eXtropia: the open web technology company
Technology | Support | Tutorials | Development | About Us | Users | Contact Us
Resources
 ::   Tutorials
 ::   Presentations
Perl & CGI tutorials
 ::   Intro to Perl/CGI and HTML Forms
 ::   Intro to Windows Perl
 ::   Intro to Perl 5
 ::   Intro to Perl
 ::   Intro to Perl Taint mode
 ::   Sherlock Holmes and the Case of the Broken CGI Script
 ::   Writing COM Components in Perl

Java tutorials
 ::   Intro to Java
 ::   Cross Browser Java

Misc technical tutorials
 ::   Intro to The Web Application Development Environment
 ::   Introduction to XML
 ::   Intro to Web Design
 ::   Intro to Web Security
 ::   Databases for Web Developers
 ::   UNIX for Web Developers
 ::   Intro to Adobe Photoshop
 ::   Web Programming 101
 ::   Introduction to Microsoft DNA

Misc non-technical tutorials
 ::   Misc Technopreneurship Docs
 ::   What is a Webmaster?
 ::   What is the open source business model?
 ::   Technical writing
 ::   Small and mid-sized businesses on the Web

Offsite tutorials
 ::   ISAPI Perl Primer
 ::   Serving up web server basics
 ::   Introduction to Java (Parts 1 and 2) in Slovak

 

eXtropia FAQs
Small and Mid-Sized Companies on the Web FAQ  

Note: This was written in January 1996

I'll begin with cost as that is most likely your primary question. My "guestimate" is that the creation of a web site would be absolutely under $2,000 and probably under $1500.00. You would also have to pay about $50.00/month to rent virtual server space and $50.00/hour for maintenance, changes and expansions.

Questions you might have regarding the guestimate:

Q: Why is it a "guestimate"?
A: I am primarily a freelance web-programmer/webmaster and not a freelance web-site developer. Having only designed one web site in my career as a freelancer, I have little experience in the marketplace regarding freelance rates. But I have been around for quite awhile and I do pay attention.

A full time webmaster could easily run you $40, 000 - $80,000/year, but for small and mid-sized companies, it does not necessarily make sense to hire a full time web developer. Instead, one should expect to hire a freelance web designer to create the initial site and then to maintain it on a project by project basis.

Obviously, freelance rates are extremeley diverse. You might hire an undergraduate hobbiest for $10/hour, or an experience multimedia artist for $200.00/hour. When I say that you should alot between $1,500.00 and $2,000.00, I am assuming that you will seek out the average range, for it is unwise to hire the undergrad and unfeasible to hire the professional.

If you want to do further research, I recommend the classified ad section of Computer World or the USENET Newsgroup comp.infosystems.authoring.html.

Q: What does the $1500.00 buy?
A: The flat fee will purchase some amount of pages that you and your designer agree upon after you talk about your goals and priorities. I "imagine" your initial goal will be to get your regular print advertisement on the web; perhaps some information about your company and its products, information on how to order, and perhaps some useful content that you could provide the web community to attract the wandering web browser.

However, don't think that it is a matter of scanning your material and translatting it to hyper-text markup language (HTML). This is why it is crucial to pay the extra money to get an HTML "Artist". The web is an entirely new medium, demanding its own unique form of presentation.

In the June 27, 1996 issue of the Wall Street Journal, on Page R27, Patterman writes, "The biggest mistake people make is that they don't treat this as a new medium. It would be as if a magazine wenty on cable TV and just flipped pages."

I cannot agree more. If you do not get the right designer, your page will flop no matter how much money you sink into it. Thus, for your $1500.00, you should expect to see a web-based version of your print material. Do not accept a series of pages mirroring your print literature, and do not bind your designer by forcing him to stay with the old standards.

Q: Why is there $100.00/month charge and what does it buy me?
A: I whole-heartedly advise that you rent space on a "virtual server", and $100.00/month is absolutely the best deal that I have seen as of June 1996.

It would be possible for you to create and maintain your own web server onsite.

