Scary exit rates! Options

[charliew]

So my e-commerce site has scarily high product detail page bounce rates but yet it is pretty successful. If i look at the exit rates I would think the site was an absolute disaster although in small terms it doesnt seem to be.

Having discussed my concerns with a few friends, they say they have the same, and I wonder, is this the norm, do e-commerce sites have high product detail exit rates? I thought my solution was to make it more user friendly, but perhaps the customers are just not finding what they want?

[cre8pc]

What can they do from that page?

A few things come to my mind, without seeing your site (or product pages).

1 - People are doing price comparisons. This means they're researching, not buying.

2 - Is there a call to action path/button/link to make a purchase from the product page (or is it just content?). Ex. Add to Cart

3 - Are there other suggested paths on the page, in addition to the buy path, such as Learn more, which goes to more detail or a price comparison chart and from THAT page, put in a call to action prompt to buy.

4. Is there a sale or special or attractive offer on the page?

5. How do you display the product itself? Is there a demo? Can images be enlarged? Rotated? In other words, is the product selling itself visually?

6. Is the page itself well marked and well labeled, so that your visitor knows where to go next, where they are and how they got there in the first place?

Like I said, I can't tell without seeing it so I'm making a lot of assumptions here Hopefully there's something in that list that fits your situation.

[Michael_Martinez]

How do you get your traffic? I send thousands of visitors to one merchant every month. Only a few of them buy, and I know they understand they are looking at and clicking on affiliate links. In order to draw them in, I do tell people to simply browse the pictures. Perhaps some of your visitors are just browsing and not really even shopping?

[charliew]

Yes, I suppose they are just having a look, which is not a bad thing.

Our product detail pages certainly offer nearly everything you suggest cre8pc so maybe there is no need for me to panic!

[AbleReach]

Just curious - are there rules of thumb for a target percentage of lookie loos versus conversions on different kinds of pages? or for clicking through to another page?


Elizabeth

[Michael_Martinez]

Do you mean target ratios people should compare their performance to? I doubt there are any very general ones. Each industry probably has its own.

Generally speaking, if I get 1% conversion (from click-through to sale) on my affiliate links, I would say that is a bit higher than normal. I have begun experimenting with higher priced, higher commission items, but this time of year it's hard to sell anything from my site. My commissions should start picking up again around August.

[bragadocchio]

...are there rules of thumb for a target percentage of lookie loos versus conversions on different kinds of pages?

When someone walks into a grocery store, chances are good that most of the times they will walk out after having purchased something. The same is true with a convenience store.

Conversion rates for online stores tend to be lower, because of a number of factors:

[list]It's much easier to go somewhere else.

[*]You can bookmark the page, and look elsewhere to make comparisons.

[*]You can gather information about a product and buy offline.

[*]You need more information before you purchase.

[*]The site doesn't show a privacy policy, shipping policy, or feel secure.

[*]The design of the site puts you off.

[*]You have a hard time comparing similar products on the site.

[*]The ordering process is confusing.

[*]The site just isn't persuasive or engaging.

[*]The pictures displayed aren't showing the product off well, or from enough angles.

[*]The right information isn't on the right part of the page.

[*]It's difficult to find the products you want because the site search doesn't work well.

[*]It's difficult to find what you want because the site navigation doesn't instill confidence in you that you will find exactly what you want.

[*]You may prefer to call and place an order.

[*]You could be making comparisons between different options for your money, considering different possible ways to spend it, and are just window shopping.

[*]Many other reasons.[list]For some reason, people with online stores tend to satisfy themselves with fairly low conversions rates. But, it can take a lot of time and effort to optimize each page for conversions.

A decent web analytics tool can help, as can usability testing. You can look at whether the main objective of the site is fulfilled - whether it's to have someone purchase a product or call about a service, or even just find the right information.

But, you need to probe deeper to really make a difference.[list]What path are people following through the site?

[*]Are they finding information aimed at helping them make a purchasing decision?

[*]Are they moving along to the next step, once they find that information, where they are faced with the ability to view the item?

[*]Can they find out more specifics if they want?

[*]Is it easy to compare similar products?

[*]Is it difficult to keep track of where they have been, and what they have looked at? (see amazon.com's - "these are items you have recently viewed")

[*]Are links to privacy policies, security policies, corporate information, contact information, shipping information, frequently asked questions, return policies, etc., in both easy to find and appropriate places?

