Better Visibility/more Visitors But Not More Sales, What to do when more visitors don't bring more sales Options

[AskJoanne]

Hello all,

Looking for advice... Has this ever happened to you?

I have several clients whom I have been able to help get better visibility for their websites, have improved their PPC campaigns (higher CTR's, lower costs, etc.) and brought more traffic to their sites. However, they are still not selling as much as they think they should be. What can I advise when they tell me that better rankings & more traffic are not bringing them more sales?

Thanks for you input. I appreciate it.

Joanne

[Randy]

You're about to get into a whole new area of SEM Joanne. It's called Conversion Analysis, and is something most sites don't get involved in and should. In plain English it's a concept of how one goes about converting more visitors into paying clients.

Lots of information about Conversion Analysis and how to construct conversion tests at MarketingExperiments.com, grokdotcom.com, and even Google has gotten into the field with some free tools and their Conversion University. They've even put up videos of one of their Conversion Univisity sessions on YouTube.

[rolf]

It's called Conversion Analysis, and is something most sites don't get involved in and should.

I've just started getting more involved with this myself (Thanks Randy et al) and have set up my first test in Google Website Optimiser.

We found, unlike your own situation, that increased traffic has brought increased sales, but only at the same percentage. This is all well and good, but to be happy with that we have to assume that our site couldn't possibly convert any better than it currently does - which I know is not true.

So anyway, long story short, we find that the suggested approach fits for our scenario as well as it fits yours.

I haven't got any practical advice at this stage, just a bit of moral support in that some of this stuff looks a bit complicated and scary to start with, but now I've got a handle on it I can see that all the principles are very simple and logical. Don't be intimidated by how much stuff there is to learn, just try and break it down into smaller steps/goals and it all becomes clear.

Good luck :-)

[Randy]

Kudos to you Rolf, for taking what I agree at first looks a lot like a leap of faith. It's always good to welcome a new convert ... pun intended. (IMG:style_emoticons/default/giggle.gif)

And you're exactly right. The hard part early on is to not get caught up in the minutae. Much like with SEO you need to process what makes sense when it makes sense to you, break it down into digestible bits and pieces, and use what you understand so that you get started. The goal should never be to try to grasp it all at once.

[AskJoanne]

Thank you!

It all seems a bit overwhelming, but I guess that kind of analysis is the only way to find out what's really going on.

Thanks for the info on the Google Videos!

Joanne

[Jill]

Joanne, sounds like a good upcoming SEMNE topic, as I'm sure you're not alone on that one!

[Ahmed]

Analytics ruins the party thats why people don't like to do it. (IMG:style_emoticons/default/tongue.gif) Who wants to see a 100% bounce rate? oh god. But jokes aside.. it has to be done even though it might be painful to see some of the results. Its best to learn as much from it as possible.

Other cool things related to analytics are heat maps and recording software. You can use them to see exactly what your users see its just incredible. You can test one out at clicktale.com. The hardest part about all of this data/video is analyzing and understanding them though so good luck (IMG:style_emoticons/default/smile.gif)

[Randy]

A year or two from now I may have a good handout booklet for you on the subject if you want it Jill. (IMG:style_emoticons/default/wink1.gif)

I've got a few new and different types of sites (one e-commerce store on an x-cart platform, one a subscription site and one a referrer/info site) that are all either about to open or in some stage of development. I have this wacky idea I have stuck in my head to set them all up with the free GoAn and WSO tools, then keep what amounts to a daily diary of what I see, what I changes I make, why, what works and what doesn't, etc. I may even open it up a bit, by taking suggestions from others as to what to test and make it something of a (small) community project.

First time I've ever considered doing such a thing., I usually keep my own sites close to the vest. But I think I've come to the conclusion that starting what amounts to an open source conversion analysis project of real world sites could be a useful resource to get more people testing. If nothing else, it should take some of the mystery out of all of it.