First you would have to buy the hardware and, if you don't use public domain server software, the sofware too. This could easily cost you over $5,000 for the server, the telecom extensions and the peripherals.

However, you would also need someone to set up and maintain the site on a full time basis. This is not a trivial process. Maintaining a web server involves software and hardware configuration, maintenance, trouble shooting, and upgrades, security issues, user support and any number of bagaboos which can pop up and grey your hair.

The only reason that you might want your own server is if 1) you already have an information systems department and you can roll web services into thier job requirements, 2) you are planning on making the web your main outlet and you plan on doing a significant amount of volume or 3) you are a big enough company with enough capital to throw around.

(BTW, if you do get your own server, I recommend you go with the traditional UNIX system. Both Mac and PC operating systems have their own motley of web server software, but from experience, NT and Mac servers are very limited and clumsy when it comes to CGI and TCP-IP applications like mail. As a result, few CGI applications are written native to those platforms. We'll see what the future holds, but for the time being, UNIX is still the best).

Finally, you would also have to purchase a T-1 or better connection to a local telecom company that offers digital transmission. Usually, this is incredibly expensive. In the US, I think it costs over $2,000 per month to rent such a line, and they are only available in large metropolitan areas. An ISDN line is also a less expensive option, but you will want the highest bandwidth you can muster.

In my experience, most companies rent space from an Internet Service Provider (ISP) who provides a high speed internet connection and alots a certain amount of disk space for each client. They also provide administration services for the servers which they host clients on. ISPs make money by hosting many clients who all share the costs of the connection and server administration. For a small company, it really does not make sense to do this by yourself, it is better to spread the costs with other small companies.

When choosing a service provider, there are a few things that you can use to discrimiate. Firstly, you should know how many clients they serve and how many servers they use to host those clients. You do not want to rent space on a server which is overloaded and which has no backup. You should ask the ISP what is the average usage of processing power daily and what means they use to backup data and backup the whole system if their main system goes down.

You should also know what kind of internet connection they have and the percentage of bandwidth typically used during the day by all of their clients combined. Afterall, it does no good to have a T-1 service provider if they are already maxed out. You may as well be using a 14,400 modem.

Of primary importance is that they provide a useful amount of customer support. If they are too large, you will not get the support you need. If they are too small, their support may not be worth beans.

Along those lines, you should find a service provider that will support Common Gateway Interface (CGI) web-based applications, Server Side Includes (SSI), Java, and other web goodies. Most likely, they will charge for consulting for the bells and whistles, but so long as they provide the means, you can do it yourself or hire someone more cheaply.

What is crucial is that their development environment is clear and easy to understand for freelance programmers you bring in. You may want to ask the ISP to refer you to one of their clients who uses CGI so that you can speak first hand with other clints who have had to work through it. As a side note, CGI development almost requires "Telnet access to your shell account". Many ISPs only grant ftp access. If this is the case, don't even bother considering them. You should have ftp AND telnet access to your site, as well as 50 - 75 megs of disk space and a local email account to process things like client orders or feedback.

Finally, if you are going to be taking orders on the web, you may be concerned about what levels of security they offer. A secure server encrypts data so that credit card information remains confidential between you and your clients.

You should ask them what type of security they offer and any specifics that you will need to know when you uitilize security (ie: aliases for secure paths, changes you might need for your url like http vrs https, and what directories are secure vrs insecure).

You should also ask what precautions they use to protect data from other clients on the same system. It does no good to secure yourself against the greater web only to leave your back wide open from competitors or hackers on the same server.

Finally, you should try to find a service provider that offers PGP service, because however secure your web server is, if you email orders from your web server to your admin, the information is accessible. PGP is a program which allows you to encrypt mail so that orders are taken securely and transferred securely. (Most companies fail to take this into account and for all their gaurantess of security, are totally open).

If the ISP you talk to cannot answer these questions, move on.

There are a host of virtual server service providers that you could choose from. I recommend you check at the USENET Newsgroup comp.infosystems.www.providers.

