The New Web Accountability by Tim Giles It was the best of times. It was the dodgiest of times. I'm sorry Mr Dickens but the birth pains of the web revolution, as with the French one, also require expression in the superlative. It was unbelievable, unaccountable, indefinable and dodgy as Hell. Good money following bad through the S-bend of e-commerce. Those were the days, but have they changed for the better? It was around about this time at the dawn of the new Century that I first heard the phrase Search Engine Optimisation (SEO). Although back then it was unsophisticated and tended to mean little more than keyword stuffing in page meta tags. This meta information being the initial lines of HTML code that serve to define and describe web page function and format. This early SEO was more geared towards getting eyeballs to the site at all costs rather than pre qualifying referrals. It was the era of dot com companies going to market with inflated values based upon artificially boosted traffic figures. When the snake oil salesmen could reel off estimates of $100 per online member, and still keep a straight face. In the race to IPO things got swept under the carpet. Rules got bent. Ethics got twisted. It was not uncommon to type a high profile company name or brand into a search engine and be presented with a selection of competitors and porn sites as rival webmasters tried to manipulate the system with keyword spamming. Traffic was God. Big numbers meant big valuations. Get to float, cash in then get the Hell out of Dodge. The metrics were flawed as well. The money managers failed to grasp the technology and were taken for a ride. Advertising revenue was based on easily manipulated statistics and was duly exploited. The biggest rort was the use of hits as a measuring stick. Even today people who should know better still refer to hits as being somehow meaningful in terms of judging a site performance without realizing what it actually means. A hit is recorded every time the host server is "hit" with a request for a specific element. Most web pages display a range of components so that for any given page display the server will be hit multiple times. It is therefore quite simple for a sneaky web master to deliberately distort this figure. I can still remember analyzing a competitor's site to find that their nice big header image had been sliced into 100 pieces and so registered at least 100 hits every time it was loaded, distorting the performance figures and hence advertising revenue for their banner ads. Other techniques involved using transparent gifs and JavaScript code to continually refresh page elements. The creative agencies either did not know or turned a blind eye as the cash rolled in until it all hit the fan in mid 2000. A lot of the slick talkers went back to selling Gold Coast swamp land but in many ways the reputation of the industry never fully recovered. That truism about damned lies and statistics could have been written specifically for the web. If it can be measured it can be tricked. Stats mean everything and nothing. The only stat that matters for a web business is the effect on the bottom line. Either your website is saving you money or it is bringing it in either directly or indirectly. Intangibles such as branding are nice spin offs but if this is the only reason you are online then you are in trouble. The web is a pull technology not a push one. Find out what people are looking for in your space and then target these keywords. It is not just about getting traffic it is about getting the right traffic. Attracting people who are in a frame of mind to buy and directing them to your sales team. It is about building specific calls to action into your site and then tracking these. The trick is that times have changed. It is now not just about stuffing your meta tags with every key term you think might relate to your business. Internet search has grown up and evolved as have the search engines. Search companies reacted against the tide of SEO spamming by moving towards relevance listings. Google came out of nowhere and blindsided the industry to dominate search. If you weren't in Google you didn't exist. The old rules no longer applied, or at least, not as much as they used to. The cataloguing and ranking techniques were shrouded in mystique and marketers were subtly pushed towards paid listings. First it was Google and Yahoo then the other search companies went "me too" and jumped on board. It is a trend that continues to define the market. The Internet has grown up as well. It is no longer the side show of the late nineties. It has infiltrated everywhere and increasingly everyone. According to Netcraft (www.netcraft.com), who run regular surveys of web server usage on the Internet, in February 2000 there were at least 10 million web sites in operation (a conservative estimate). In its May 2004 report this result has grown significantly to over 50 million sites. It is little wonder that on the web you can be lost in the crowd and the search engines are very quick to have their hands out for your hard earned lucre. Looking at the main players, Google has Adwords and Yahoo has Overture. New kids on the block Sensis are making noises on the fringes. There is money to be made in paid search listings. Every day more and more people are making purchasing decisions through their mouse and keyboard, whether at work or at home. You might have the best products or services but are you getting this message to them before your competitor? Paid listings can help at least get you on the same page, even if it is in a small advertisement on the side or above the main listings. The key is however to choose the right horses for each course. A smart marketer should choose wisely. Paid listings are not the be all and end all of Search Engine Optimisation. A good "natural" listing will always generate better results than a contrived paid one, however obtaining good rankings are an ongoing process. The paid links are ads and look like ads, however they are at least contextual to the search term rather than the splatter gun approach of other forms of Internet advertising. Anecdotal evidence suggests that people are still hesitant about clicking on these advertisements in the side bars however the Google and Overture statistics (that word again) indicate that a significant number of users of search engines do. It would be silly to ignore this technique entirely however it is equally folly to limit the exploration of search engine marketing to the paid services. The nature of pay per click means that you are in a bidding war for your chosen keywords from the outset. The system works easy enough allowing you to set bid limits and play a range of keyword combinations. How much you will have to pay to secure a top listing will depend upon the popularity of the term. Want to nail down top spot for Tractor Maintenance and it could cost you only $0.10US per click. Try for something competitive like Web Design and you will be stung around $2.54US. Top of the scale is Mesotheliona, that nasty cancer you get from asbestos dust. Fuelled by litigation payouts the top bid is currently $100US per click on Overture (www.overture.com). Unbelievably four US law firms are paying this price. According to Overture 36,695 searches were done for this term in June 