[*]Is the registration process easy, and does it ask for the appropriate information and the right time and place?

[*]How difficult is it to see what is in your shopping cart?

[*]How hard is it to choose between sizes, colors, and other variations of a product?

[*]Is the purchasing process confidence building, or does it create confusion and uncertainty? If so, where?

[*]Are there other ways to get people to purchase, such as wish lists, gift registries, and if there are, how well does that process flow for people to set those up, and for people to use them?

[*]Are there ways for people to interact with the site in other ways, such as personalized reviews - the chance to write them, read them, and rate them, and find out more about the reviewers? How difficult or easy are those to use?[list]There are many things that can be done to improve the conversion rate on a site, but most of them involve a fair amount of thought, and effort, measuring movements carefully, and paying attention to outside factors, too. You have to look at each tiny step in the process.

But being satisfied with one or two percent is a failure to take advantage of the fact that people are finding your site, and actually looking at the pages, often with a credit card in hand.

[Michael_Martinez]

But being satisfied with one or two percent is a failure to take advantage of the fact that people are finding your site, and actually looking at the pages, often with a credit card in hand.

I don't know if the merchants are satisfied with a 1% conversion rate or not. But people don't come to my sites (generally) looking to buy anything. Most of my affiliate page marketing is done in-network to my visitors. They may or may not have a credit card ready to use.

There are two schools of thought on affiliate marketing. One holds the view that the affiliate should not be doing the selling, just recommending. The other school holds the view that the affiliate should be pre-qualifying the sale.

I find I earn better commissions when I just integrate the affiliate links into my content and do nothing. Trying to presell stuff just doesn't work for me.

[bragadocchio]

Those are good points, Michael.

charliew was asking about ecommerce sites, and people leaving on product detail pages, so his question possibly had more to do with a non-affiliate based site.

But, I do think that it's important to bring up the differences in different modes of ecommerce like you have. The distinction between the different types of approaches to affiliate marketing is interesting.

[Respree]

Really excellent post, Bill. If I didn't know any better, I'd say you were an E-tailer. =)

Haven't looked at the stats recently, but the last time I did, conversion rates for E-tailers, on the average, were about 2%. Yes, that's right. For every 100 visitors you get, 98 will leave without purchasing anything.

So the question is, are you 'above' or 'below' the norm? If you're drastically below the norm, this is suggestive that there are a lot of things you could do, in in terms of your site, to improve these rates. Wherever you are, you should never give up on the idea that converison rates can be improved. Bill hits the nail on the head by mentioning specific things you can do to improve conversion rates. Unless you've got something that you have designed and exclusive market, competition is, and will continue to be, intensely competitive.

Ask yourself the question, why should your visitor buy from you, versus your competitor. That is where a Unique Selling Proposition (aka "USP") comes into play. Think of ways to differentiate yourself from the rest of the pack. Look at what the competition is doing and go one step further. Some of the ways you can do this are:

- Better pricing
- Liberal returns policy
- Better customer service

Get 10 of your good friends to use the site. Ask them what they liked and disliked. What did they find frustrating? Was it 'easy' to find information they were looking for? Was it even on the site? Feedback can be a powerful tool, but you need to ask for it. In my experience, few people will just 'volunteer' suggestions for improvement.

[AbleReach]

When I was doing face to face shows one rule of thumb was that generally ten out of 100 would stop at a booth. Discouragement about a slow day at a show with low traffic is pointless. Less than 10/100 and the booth itself probably looked bad/boring/unstocked or the people in it seemed unavailable or unsavory - both in my control.

Would the 10/100 generality be parallel to getting clicked on when coming up at the top of organic search results?

What happened next was up to what was available and who's doing what. I treated each person as an opportunity to visit, help, learn from, serve, educate, with an open door to sales - all in the same family IMO. All could potentially lead to sales or referrals then or later, from the impression I left at the time or from what I learned by interacting with whoever stopped.

If I were in that world now, I'd have worked a combo of face-to-face with taste-test handouts/brochure featuring prominently displayed URL. I have 101 ideas for what I would do with the brochure. The presentation must satisfy while creating curiousity about more at every turn.

Elizabeth

[bragadocchio]

Thanks, Garrick

I've heard that 2% is pretty typical, too. But higher numbers can be obtainable. A unique selling proposition is a good start.