Q: What is the $50.00/hour for?
A: It should be calculated into your expense that your page will undergo many changes after it is set up. The technology of the web is changing at an incredible rate. As time goes on, you will want to add as many bells and whistles to your site as possible, taking advantage of emerging technologies like JAVA, or Frames, or VRML. $50.00/hour is the kind've rate that you can get a grad student for. I think most web-development companies work for more like $150.00, but if you find "starving students" you can get quite a discount. Again, I recommend going with the established multimedia artist rather than the starving student. Considering how cheap web defvelopment is compared to other forms of advertising, you can afford to be extravegant and go with the best. However, as with all else, the starving student may actually be the better designer.

CGI programs will be more expensive. For example, if you want to put your database online or create a groupware aplpication for your distributors globally, it may run 2,000 - 3,000.00 per project depending on the complexity. If you want to see examples of CGI applications available, try http://www.extropia.com/Scripts/ and if you are looking for freelance CGI consultants, try comp.infosystem.www.authoring.cgi.

Another project which you may want to contract out is that of proactively adding your web site to search engines. Most engines have "robots" which seek out the web for new sites which they can index. However, you may want to actually submit your web site to the search engines to more quickly get your site indexed.

Along those lines, a good marketing strategy is to proactively seek out related USENET groups and Listservs on which it is "appropriate" to advertise, and let the participants know of your existance. It is crucial that you do not "spam". Internet communities should only be approached if they are looking for the sort of information you want to provide. Thus, an internet marketting person, who is savvy with USENET is crucial for your companies reputation and for the efficient distribution of your information online.

Whatever you hire a freelance consultant for (HTML, CGI, advertising) you should make sure they know what they are talking about. Ask them what books they have read or what sites have they designed. If they supply you with a sample URL, you should check it for errors at Harbinger (http://www.harbinger.net/html-val-svc/)

Also, ask them for samples of their code. Does it look clear? Do they code in such a way that others could quickly take from where they left off? Ask them what "standards" they use to develop. Do they use all capital letters for commadns and arguments? Do they indent for lists? No standard is necessarily the best, but some standard is essential. Your designer must be as anal anal as possible when it comes to following the same style for every page. If you know nothing of HTML, you can probably find good samaritans online who will help you critique, especially if you pay them a small fee.

Q: What else must I provide?
A: It really depends on what goals you have.

The easy answer is that you should provide all of the print media material that you currently use for advertising. You should also provide any company graphics/logos/type faces you've had made. These will be transferred to HTML so that they will be available digitally.

It is best to supply text as plain ascii; perhaps have someone type in the text to an email. A hardcopy should also be sent so that the designer can see the original layout. Graphics should be sent digitsally in a common format like .gif, .jpeg, or .pict.

The hard answer is that you should provide "vision". Web development is an art. And like any art, the best work is the work that "fits" intuitively. There is a billion and one ways to present a company's information, but there may only be one way that corresponds to the "soul" of the company.

In order for the web designer to really create the site that is right for you, he/she needs to know the soul of your company. It is helpful for you to do your best to communicate this soul to the web developer. Why did you found the company? What are the goals of the company? How does the company relate to its client, society and world? You should be prepared to talk wiuth the designer for a good hour if possible.

Such knowledge could help the web developer decide whether your page should use red text on a black background or earthy colors on a green background. Perhaps you could imagine what kind've different moods these two different color schemes would set. Not being a graphic artists myself, I do not know many of these graphic tricks used to communicate the soul of a company, a good one would...but it would certainly at least help me to know.

You should also make time to define your project requirements and the time limit. It is likely that this will be extremely amorphous at first since you won't be sure exactly what you want. Thus it is extrea worth it to discuss this with your designer. Help him/her create a requirements list and a project lifetime summary. Use your intuition, leave no room for ongoing projects, define what it is you hope to see and make your designer spell out exactly how he/she is going to meet those desires and how long it will take.

