FACULTY OF ARTS, COMPUTING, ENGINEERING AND SCIENCES

 

BROADBAND INTERNET ACCESS IN DEVELOPING WORLD ECONOMIES:

AN INVESTIGATION OF THE FACTORS AFFECTING VIABILITY

     
 Chapter 7: Discussion and conclusions (Part 2)    
     
Home | Abstract | Table of contents Back
DRASTIC Website

The Bigger Picture –transferability of conclusions

 

Conclusions have been presented, especially in chapter 6, concerning the market, the user demographics, business scale and utilization levels necessary for a viable Internet business in Aceh. The discussion must now turn to the wider applicability of these conclusions.

 

The author proposes three steps for examining the potential viability of similar businesses in other locations:

 

1. Determination of the general conditions

The climate for businesses in general should be reviewed for the country under consideration.  These can be researched using websites such as Doing Business (World Bank, 2006) and Corruption Perceptions Index (Transparency International, 2005).  These will highlight any especially risky or problematic issues.

 

The maturity of the economy should also be reviewed, in terms of its readiness to embrace and benefit from improved Internet access.  The framework proposed by the Center for International Development at Harvard University (2000) could be adapted to evaluate the "Network Readiness" of the community that is being studied.  This framework describes a number of areas in which an economy should work to benefit from the Internet economy.  For each area, criteria are given, allowing one to determine how mature a particular economy is. This exercise was applied to Aceh, and the results are shown in Appendix E.  Assigning the —stage? of development for these criteria was difficult.  There would probably be some benefit in updating the criteria, and maybe having more than four stages for each.  However, the exercise is useful in highlighting areas that might hold back a potential business.

 

2. Reviewing a full range of viability factors 

 

Returning to the work of Alan Thompson (2005) on the Dimensions of Viability, a more detailed analysis of the factors affecting viability was prepared. This involves determining how each dimension of viability will be measured and how each will be weighted. (See Appendix F).  This framework was found to be helpful because it encourages a holistic, “Systems Thinking” approach. Some of the answers obtained may be subjective, but nevertheless are likely to illustrate areas of business planning where extra effort will be needed.

 

3. Business model development

 

The author developed a spreadsheet model that allowed different potential solutions to be modeled in terms of capital requirement, Profit and Loss, and cash flow.  Revenue projections were based on competitor pricing and assumptions about realizable utilization based on previous experience.  Costs were based on quotations from suppliers for equipment and services. It is believed that this spreadsheet model will be able to reliably predict the potential viability of similar businesses in other locations, given data about proposed scale and utilization.

  

Evaluation of the Research

 

In the area of International Development, there is a rich and extensive literature in the area of microfinance – using small loans to assist micro-entrepreneurs as they set up and expand small businesses.  The impact of this approach has been substantial and widespread.  The need for developing nations to participate in a global economy has been emphasized by many governments and by International organizations such as the United Nations.  Widespread access to the Internet and the resources available on it are thought to be critically important.  Many organizations have tried to introduce programs that will increase accessibility to the Internet in developing countries, but few of these programs have proven to be sustainable.  Given the success of microfinance, an essentially business approach to  Development, one has to question why a more business approach to Internet provision has not been widely adopted.

 

This research was sparked by the idea that if it is possible to demonstrate a highly profitable model for providing Internet access in the developing world, it will spontaneously be replicated.  The research therefore set out to understand who the current Internet users are and what they want.  The principal task was to investigate how a viable business could be built, and whether the business model that had been conceived for the purpose would produce acceptable results.

 

The questionnaire that was used to gather demographic and customer preference data was administered successfully, but some of the questions were probably too complex and the author was not content with the quality of the responses to these.  Rather than being left with inconclusive data, an alternative approach was developed, allowing more accurate data to be collected by directly monitoring Internet café usage.  Other than the problem questions, the questionnaire data produced useful results.  The sample size would suggest that these are sufficiently significant to at least provide guidance for other Internet businesses in the same region.

 

The data acquired through direct monitoring of the usage patterns of Internet café users was helpful –especially in terms of managing bandwidth and capacity.  It also highlighted the reliability problems that exist.  The method of data collection was thought to be very reliable as it was completely automated. However, it produced a very high number of detailed transactions.  The decision to categorize the URL destinations visited through browsers was reasonable, although the need to manually evaluate each address possibly introduced errors into the categorization process.  Making consistent decisions about these categories was also problematic.  For example, the decision about the point at which a URL is categorized as pornography is subjective and subject to personal bias.  For this reason the categorization can only be considered to be indicative.

 

The direct measurement of bandwidth consumption to determine overall capacity utilization was also automated.  This measurement was possible for the whole network, and for each location, down to five minute granularity.  It was therefore possible to compare and contrast the consumption of each Internet café and each re-sold bandwidth customer separately.  This allowed one basic idea to be proven: that different user types would have differing and complementary peak usage times.

 

The collection of Supernet’s Profit and Loss data allowed the costs and revenues that was assumed at the planning stage to compared to actual results.  Since this data had been externally audited, there was confidence that the data was valid.

 

Taking all of the research data together enabled a rounded picture of the case study business to be formed.  None of the data collected led to any big surprises, although there was some variance between the planned and actual costs and revenues.  The result of this was that the scale of business needed to achieve viability is slightly larger than anticipated.

