William Charlwood

I released my new video course – Blog Email Alert – about 36 hours ago with a small promotion to one of my smaller lists. I never like to launch things to a big list until I have had a few successful sales go through so that I can be sure the sales order processing side of things does what it’s meant to do.

So far no problems though so I’m pleased. The other thing I’m pleased about is that it is the first time I’ve used MuVar with any degree of success. MuVar is an interesting multi-variate testing tool that lets you break down web pages into sections, create different versions for each of those sections and test how well they perform. So for example, you can try out 6 different headlines with 3 different headline fonts and 2 different guarantees etc. and then it will display various combinations of these items whilst recording the outcome in terms of a sale or a no sale. It then uses this conversion data to display the most profitable variable combinations more and more thus increasing your overall conversion rate over time.

MuVar is not as easy to use as it should be – for one thing, you can’t easily edit versions of a variable once you’ve entered it and so if you put a typo in the headline for instance, it will only disappear from the sales page if it turns out that it reduces conversion rates. Now a purist might say that testing typos is a sensible thing to do. After all, who knows if typos are good or bad for conversions. We might think we know but we could be wrong.

But there are other things you might want to put into your sales page that you can’t afford to get wrong – such as using scripts. When you create a sales page you might want to add a script to it but forget to do so until you’ve already entered a number of variables into MuVar. For example, I use statcounter quite a bit to track visitors and it requires a piece of JavaScript to be added to the page. However, if I decide that I want to use statcounter with MuVar after I’ve already set up a page it is a slog. It should be a case of just editing a variable.

The other problem with it is that until you have some data, some of the reports seem to go into endless loops. However, as soon as sales start to come in, this gets sorted out. I also read somewhere (can’t find it now) that you need run the batch processing function repeatedly if you want all the variable optimisation to take place. I think the comment I read was that it didn’t optimise all variables at once to stop overloading your server. Personally I think it should self-optimise after a pre-set number of actions rather than wait for you to do it yourself. Having said that, it may already do this – it’s just that the set up videos don’t make clear whether you need to run batch processing for it to optimise or whether you just need to do this to generate reports.

In terms of value for money I think MuVar is good but a little bit more attention to making it user-friendly would make it excellent.

Looking back over a number of years of product creation and product selling, it is clear to me that the three most profitable things you can do to grow your own information-based online business are

1. Create good quality information products with a known market
2. Establish good relationships with a wide range of JV partners in those markets
3. Write effective sales pages for your products

You do need to master some of the basics of selling things online: setting up order processing systems, product delivery systems and registration systems to capture the contact details of people who are interested in or who have bought your products but this is just the plumbing of the internet and many sites exist to help you along the way.

Once you’ve sorted the plumbing out, building a portfolio of products and working with JV partners is the best thing to do. JV partners are always looking for good products to sell to their lists to make easy commission from so if you approach them with a sensible proposal that really is “win win” you’ll both win.

Of course, a poor product will destroy a JV relationship for two reasons: too many refunds will slash commission payments and at the same time will erode the standing of your partner in the eyes of his subscribers which will make it harder for him to sell anything when he next sends out an email.

So focus on good quality product creation and then take those good quality products to people with lists.

That then leads onto effective sales pages and the most productive method of sorting this out is to study pages that you know to be successful and copy their style. Then you need to test alternatives: try changing the headline, the images, the body copy, the order link, the guarantee, the font, the font colour. Everything.

But people don’t because they can’t be bothered. Yet testing is where you can make half of your money.

I’m using a testing system on my blog email alert site where I sell a video course that shows you how to send out email alerts every time you update your blog. The system tries out different combinations of the key elements in the sales page and uses conversion statistics to work out which combination is the most profitable. It then starts to show that highest profit combination more of the time.

The outcome of this is that the conversion rate for my site will increase over time. The stronger it gets, the more my JV partners will be interested in promoting my video course and the more money we will all make.

I took a look at some interesting numbers this morning relating to Traffic ReGenerator, a new service that I mentioned in two recent emails to my lists.

The raw stats are as follows.

Of my readers (and I’m sorry to treat them as a statistic here) 185 signed up for more information about the service when they visited the home page. A further 91 people signed up when they were about to leave the page.

That is an extra 49%.

Why did they do this?

Because Traffic ReGenerator gets more people to take the action you want them to take.  It detects when people are about to leave your page without taking the action you want and it then gets them to take that action.

So if you want people to visit another page on your site where they can click AdSense ads or buy something from you instead of leaving, you should at least take a good hard look at this service.  The statistics from this case study speak for themselves: by using this system, my opt in rates jumped 49%.

To speak statistically for a moment, the data sample is also big enough to be significant.

In other words, a statistician would say that these results are likely to be repeatable with a strong degree of confidence.

That’s math speak. Not marketing speak.

Clearly I can’t speak for sales rates or AdSense clicks because I haven’t tested it with them. But for opt ins, it works and it works well. I see no reason why it shouldn’t work as well for other actions.

Worth a trial in my view. Traffic ReGenerator