2004, not to mention the other possible keyword combinations. Expensive stuff if you get the clicks, it all boils down to how much a click through is worth to your business. That difficult concept for old school web marketers, return on investment. Running a paid campaign will help boost your overall site ranking in the search engines. Doing this alone, however is unlikely to get you the prized top spots for your chosen keyword string. That is, unless you have chosen a very niche term with little competition. In which case you could probably have got there without pay per click. Another key component in this process is to make sure that when choosing the target keyword combination (known as a string) that you are realistic. If you are a small shoe store based in outer suburban Melbourne don't shoot for a highly singular generic term like "Shoes" by itself. Be more specific regarding product and location. The type of person that will respond to any online pitch is likely to know how to use a search engine and use keyword combinations to narrow their search results. Better to target "Ladies Shoes Melbourne" or similar combinations thereof to make better use of your paid listing spending. It is no point boxing out of your division on a global scale if you are targeting local business. Accountable web marketing is achievable, however you do get what you pay for. After the crash when the bottom fell out of the market companies retreated in-house or farmed out work to back yarders that were prepared to work for food. Cutting corners and scrimping delivered cheap sites and even cheaper results. The web site became a necessary evil rather than a valuable tool, the thunder box out the back rather than the en-suite. When it failed to work miracles it was fobbed off as a white elephant. It became a classic Catch 22 situation. Content is and always was king, but once bitten and several times shy the marketing managers, who could have made best use of the technology, refused to spend their budgets on updates. So, at the very time when the search engines started rewarding context and content, the corporate world retreated into safe static options or the short term fix of high profile flash dynamics. Two extremes, neither of which deliver good long term search engine performance. If your site is not optimized then you have no alternative but to pay for referrals. If your content is stale and boring then who is going to buy from you. The site is hamstrung from the start and destined to fail. So the cycle continues. You get what you pay for. That is the e-marketing lesson. That doesn't mean you should throw bucket loads of cash at the nearest pony tailed web designer bouncing on his Pilates ball. What it does mean is that you should make a realistic online plan complete with goals and targets and exit strategy and then commit to it. If you are going to dabble in pay per click then be prepared to fight for a first page listing. If you are number 22 on Google Adwords for your primary key term then your half hearted listing will deliver half hearted results at best. More dollars round the S- bend. Use the tool aggressively. Do not expect miracles from token outlays. Mostly, however the lesson is to use multi channels. The starting point should always be your site. The mistake of the late nineties was that companies tried to trick search engines to deliver traffic. The task now is not to fool Google but to work with it. Search engines continually espouse the virtues of their technology in producing relevant results so work to make your site as relevant to your target key terms as you can. A keyword spammer will soon be exposed for what they are but a site tailored to reinforce its core contextual message will prevail. The key to this is content and it is important to realize that content imbedded in images and flash movies cannot be read by search engines. It might be painful but some surgery may be required. A good URL is also important but it requires a user to be already exposed to your branding. This is no problem if you are Coca Cola but a worry if you are Fred's Smash Repairs. New business is unlikely to come directly through your URL. Think holistically. Developing an effective website is a balancing act between four competing elements, Design, Usability, Bandwidth and Search Engine Optimisation. Try too hard on one and you will impact on another. Jacob Neilsen is often held up as the guru of web site usability however anyone who has been to his site (www.useit.com) and implemented his suggestions without question will soon recognise that he is to design what Dr Pritican is to Christmas dinner. On the flip side how many times have you been frustrated by the inability to print off a key piece of information from the funky flash site that you just stumbled upon, or left in frustration whilst that self indulgent animated splash screen took forever to load on your dial up connection. Similarly a perfectly optimised web site for search engines would consist of nothing but ugly easily searched and catalogued plain HTML text. The art is therefore a balancing act in determining the level of compromise to suit your business needs. This then is the key question. What is your core business requirement from your website? Unless your site achieves this you will not be satisfied so define it and then tailor your site to achieving it. Return on Investment for a website is not just about "paying for the bloody thing." It is about sustaining it at a level that maximises its effectiveness. Set and forget is a recipe for failure and different businesses will have different capacities for achieving this quickly. The specific goals may vary depending on the industry and the individual business concerned. A few points to consider… • Put specific calls to action on your site. The best measure of success for a website is tangible client contacts. Encourage visitors to use the website contact information by offering web only deals or other incentives. • Target your primary selling points and products in your keyword strings and cover appropriate cultural/market terminology variations. (For example Mobile Phone in Australia, Cell Phone in US, and Hand Phone in Japan; or Crayfish in Victoria and Lobster in NSW) • Don't put too much information on your site. People use the web increasingly for research. Put enough content up to convince them of your credentials then encourage visitors to contact you directly for more information and make it as easy as possible to do so. Put your phone number on the site. Monitor and respond to email enquiries immediately. • Set realistic review periods. Different search engines have differing lead times to catalogue new sites and index. A new site can often take up to three months to fully make its way through the system. Be patient … to a point. • Monitor frequently. Search engines and their listing criteria change frequently and often arbitrarily. Tweaks and changes will be required sooner or later. Have a plan in place with your web developer for reacting to any sizeable position drops. Sooner or later your competitors will react and you have to be ready. About the Author Tim Giles is a Pre Marketing Consultant for Enedia (www.enedia.com). Enedia's client's include Ansearch (www.ansearch.com.au), an Australian search engine and directory.