425_n

V. How Bridge uses data to perfect the model

In the last post, I talked about how Bridge is able to leverage its economies of scale to both utilize huge amounts of data to make decisions and, once those decisions are made, they can be rolled out en masse.  I will give a few concrete examples of how this works in practice.

Last September, we wanted to see whether offering a free month of school and having a grand opening ceremony with a bouncy castle would boost enrollment.  So we did what most respectable startups exploring a new product or market would do: we tested it.  Of the nine schools we opened last September, four had a grand opening ceremony (GOC) and first month free (FMF), two had only FMF, one had only GOC, and two had neither.  When I looked at the numbers, the results were amazing.  Not only was initial enrollment nearly three times what we had experienced in the past, but the conversion rate – the most important factor in measuring the efficacy of a marketing promotion in retail – was 85%.  This is practically unheard of in retail.  In other words, 85% of people tried the product and decided to buy it.  When was the last time you started paying after the free trial expired?

When I shared the results with the management team, the action was relatively decisive.  With the 30 January-2012 schools scheduled to be opened in only eight weeks, they changed everything.  Effective as soon as possible, every new school would have a grand opening ceremony and every new student would be given a free month of school.  And, to make it fair, all 60 schools would have a GOC in January and every student would receive January free.   One by one, the managers detailed what needed to be done and set to work.

The IT team began making changes to the billing system and the smartphone application; the training team began prepping the training facilitators to communicate the new policy, and the operations team went out to each school to explain the changes directly.  Marketing began contacting companies that rent bouncy castles and negotiating prices, while government relations reached out to the elders in the community and invited them to attend as “Friends of the Academy.” Within 24 hours of my sharing the analysis, the company began preparing for a monumental change in the way things were done.  In January 2011, our largest school opened with 200 students.  In January 2012, the biggest had more than 700.

For me, the policy change had even greater implications.  Since each cohort of schools opened with different policies, regulations, and circumstances, it was difficult to isolate determinants of performance without introducing incredible amounts of bias.  But now, every school had a grand opening ceremony and January became a free month for every single student.  Therefore, the maximum attendance in January effectively equalized every school and made them as close to comparable as they would ever get.  Now, all of a sudden, we were able to actually measure how factors like population density, school location, cost competitiveness, income levels, urban/rural, and relative importance of education in the community influence school size and profitability.

Example of a regression model built to predict enrollment.

Example of a regression model built to predict enrollment.

From our market research, we had hundreds of consistent variables about each community.   So I built a massive Excel model and ran some basic correlation analyses and scatter plots to identify the most important factors in determining where to open a school.  Based on the analysis, I created an algorithm to actually project the size of the school after one year that was accurate within a range of 100 students at 80% of schools.  We automated the report creation and incorporated a profitability model into each one, which would dictate land price and school size.  And, just like everything else at Bridge, once we had it right, the new report became part of the Bridge model, and is there to stay until the data proves it wrong.

579097_10100100147879654_1774975841_n

VI: Updated conclusion from November 2013

I feel largely the same way about Bridge today as I did back in 2012 when I left to return to the U.S.  I still believe it is a revolutionary idea that has the potential to re-define how companies serve the poorest segments of the population around the world.  It’s laser-focus on scalability and and its systems-orientation have made it a model for other burgeoning social enterprises, and not just in education. A standardized approach designed for replication has applications in healthcare (see Penda Health), water and sanitation (see Sanergy), agriculture (see One Acre Fund), and other sectors.

Bridge has strong partners in its investors and has some very talented individuals leading different segments of the company.  If it is successful, it has the potential to prove out the social enterprise model as a viable approach to economic development.  In doing so, it would open up funding for similar companies around the world.

And perhaps most importantly – at least to me – it has the potential to prove that the conventional wisdom can and should be challenged.  When James Tooley and others first began advocating for private schools in the slums and talking about charging poor families for an education, he was lambasted by the education establishment for trying to charge for what they considered to be a public good.  But over time, people came to realize that theory does not always jive with the realities of the situation on the ground.

In the specific case of Ed Gragert, the man criticizing the model in the NPR piece, he should go to Mathare, Kawangware, or any of the other slums where Bridge schools are located and visit the schools in the area. Before passing judgment, he should look at the alternatives and judge for himself whether Bridge offers a decent education relative to other schools. He should think about the availability of talented teachers in the communities where Bridge builds schools, and look at the incredibly rigorous screening process they use to identify the best candidates in the community.  Because then, he would realize that what people would say about a school like Bridge in the U.S. means next to nothing.  If there is one thing I learned during my years working in international development, it is that context is key.

In reality, Bridge – like many of the innovations in Africa – is far ahead of the curve of the U.S. education system.  Contrary to Mr. Gragert’s assumption that a person would be laughed out of the room for proposing a Bridge-like model, the arc of education is bending closer to the Bridge model than he would like.  MOOCs, Khan Academy, “flip teaching” and the entire blended learning model are an attempt to leverage economies of scale and domain expertise to improve the quality of education in the U.S.  Because Bridge is deeply constrained by a lack of resources, it is required to innovate in ways American educators do not.

For example, the scripted curriculum allows Bridge to thrive in the absence of a robust higher education infrastructure in low-income areas of Kenya. If they can leverage the expertise of skilled educators, Bridge teachers can focus on controlling the classroom and ensuring the students are at least reasonably engaged.  Without electricity or Internet, Bridge schools cannot simply project Khan Academy lessons.  So, in reality, the e-readers are an ingenious way of delivering world-class lessons without the resources of an American classroom.  These innovations may seem trivial, but, in reality, are quite ahead of their time.

So, a year and a half later, I feel the same way about Bridge that I did when I finished working there back in 2012.  I wish them all the best in their expansion to Nigeria, and can’t wait to see what they do in the future.


Develop Economies’ Music Recommendation