Time. Most web sites can be initially built in about one month. Let this figure guide you but not bind you.

Q: What will the designer do in a month?
A: Some of the work time is spent in digitizing web graphics and in translating text (informational brouchures, order forms, employee bios, etc) into Hyper Text Markup Langugae (HTML). This is not a terribly time intensive process, but it is commonly thought to be. The truth is that with modern HTML development software and graphics programs like Adobe Photoshop or Corel Draw, the text and images can be developed very quickly. A good designer will have an arsenal of tools available to him or her which can be used to translate mass documents or images to HTML native format.

What really takes the brunt of the time is experimentation and visualization. As I have said before, web development is an art. It is an art which takes a great deal of experimentation and on the fly manipulations of images and text.

It is reasonable for the designer to take a whole week just to "think" about the site, musing about various ways it might be arranged and integrated. This time is very important. You will know you have found a good designer if they take this time. Rushing straight into a project leads only to sloppy directory structures and a confusing nest of links with little internal logic or flow.

The next week might be spent converting the text and images you have supplied.

Thus, it is reasonable for you to ask for a demo after a couple of weeks. However, expect the format to change drastically over the following two weeks. This evolution/fermentation is essential and important for the birth of your web site, so do not stunt it.

After the site has been competed per the intial requirements, most changes that you will want will probably take under 5 -40 hours each, thus, most companies hire freelance consultants or web development companies on a project by project basis rather than hire a full time staff.

Time will depend on the complexity of the project, but most things can be done in spurts of 5 hours (usually less), provided that your intial code is clear and well written. Obviously it is hard to make estimates without knowing anything about your product, but I think these are good "general" rules.

Q: What should I expect to see by the end of the month?
A: In my opinion, you should have a web site which is globally accessible 24 hours/day which distributes your information in an effective way to potential clients, partners, and employees.

By "distributing information in an effective way" I mean that...

  • your information should be accessible quickly (relative to the slow speed afforded by current technology). Users should not have to wait an unbearable amount of time to download the graphics on your page. This is a benefit of going with a company that specializes in graphic arts for the web. Graphics and text must be balanced carefully.

  • your information should be presented clearly and logically. Users should be able to get to the information they want quickly and intuitively. Extracting information should not be chore.

  • your information should be presented in a way which entices users to read more and become involved. Extracting information should, in fact, be a pleasure. Again, a real web developer will be able to blend advertising, art, and technology into a "Real" place to visit.

Now, the most difficult questions of all. Why should you have a web site in the first place! You have already addressed this to a limited degree because you have come to me for advice. But, you probably have not really answered the question for yourself, even though you want to move forward and learn more in the meantime. Let me try to address some of the questions which may have stumped you.

Q: How do I measure returns on my investment?
A: At this time, there is no dependable, statistical method for gauging returns.

Many sites use "hit counters" that keep track of how many times someone has accessed your page or "site statistics applications" that generate reports on who visited, what they saw, etc. Such programs are not generally considered part of web page development as they require programing far more complex than HTML.

However, often interesting and sometimes useful, hit counters and site statistics are not good measures of how profitable your web site is since they cannot be correlated to sales. For example, there is no way to tell the difference between someone who accidentally accessed your page, and someone who became a hundred million dollar client.

Hit counters can be supplemented with other programs like a client "Guestbook" which might allow clients to leave messages about how useful they found your site or other information useful for a "qualitative" analysis of your site's return. Similarly, you could implement a bulletin board system so that you could maintain an interactive dialogue with your clients enhancing customer service or technical support. Hopefully, at very least, your email communications with your clients will rise.

Another measure of your site's return might be found in the number of other pages which refer to yours (such as search engines). If you are listed everywhere, your site and your company's name will have higher visibility. But again, this is not very quantitative.

Faced with this situation, most companies have looked for other justifications for having a web site for not everything a company does is statistically correlated to profits.

The most common justification for developing a company web site is that of advertising. In traditional advertising a company assumes that its opportunity costs are greater than its advertising costs. For one, advertising will bring in customers that would otherwise never have known of the company's existance.

