|
 |
|
BUSINESS BASICS CHANNELS 
|
Judging the effectiveness Judging the effectiveness of your website is perhaps one of the hardest things you need to do when you own a site. There are several standards of measure you can rely on, but in this article we'll explain why you still need to resort to an old fashioned hand method. First off we need to define the criteria by which you can evaluate your site. In this article we'll deal primarily with business related sites. Portals, homepages and informational sites have a differing set of criteria which will not be dealt with here. What do we know about a site? Well as the site owner, your designer and or hosting company should have made a log analyzer program available to you. If your host doesn't make this option available (and most do at no extra cost), then its time to consider switching hosts. The log analyzer provides statistics concerning your website. Plain and simple. The problem is that in the continuous rush to provide snazzier graphics and a slicker program, the analyzer tends to hide key information from you behind a morass of pretty pictures and watered down summaries. Most analyzers will tell you all sorts of fascinating trivia about your site. How many people visited, how many people from each country, how many browser types etc.. Interesting yes, useful? Well only sometimes. Let's digress briefly and talk about some of the more common stats provided by log analyzers and how people confuse what information the analyzer is telling them. DON'T HIT! Some designers, who have few scruples, will artificially inflate the counts for their clients by telling them "hits" instead of actual visitors. A "hit" is a request to fetch a file from a web server. Therefore if you have a page which contains ten images on it, one visitor to that page would generate eleven "hits". Ten people visiting would mean 110 "hits". Hits are not visitors, so don't let someone confuse the two metrics. Furthermore, most analyzer programs report both "hits" and visitors, sometimes in a manner which it becomes easy to mix the two metrics. "Hits" will typically be substantially higher than visitors. For example a one page website with ten images might see 1000 visitors in a month, or 11,000 "hits". The most elementary advice I can impart here is "don't trust the designer or host to give you the information". Accept it if its from an analyzer or available in raw log format. Now having said that, let me assure you that most designers and hosts are honest, and to be honest, as a designer, we don't want to be bogged down handling analyzer details for a client. Most designers and hosts will answer questions concerning the analyzer, but they don't really want to be responsible for delivering the data to you. PAGEVIEWS Pageviews are another statistic which are reported by most analyzers. Its a great little statistic, but it suffers from one fatal flaw. The pageview tells you the number of times all the pages of your site have been viewed. For example, a ten page website might have 1,000 visitors in a month, with each visitor reviewing an average of 3-4 pages each, generates a pageview number of 3,000 to 4,000. Again, its providing an inflated number which doesn't reflect the reality of the site. VISITORS Most people judge a site by the total number of visitors the site receives in a given time frame. The question that begs to be asked is "Isn't this a valid metric to measure a site?". The answer to this is both yes and no. If your site generates its revenue by the placement of banner ads, then obviously the goal of your site should be generating the maximum amount of traffic. But if you're selling real estate, or fly fishing rods or books, then this number still isn't answering your question. The other problem with visitors is the number doesn't reflect the amount of qualified traffic. How many of the people visiting the website are turning into paying customers? What's the conversion ratio? We've built websites where the client insisted on maximizing the traffic at the expense of qualified traffic. If a website is dedicated to selling fishing poles, and 500 sports fishermen (and women) visit it each month, that's 500 possible qualified sales. Bringing in 5000 people, of whom only a small percentage are qualified buyers means a loss of revenue from the site. On the other end of the spectrum, we have clients for whom 10 qualified sale leads in a single month is more than they can handle. Focus is key here. The traffic brought into a site should, and must, be qualified traffic. Sure its nice to see your page counter spin off numbers, but when push comes to shove, the bottom line is you want a customer to walk into that door, not some idle window shopper that pops in out of curiosity, then wanders away. REMOTE LOGGING Finally in regard to analyzers, lets state one critical fact concerning the remote analyzer. These systems are usually provided free by companies. For example, HitBox (http://www.hitbox.com) provides an interesting and unique service which is useful if your site is housed on a free website server. But for a hosted domain we do not recommend it. Your hosting company, if they have set up the server correctly, is automatically logging all accesses, errors and referrers to a log file. This is the most accurate snapshot of the activity which is occurring on your domain. On the other hand, a remote logging system suffers from the problems of net connectivity, server congestion and other similar problems. Should for some reason, the connection between your server and your remote logger go off-line, then all your site traffic during the downtime will not be recorded. Some of the more fancier remote logging systems (like hitbox) rely on a java applet which now means that not only do you have to worry about the connection between your server and the remote logger, but also the remote logger's connection to the visitor as well. Should either of these connections go down, valuable information is lost. In essence, you're paying for a professional level hosting company to host your domain. Most hosts provide log files and log analyzers, take advantage of the best picture you can get. Use what the host provides. Which brings us back to the question "How can we judge the effectiveness of a website?" Well lets start with the basics. Your website is selling something. Be it a product, or a string of products, or some sort of service. Websites that are POS (points of sale) in format are easier to judge because some of the key data you need is right in front of you (only you don't know it yet). POC (Points of Contact) sites are harder to judge because the site generates leads, not sales. And each lead needs to be finessed into a purchase. Again, the data is there, but its a little harder to track. Lets take the case of the POS sites first. Whether you are selling a single product or you are selling a whole catalog of items, most websites which accept orders online have two common elements. They have an