I haven't convinced myself 100% to do it yet, because it's a lot of extra work for yours truly to document everything and the whole concept runs directly counter to my normal mode of operation. But I'm leaning towards doing it so that others can have a good foundation from which to start their own testing. It's not like I don't already own a couple of domains I could use to publicize the info. (IMG:style_emoticons/default/lol.gif)

[oneofthe3lions]

Just start with the basics..

Look at what pages you are losing a lot of visitors. Check the bounce rates for each of the pages, if one page in particular is losing more viewers then alter the page to see if it helps.

Set up conversion goals on your ppc as well as your natural.

See the popular pages viewers want and make those pages more accessible. using site overlay for this is good.

Initially you want to see the way that visitors go through your site and where you lose them.

Low conversions can mean a poor page, badly designed, mis information , uncompetetive prices, a variety of reasons.

or

It could also mean that you are getting traffic that is too generic and will never achieve good conversions, look at your target phrases and niche them as best as possible so that you know they definatly are searching for your goods, services.

[Randy]

In additon to what oot3l's laid out, I'll warn you now that you should 100% expect to make mistakes. Be willing to accept those and learn from them.

To give you a brief story, several months ago I totally revamped the design and content of one of my sites. I'd already been testing things on it for ages, and tried to extrapolate some data I'd already seen from test for different things and incorporate some new concepts into the revamped design and brand spanking new copy.

Lo and behold, I was wrong. Totally wrong as it turns out. (IMG:style_emoticons/default/lol.gif)

I took a site that had been converting at around a 14% clip for all new visitors and managed to decrease conversions back down into the 4% range. Same keywords, same basic traffic numbers, just a new page design, a changed navigation structure and some "better" copy.

It took me a couple of months to identify the issues, but with further testing with the new design I eventually did. The new design was fine, overall. The new copy was great after it was all sorted. The enitre problem tuned out to be that I'd complicated a little of the navigation (needlessly) and inserted some confusion into the buying process by introducing some things that just didn't need to be there.

At the end of the day it all turned out okay. This particular site is now converting at around 22% of all new visitors, which is quite healthy IMO. But my initial attempt during the redesign could only be categorized as a complete and total failure. Thankfully I didn't make a rash decision and simply junk the whole thing and write off the time and expense I'd already spent on the revamp, because my initial data turned out to be pretty good. I just didn't interpret it well enough to keep myself from introducing other problems into the picture.

Be prepared for some failures. They'll happen. Accept them, learn from them and move forward.

[chris1]

Set up conversion tracking and make sure you get as granular as possible. You must track down to the keyword level if you want to be successful in paid search. Test ads mainly for conversion rate, with CTR being secondary consideration. Find out the campaigns, adgroups, ads, keywords that convert at the best ROI - expand on these and raise bids. Lower bids for those that are you worst performers and test new ads.

Maybe the most ioverlooked thing for PPC managers is to test and upgrade landing pages. The majority of sites (as they are) are just not going to work in search because chances are someone at the top of the listings is doing a better job to convert ppc traffic. Take a look at the top listed sites for...
- price
- product selection
- estimate how much $$ they make per conversion
- usability
- conversion path: ease of and amount of steps to get to conversion
- are they monetizing their traffic in other ways (related products/services, selection, provide a reason to return, partnerships, affiliate links, adsense, etc.)

And test everything!

[Webmaster T]

All the above is fine and well but if you're selling $10 bills for $12 no amount of landing page or any other SEM wizardry is likely to end in a sale! Before I even do keyword research I go to Froogle er... Gbase and see what the products are selling for. One client raised his conversions and ROI significantly by not advertising low margin or non competitive products... especially when advertising on the clipjoints er... comparison shopping engines! It also helped him set prices that won on price point. What you must remember is many people are shopping online because the price points are lower if it's not cheaper than it is at the mall then they are going to go to the mall and buy it!

[ACpeter]

Just my 2 cents worth... but when you get into site design, it's been proven to me that you have to make your website more like a billboard than an article. Quit trying to make the visitor think. If you've provided the details they need, keep the site design clean and continue to move towards the sale. Some sites add a 3 ring circus to the web page that just distracts from what you want the visitor to do...and that is, click the 'Buy Now' button.