Getting Traffic That Works

There are a few fundamental truths when selling things online. The first is that you need visitors to your website. The second is that you need those visitors to buy something. Anything else they do, such as blog or tweet about you, is just a bonus.

A lot of times people concentrate on getting traffic, instead of on obtaining quality traffic.

This article contains most of the traffic generation techniques I've used in the last few years, along with some statistics on how well they perform. Naturally, your results may differ from these, but they're a good starting point for evaluating different methods.

All statistics are taken from an 18 month period (September 2007 – February 2009) to give as large a sample size as possible. There are nearly 1,000 different referrers for this period, so these statistics are certainly not 100% accurate. They do give a good picture of what works, which is the important thing.

Anyway, enough with the disclaimers. Each table contains four columns:

Pages Per Visit

This is the average number of pages a visitor looks at. A higher number means they look around your site more, so keep an eye on this number.

Adding relevant links to the bottom of pages, or sprinkling them throughout your content can help. Be careful not to add too many links though, so you don’t distract visitors from downloading or buying your product.

Time on Site

The average amount of time a visitor spends on your site. Again, higher numbers are better, as it generally means they’re reading and not just skimming.

Bounce Rate

The bounce rate is the percentage of visitors that leave your site after a single page. For example, a visitor who enters your homepage and then types another link in the address bar without following your links has "bounced". The lower this number, the better.

Download Rate

This is from a goal I set up in Google Analytics to track downloads. A lot of my sales come from people who buy from within the demo, so the higher this number the better. If a source of traffic doesn’t produce any downloads, it’s a sign that I’m advertising to the wrong people.

As you’ll see, there’s a big difference between the best traffic sources and the worst. The most traffic to sodaware.net comes from search engines, but search engine optimisation is a big topic that I wanted to avoid, so this article concentrates on sources that are easier to control.

Paid Advertising

Google AdWords

This is one of the easiest ways to buy traffic. All you have to do is write a short textual advert about your product or website, and then target keywords that people search for. It's quick to set up, and you know visitors are looking for what you’re offering because they searched for it.

On the flipside, there's a lot of depth to AdWords, and it takes a long time to get really great results. Throwing down a few dollars is only the start, and you’ll lot of your time will be spent on optimising your campaigns and weeding out under-performers.

Pages Per Visit Time on Site Bounce Rate Download Rate
3.18 01:24 32.6% 17.44%

Project Wonderful

Project Wonderful is even easier to use than Google AdWords. Create a graphical button or banner, select the criteria for sites you want to advertise on and you're away.

Project Wonderful has the advantage that ads are extremely cheap, as you pay per day, not per click or impression. It's a good way of getting lots of hits, and the campaign option makes things even easier. However, it's quite easy to spend a lot of money and end up with little to show if you're not careful.

Manual bids that are tailored to each site you're bidding on seem to work the best. They takes more work, but the responses are generally higher.

Below are stats for the top three bids I ran.

Pages Per Visit Time on Site Bounce Rate Download Rate
3.52 01:11 61.0% 9.76%
2.42 00:50 59.3% 6.44%
2.08 00:47 59.8% 2.06%

Text Link Ads

According to Text Link Ads, the real power of this method comes from the increase in search engine ranking for your site. They recommend you leave your ad running for several months to evaluate it, as search engines take a while to update.

Over a five month period, I saw no change in my ranking. Having said that, I was trying to rank on very competitive terms, so it may be better for niche phrases instead.

These are statistics from the sites that ads were placed on. As you can see, the links themselves didn’t bring particularly good traffic compared to other methods.

Pages Per Visit Time on Site Bounce Rate Download Rate
1.35 00:05 88.4% 3.49%

Free Advertising

StumbleUpon

Although you can pay for Stumbles, I'm putting it in the free category as you can get good results without paying. Stumblers generally fall into two categories: People who will stay and read, and people who will leave within a few seconds. This dramatically skews the visit time down and bounce rate.

StumbleUpon brings in a big spike of traffic at the beginning, and quickly tales off after a few days. It continues to bring in visitors over time, and sometimes you’ll get secondary spikes. There’s also the chance that visitors will stumble your other content, leading to more spikes.

For the main page, they were quite poor, but for articles and resources it works much better. Your best option is to add tips, hints or guides to your site and have them stumbled.

Articles about unique qualities of your product will perform better, as long as you go easy on the sales talk and concentrate on writing something of value.

Pages Per Visit Time on Site Bounce Rate Download Rate
1.34 00:23 74.4% 0.58%

Article Marketing

This is a simple technique, but it can have good results. The process involves writing articles that target certain keywords. Each article will contain a link to your site in the article footer. This article is then submitted to article directories, which generally rank high in the search engines. For example, if you're selling a solitaire game you could write about tips for playing the game and include a link at the bottom of the article.