order form and a response or results page. Now the key information you want to gleam from your analyzer would be how many times the order form has been viewed in a specific time frame, versus how many times the order response page has been viewed. This comparison yields up some interesting information. This might mean some digging, and you may have to turn to your designer with a little cash in hand to get them to detail exactly what information you want from the analyzer and where to find it. But you need to know, where to find the number of times the order page is accessed, and the number of times the response page has been accessed. Check with your designer, he/she may or may not charge you for this level of information. Remember when we talked about qualified traffic? Well the true qualified individuals are those that click on your order form. Up to that point, everyone is merely browsing. Once you know the number of times the order page was loaded, and you should already know the number of orders you get in any particular month from the site, it becomes simple math to figure out how effective the site has become. Percentage of Orders vs. Loads of the Order form Orders
% = ------------------------- * 100
Order page loads
Now I know you're sitting there wondering why there is such a difference between the number of orders and the OPL (Order Page Loads). Our experience has shown that the number of people loading an order form and the number that actually complete the form can vary greatly. It's almost as if people get cold feet and decide at the last minute to back out. That's one reason, but there could be others. For example, is your form too complex? A daunting, comprehensive form can put people off. Perhaps your sales pitch wasn't clear or compelling enough or worse yet, your order page isn't a secured form. No one should be asking for credit card numbers via an unsecured form. Now we get into the sticky part. What should this percentage be? Well as an old teacher of mine was fond of saying, "Success is a rubber yardstick". Meaning a percentage that might be acceptable to one person isn't always acceptable to another. At this point I would have to say that if the website is generating enough income to pay for its hosting and upkeep costs as well as leaving money left over, you're in the plus column of the equation. But there is always room for improvement. The real question becomes how to improve and what to improve. For instance our Set Sim Pro product currently is seeing a 17.14% conversion of Order Page loads to orders. Thanks in part to the friendly folks at VentureConsult.com. Its not too shabby, but there is plenty of room to improve on that number. But even with a 17.14% conversion, Set Sim Pro is more than meeting the basic requirements of paying for hosting and system upkeep. Now comes the case of too much improvement might actually have a negative impact on a business. While we all dream of bringing in so much business that the business is forced to expand and grow, one must temper the need for growth with a solid understanding that too much growth can be harmful. That old rubber yardstick again. Here are a couple of simple tips that will help you boost your numbers some. * Make sure your have links on every
page to your order page. In the case of POC (Point of contact) sites, the hard data is already in your hands, you just have to recognize it as such. In fact, each time someone has contacted you from your site to inquire about using your services is the same as someone loading that order form. We often suggest to our clients that they use a special email address for contacting them via the business site. Keep this address on a different email program (like the one you downloaded with netscape communicator), this way you can keep your business email separate from your personal email. Its easier to track when you don't have to discount emails from Uncle Frank. So far we have seen the method of judging the effectiveness of the site based on the number of orders it pulls in. But there is still another metric to develop. For this one we need to know the number of times the order page has been loaded, AND the number of times the index page has been loaded. Why the index page? Well while its true that if your visitor comes in from a search engine, they are likely to come into your site on any page, in most cases people tend to back out to the main page of a site before proceeding onto ordering something. And there are far more directory systems which accept only main page urls than search engines which list all the urls in a particular website. Given the main page load number and the order page load, you can develop a percentage which gives you an idea of the amount of qualified traffic your site is generating. For example; Its easy to see that a lot of people are driven to a website by sheer curiosity and aren't qualified traffic. Improving your percentage here can be achieved by (a) fine tuning your site and meta tags for qualified queries at the search engines and (b) checking all the directories your listed in and making sure you're in the correct category. The other method for increasing your qualified traffic involves doing something that many business site owners are not willing to do. That is, turning your site from a hard sell site into an informational resource. Take for example the case of ABCs of Small Business. Had the company erected a website which was merely a listing of their services, it would have been far smaller in size, and more likely to be over looked in the grand scheme of things. By including articles, such as this one, and listing resources, ABC's of Small Business has significantly raised the value of their website. More people looking for business information will be inclined to go to their site, and more importantly to ABC's of Small Business, more visitors will be inclined to use their services because the site lends an air of credibility to the company (not that they didn't have that already). Think on it. If you were in the market for a fishing pole, would you be more or less inclined to purchase from a site which included detailed articles describing the product and how to use it, or the site that just shows a picture and lists a price? There is an added and often overlooked bonus to adding more content to your site. The more pages a website has, the more pages will be indexed within the search engines. The more pages, the better the chances of people finding you. Judging the effectiveness of a site, in our opinion, lies firmly in the goals of the site owner. If the site owner wanted maximum traffic then count the page loads, if the site owner wanted sales, then do the math, or sales leads, again do the math. If you ignore the numbers, or never bother to check, you'll never really know how well you're doing. Websites are a form of advertising where nothing is cast in concrete. If you know what your current status is, you can take steps to improve it. - Northern Webs
|
|