 

Reflections and Implications

If the results of this research are correct, then faster internet connectivity is going to continually be in demand.  This is in line with developed countries.  According to Jackson (2006), faster connectivity means that British internet users do more things online and spend more time connected to the Internet.  In the developing world, it looks like investment in widespread infrastructure is unlikely to keep pace with demand.  Only satellite connectivity overcomes these bottleneck problem by bypassing them.   Where VSAT is used,  Internet access is going to require larger scale Internet businesses than those commonly found today in the developing world.  In locations where high cost VSAT licenses and customs duties apply and where expensive electrical generation / storage and management equipment needs to be used, it is hard to see that a viable business can be developed.  This is especially true in the rural areas, where there may not be sufficient  potential users to produce enough revenue.  Countries where such conditions apply continue to look to the international community to assist them in bridging the Digital Divide.  Yet part of the solution is in their own hands. Removing import duties and license fees would improve the possibility that indigenous enterprises could form viable Internet businesses across these lands.

 

There is a possibility that emerging technologies will also tilt the cost:revenue balance, allowing viability to be achieved while operating a smaller scale of business.  Such technologies would potentially include WiMAX (802.16) wireless.  Anything which reliably allows high speed access to be distributed with relatively low equipment cost would be helpful.

 

It is suspected that the viability criteria laid out at the end of chapter three are valid where a business would be borrowing from a bank or similar external source. Money raised within a family (or other close personal network) may not carry the same obligations in terms of interest rates, payback period and expected return on investment.  With more benevolent terms, smaller scale businesses could still thrive.

 

Technologies and economies continue to evolve, and it would be wise to continually work on value innovations that will make a business model more competitive, thus increasing the range of circumstances where it could thrive.

 

Suggestions for further research

 

It would seem that there are further areas to be investigated to improve the viability of this kind of business, especially, if possible to reduce the scale at which the model is viable.  These areas would include the following:

Pricing of Internet café access

 

The price elasticity of demand for Internet access is unknown.  Given that users are clear about their preference for fast access, it would be interesting to determine how much of a premium could be charged for faster access.  Furthermore, if access speed is much higher than the competition, there may be an opportunity to increase the price quite substantially. Some of the individual users were questioned on this informally, and some said they would be happy to pay twice the price for guaranteed higher speed access.  This could quite easily be tested.  The Mikrotik bandwidth management router allows bandwidth allocations to be made to specific computers.  It would therefore be possible to set up a number of computers in an area of the Internet café for premium access speeds and price the use of those computers differently.  Furthermore, access speed availability and pricing could be different at different times of the day.  It might be worth introducing premium services only at times of the day when total bandwidth consumption on the network is normally low. This would then avoid cannibalizing revenue which is normally generated at peak hours.

 

Web Content filtering

 

There has been a suggestion that content filtering is introduced at the egress point from the satellite system on to the public Internet.  This would restrict access to sites that are considered undesirable by some and may be high risk for malware; this would mainly impact pornographic images and video material. Although this would probably deter some users (maybe the same ones that would pay more for faster access) it might be considered that the resultant reduction in problems with computers and the availability of more bandwidth for more wholesome activities would benefit the business, improving utilization overall.  Again, this could be tested by passing data for an Internet café through content filtering for a period, with a survey before and another after.

 

Power management

 

The capital cost of electrical power generation and management equipment and the ongoing cost of power to run an Internet café is not inconsiderable.  It would be worthwhile investigating a range of options for reducing the power requirement whilst keeping the user experience competitive.  An example would be the use of multiple users working from one PC.  A system developed by the Ndiyo project in Cambridge shows promise (Stafford-Fraser, 2006)  This can run 6-10 users from one PC running Ubuntu Linux.  With all screens being low consumption LCD Flat Panels, the overall power consumption is much reduced.  On paper, this solution also reduces capital costs and improves manageability.  This equipment would probably not have the same vulnerability to malware as Windows XP based systems.  

 

Transferability / Replicability

 

It would be good to have the opportunity to test the same basic business model in some different countries to see if the results were similar, or if different factors come into play.


References:

CENTER FOR INTERNATIONAL DEVELOPMENT AT HARVARD UNIVERSITY (2000) Readiness for the Networked World: A Guide for Developing

Countries. [online] Last accessed on 6/28/2006 at URL: http://cyber.law.harvard.edu/readinessguide/guide.pdf

JACKSON, B. (2006) British internet users spend 50 days a year surfing web. Guardian Unlimited [online]  Last accessed on 8/14/2006 at URL: 

http://www.guardian.co.uk/uk_news/story/0,,1839237,00.html

STAFFORD-FRASER, Q. (2006) Executive Summary [online] Last accessed on 9/16/2006 at URL: http://www.ndiyo.org/intro/summary

THOMPSON, A (2005) Dimensions of Business Viability. Murdoch University [online] Last accessed on 6/29/2006 at URL:

http://bestentrepreneur.murdoch.edu.au/Understanding_Dimensions_of_Business_Viability.pdf 

TRANSPARENCY INTERNATIONAL (2005) Transparency International (TI) 2005 Corruption Perceptions Index (CPI)[online] Last accessed on 7/21/2006

at URL: http://www.icgg.org/corruption.cpi_2005.html

WORLD BANK (2006) Doing Business[online] Last accessed on 7/23/2006 at URL: http://www.doingbusiness.org

 



Edited by the author for the web.

© Copyright, 2006  Rob Longhurst (rlonghurst@drasticom.org)

 

 

 


Ó