We do try to ask people about the web site when they call on the phone. It's funny, but nice, when you get answers like "I called because I like your privacy policy."

Those are great suggestions about a show booth, Elizabeth. I do think that the experience can translate over. Maybe not step for step, but if you think about some of the different ways that you can interact with different customers on your web site, and try some of them out, you may find some positive results.

[bwelford]

I think another aspect of the show booth analogy, Elizabeth, is that people come to the show for different purposes. Simplifying you could split them into three groups:
A - People who will definitely not purchase but are looking around for information, pleasure, curiosity, or whatever.
B - People who are not looking to purchase but might suddenly decide to buy given something that triggers them to do so (impulse purchasers), and
C - People who know they want to purchase and are looking for the right supplier.

Before thinking about conversion rates, you may want to consider how your website (show booth) will bring in A's, B's or C's. Does it look like an information booth or a retail outlet? Depending on how your website (sales booth) performs, you will get differing numbers of A's, B's and C's. If you get all A's and no C's, you have a problem.

You then have different objectives for each type. For the A's you're trying to develop a positive impression so that they may buy down the road or may tell a friend about you so that the friend buys. What you can do with the B's depends on your product. It may or may not be easy to create that impulse purchase.

Where it really counts is with the C's. What proportion of them convert into purchasers? This is where Bill's list is critical.

[AbleReach]

C - People who know they want to purchase and are looking for the right supplier.

People who know they want to purchase can be broken down into:

1 - People looking for something for an occasion, not a specific product - a birthday present for Mom (great opportunity to ask what Mom likes and get ideas)

2 - People looking for a general category - value (meaning/moral,) price or product oriented "Is this organic cotton?" "Can I get a getter deal elsewhere?" "I like candles." (notice preferences that the product niche seems to attract) Utility comes in here, too.

3 - People looking for specific items - I want a Brooks Brothers pinpoint oxford shirt with a button down collar in size thus and such (can't beat anyone at their own game, but ya sure as sugar can beat em at yours - compare, contrast, innovate)

Just a few thoughts.

To get back to the origin of this thread, what are some ways to determine if exit rates are driven by site design or site content?


Elizabeth

[bragadocchio]

To get back to the origin of this thread, what are some ways to determine if exit rates are driven by site design or site content?

Probably a combination of both design and content have an impact, Elizabeth. Design has at least as large an impact upon how people perceive a site, and the owners of that site as the words used on the page.

For example, credibility is an important part of whether or not people will make a purchase on a site. The large studies linked to at the bottom of this page from the Stanford Persuasive Technology laboratories, in conjection with the Consumer Web Watch folks, show that design has a very large impact on whether or not people believe that a web site is credible:

http://credibility.stanford.edu/

The leader of the Stanford group has a book out called Persuasive Technology: Using Computers to Change What We Think and Do. There's a great example from there that I believe illustrates this really well.

Scenerio 1: A man wearing a suit knocks at your door. His face is familiar and he says, \"You've won our Sweepstakes!\" He hands you a big check, and the TV cameras are rolling. Outside your house, three reporters compete for your attention.  

Scenerio 2: You receive a letter in the mail, sent using a bulk mail stamp. The letter inside says,\"You've won our sweepstakes!\" The letter has your name spelled incorrectly, and you notice the signature at the bottom is not an original.

The words in both are the same. The methods of delivery are every different. One enhances trust. The other detracts from it. In many respects, that's the role of design.

[cre8pc]

Elizabeth wrote:

To get back to the origin of this thread, what are some ways to determine if exit rates are driven by site design or site content?

From a usability perspective, content is part of the design. I find in my site testing work that sites with content still have high abandonment rates because the content isn't answering visitor questions or aiding the visitor in doing something productive (ex. providing direction on where to go next, to do such and such.)

Task analysis uncovers holes. Send someoneon a mission and watch to see what they do and how they do it. Don't guide them in any way.

Understanding how people use web sites is key to designing them well. Understanding how *you* (owners, developers, stakeholders) would use your site limits options and considerations for everyone else.

[AbleReach]

So, Kim, the best way to analyze exit rates is:
1. Identify the site's specific goals
2. Identify the page's specific goals
3. Send live bodies on a mission to see how/if they achieve those goals
?