It may sound a little bit spammy, but it isn’t if you provide value to your readers.

So far this has been one of the best methods of traffic generation I've used. It takes effort to research and write your article, but the traffic quality is extremely high. An article I wrote over a year ago still sends a good amount of traffic which converts extremely well into downloads and sales.

Article marketing is a topic I’ll be writing more about in the future.

Pages Per Visit Time on Site Bounce Rate Download Rate
2.37 00:46 42.8% 24.28%

Blog Commenting

Find a blog that relates to your topic, and make a relevant comment that contains a link to your website. As long as your comment is useful and the blog gets a reasonable amount of traffic, you'll get visits. Depending on your contribution, you may also develop a relationship with the blogger which can bring more traffic.

The first result in the table below is the result of a single trackback I made three years ago. It's brought several hundred visits (and still brings in a dozen or so a week), but the traffic doesn't convert. This technique works much better for getting blog readers rather than customers.

Pages Per Visit Time on Site Bounce Rate Download Rate
2.60 02:28 48.3% 0.00%
1.83 00:09 66.7% 0.00%

Forum Posting

This is similar to blog commenting, but the link to your site goes in your forum signature. Again, the benefit here comes from generating an online presence, rather than raw traffic power.

These are the top three forum referrals to sodaware.net

Pages Per Visit Time on Site Bounce Rate Download Rate
2.39 01:28 73.4% 1.06%
2.96 01:41 62.7% 0.00%
1.36 01:21 72.7% 0.00%

"How To" blog articles

Think of a problem you face daily, and write a blog article about solving it. Its simple, quick, and if you get the title (and problem) right you can get a lot of exposure and links.

I get a lot of traffic to one "how to" in particular, but the vast majority of visitors don’t stay for anything else. This makes sense, seeing as they're searching for a solution to a problem and not for games.

The best advice I can give is to treat this method like writing a recipe with one of your products as the ingredient. This way you’re providing something valuable, and also encouraging downloads and sales.

Pages Per Visit Time on Site Bounce Rate Download Rate
1.04 05:29 94.3% 0.00%

Link directories

I tried an automated submission programme a while ago, but I’m yet to see a single visitor from any directory.

dmoz.org is worth submitting to, but as it's human edited it can take a while to get added.

Pages Per Visit Time on Site Bounce Rate Download Rate
0 00:00 n/a 0.00%

Blog Carnivals

I've entered blog carnivals as well as hosted them, and the results have been a little mixed. Hosting brings in a decent amount of short term traffic, but I suspect it's mainly people who submitted articles that visit.

It’s a good way of getting a few more readers for your blog, and be good for spreading your site around.

Pages Per Visit Time on Site Bounce Rate Download Rate
1.50 00:21 75.0% 0.00%

Game Buttons

I tried these out in my $19.95 experiment and absolutely hated them. I think they cheapened the look of the site, and most of the banners contained images I didn’t really want to see. The other problem is you have to make a lot of impressions to generate any decent amount of traffic.

Pages Per Visit Time on Site Bounce Rate Download Rate
2.84 00:59 41.3% 13.99%
1.67 00:21 71.4% 0.00%

Free Developer Resources

The "For Developers" section is where I put articles, source code and tools for other developers to use. It didn't bring in much traffic to begin with, but posting about some of the resources on forums helped.

Pages Per Visit Time on Site Bounce Rate Download Rate
2.43 01:39 73.5% 0.86%

WikiPedia

This is a five minute method that works well, but if done incorrectly you’ll end up looking like a spammer, so use it wisely.

Find an article on WikiPedia that relates to something on your site, and add a link to it in the "External Links" section of the page. Remember that WikiPedia monitors links that are added, so if you don't add value to the topic you'll find your link nuked (and you could also be banned). For example, my interview with Hanako Games is on the Hanako Games WikiPedia page.

As with text link ads, the real benefit here is getting links and increasing your search engine ranking.

Pages Per Visit Time on Site Bounce Rate Download Rate
1.31 01:03 76.9% 0.00%

Traffic Exchanges

This one is a little hit and miss, although it's not as bad as some of the other methods. You add your link to a directory, and its position is dependent on how many other sites you visit. For example, visiting 30 sites may put you on the first page, which will get you more visitors.

The downside is that most people are only visiting your page to increase their own rank, but the traffic isn't as bad as I expected.

Pages Per Visit Time on Site Bounce Rate Download Rate
1.42 00:27 81.4% 1.53%

Conclusion

This is just a small sample of the various method of traffic generation techniques out there. If you’re trying to increase your site’s traffic, just remember to measure what your visitors are doing, not just how many you’re getting. Sure, it’s nice to see big numbers in your hit counter, but that means nothing if they don’t stay and buy from you.

2 Comments

Post a comment

org-mode tags allowed: /italic/, *bold* and =code=