Just the high profile of your name caused by advertising can increase profits. Clients may buy your product just because they recognize your name.

Advertising has a defensive nature too. If your competitors are advertising, you may need to advertise yourself just so that they do not become the "standard" in the eyes of the market.

Though the web is new, don't expect your competitors to be sitting on their hands much longer. Everyone will be colonizing the web in the next ten years at a break neck pace, attempting to set up their shop before others steal the best "real estate". Even if a web site was run at a loss, if the other companies do it, you might have to bite the bullet, just to keep the battlegorund in stalemate.

Advertising has an even more important role for small and mid-sized companies who cannot compete with the economies of scales of largeer companies and who must reach out to niche markets unaddressed by the larger companies. It is crucial that you communicate with these niche markets. Unfortunately, traditionally, advertising to niche markets has been difficult. Mass advertising via mailings, TV, or radio is expensive because most of the people who you advertise to are not even closely related to your niche market. Targetting advertisment is extremely difficult.

Advertising on the web, however, has some advantages over traditional mediums.

Advertising on the web allows smaller companies to more efficiently extract profits from the global market. What was once monetarily infeasible (worldwide advertising) is not commonplace. For a one time setup fee ($1500.00) you have a worldwide accessible add permanently. Further, because of the nature of the web, those that are interested in your niche will come to you rather than you having to find them. As you discovered on the search engines, it is very easy for cliennts to find the information they want. If you come up in the searches, they will find you too. Smaller companies can thus extract profits from niche markets unaddressed by the larger companies.

Obviously, this means that you are going to have to learn how to maximize this ability. Thus, the decision to move wuickly into the new technology is a good one because you will need a large amount of "organizationl learning" when it comes to figuring out how to evolve your comany in the new global economy. All companies will be forced to change, and those which set themselves up to learn early will fare the best in the long run.

Related to that is the fact that by creating a web page, your company will present an image of cutting edge technological stregth. Yet this is another reason to do it well the first time, you don't want to look like a loser company, you want to come out strong and interactive, never showing weakness, insecurity, or naivite.

Q: What are the benefits of the web besides as a more efficient medium for advertising?
A: Eventually, a web site can allow clients to buy directly from you. Bypassing distribution channels, this can greatly reduce your costs such that you can lower your prices and more effectively compete with the larger companies. Many companies are utilizing online warehouses or shopping carts which allow clients to actually browse the companies product offerings and order them while online.

It also provides a medium to enhance the relationship between you and your clients who can easily, and at no cost contact you electronically for product information and post-purchase support. The company Saturn has put a great deal of effort for example, into building client groups and creating an image of a "Saturn community".

Once you have created your public face, you can move to enhance your inner workings. An internet web site can easily evolve into an internal "intranet". Such an intranet could be used to manage inventory or facilitate internal communications between R&D, manufacturing, management, and distribution with such groupware applications as a database, calendar, real time chat, or a bulletin board system.

Q: How many people will "actually" be able to see my web page?
A: Well, internet sales are still farily concentrated in the US because that is where the internet is most well developed and because internet technoogies are still biased towards the English language. This audience will also typically be upper middle class. male, and college educated. Thus, if your products are geared for another niche, you should not be expecting much for another two or three years.

I say two or three yars because that seems to be a comon estimate regarding when the internet will infect the Asian and European markets. Many companies, I studied Thailand at GW, are investing huge amounts of money into information/telecommunications infrastructure, and because such development takes a fraction of the time it takes to develop older infrastructures like roads or railways, the information revolution looms right around the corner. You do not have as much time as it seems to prepare for the revolution. Countries like China, Thailand and Singapore are going straight into fibre optic, bypassing twisted pair technology altogether. Thus, when they get online, they will be doing so with the latest and best technology and will be able to take full advantage of web-based technology immediately.

Thus, "getting your feet wet "has a very serious side to it. You won't have long to wade because the flood is coming.