Do you run into sites that don't have a clear goal for a particular page?
Is site owner motivation difficult to inspire after a bad useability review?

Where do you hit the most resistance from site owners -- possibly owners who know what *they* meant but may not see that others don't get it, *until* they see the live body results?

How do you find your testers? The more practice they get, the better they'd get at second-guessing site design, right? To get accurate average user results you'd probably need a fresh pool once in a while.


Thanks (*cool topic)

Elizabeth

[cre8pc]

Elizabeth asked:

So, Kim, the best way to analyze exit rates is:
1. Identify the site's specific goals
2. Identify the page's specific goals
3. Send live bodies on a mission to see how/if they achieve those goals

There's more, but what you note offers a way to measure and compare requirements to what's really happening. Tracking software is useful for seeing page flow (in/out, length of stay) and understanding what traffic is a robot and what is human.

Do you run into sites that don't have a clear goal for a particular page?

Constantly.

Is site owner motivation difficult to inspire after a bad useability review?

Not the way I do them (site reviews). The common remark is "I was discouraged at first, and then I realized how much information you supplied us with and we're making all the changes now."

There is no "bad review". There is only education and solutions to correct boo boo's or make upgrades and enhancements. The other key is telling them why a specific solution will benefit them. This always gets owners jazzed.

Where do you hit the most resistance from site owners -- possibly owners who know what *they* meant but may not see that others don't get it, *until* they see the live body results?

I can't remember any resistance. They always have a choice on what they do after they receive an evaluation (usability). When they see me apply a task and user persona, they see first-hand why and where something doesn't work. It's their choice on whether to make repairs.

Most site owners/design teams understand how to use their site because they built it. This makes them completely blind to their end users and how *they* might use a site. People always do things you would never expect them to do on your web site.

Should be mentioned that no site is perfect. Having worked on sites like Iomega, Gurunet and Geico, I can vouch that even those with staffs and budgets don't get it right the first time. Understanding end users is often the last thing anyone considers, unfortunately (though I see this is improving!). Even if they have data, from marketing, etc., this doesn't convert to automatic understanding of how someone will make a purchase with their shopping cart, for example.

How do you find your testers?

Because I'm a one-person act, I refer clients to companies that do user testing in labs, or virtually, like Site-Report.com. Some companies will fly in a team and do this in-house, on site.

I've led user testing in a virtual setting for AT&T Worldnet. With some of the new software out for remote testing, this will get easier (and less expensive than hiring usability labs.)

The more practice they get, the better they'd get at second-guessing site design, right? To get accurate average user results you'd probably need a fresh pool once in a while.

Depends on what tasks are being tested. First time usage is one angle. Repeat visitors have their needs too. As do special needs, as do seeing impaired.

Actually, user testing is quite a skill. The environment has to be "just so", the tasks or questions constructed "just so". Data can be influenced by the tiny things.

One of the suggestions I have for everyone is to put a feedback form/survey on your site. It's free feedback, from any possible type of end user. Ask specific questions ("Did you find what you were looking for?", "Please rate the font size"...) and add a field for comments.

The information gathered from these forms answers some of the questions site owners have when they review their traffic stats. Especially useful for understanding exit rates and page abandonment.

[AbleReach]

Thanks for the thorough and matter-of-fact post, Kim.

There is no \"bad review\". There is only education and solutions to correct boo boo's or make upgrades and enhancements. The other key is telling them why a specific solution will benefit them. This always gets owners jazzed.

Where do you hit the most resistance from site owners -- possibly owners who know what *they* meant but may not see that others don't get it, *until* they see the live body results?

I can't remember any resistance. They always have a choice on what they do after they receive an evaluation (usability). When they see me apply a task and user persona, they see first-hand why and where something doesn't work. It's their choice on whether to make repairs.

Ahhhh, now you're talking straight to my heart. This is how I would like to approach a need for repairs, or really any need. Sometimes I've forgotten to put myself in the site owner's shoes. I've heard "this is almost too much to work with," at first, then been sad when the prospective client vanished. Over the last few weeks a couple of them have dropped me a tentative line again, and they're different - as if they've incubated. I think I was probably more in tune with those two - and the neccesity for self employed perfectionists :oops: to have reachable-feeling goals.

Every few days that I am in this forum another lightbulb comes on.
Wow.

Thanks again,

Elizabeth