What is a “Good” Conversion Rate?

Posted on June 12, 2012 by Shannon

eCommerce folks often ask the same question about conversion rates – what’s a good rate to have? The answer unfortunately is elusive, as it varies from merchant to merchant and industry to industry segment.

2.9%
A 2007 Article in Target Marketing cites 2.9% as the average conversion rate. Most numbers that you see fall somewhere around the 2-3% range, which hasn’t changed much for a decade. The answer however isn’t that simple.

Context matters
You can’t really look at your site and say “I’m at 2.0%, I must be doing poorly.” I’ve seen viable businesses that sell high end goods (greater than $1000 AOV) doing just fine at 1/10th of one percent. On the flip side, some businesses run north of 10%, all the way up to 40% in some cases. Those high numbers are often for a reason – for example the 40% data point is Schwan’s which has a lot of repeat customers who are there to just re-order goods that they already know about. So again, context really matters.

Filter on Average Order Value (AOV)
One interesting way to look at the data is to plot conversion rates by a site’s Average Order Value (AOV). At some level it’s pretty intuitive – it’s easier to get someone to buy when the purchase price is $1, than when it’s $1000.

The chart shown here has some actual website conversion rates plotted against their AOV. We see a cluster of sites in the $100 or less range who are seeing rates of less than 2%. These sites are often struggling to make ends meet. In the green band we start seeing a bunch of sites with higher conversion rates and/or higher AOV. These companies are generally doing brisk business. In the blue are some outliers who are really excelling by having reasonably high AOV combined with reasonably high conversion success.

Again – we’ll mention that even here, it all depends. If it’s costing you a fortune to get that inbound traffic, it doesn’t matter if you are getting double digit conversion rates. However, using the AOV might get you a bit closer to understanding if the conversion rate on your website is under performing or is providing an ROI.

Increasing Social Engagement at the Time of Purchase

Posted on May 31, 2012 by Peter

Here at Spring Metrics, we’ve been helping our customers increase their social engagement and we’ve been seeing some noteworthy results. After gathering data for a few months now, we’re able to share some early and surprising figures.

While the public debate about the impact of social media on your online business continues (our data shows they are clearly positively correlated, and the only question should be how much a Like or Follow is worth to your business), many e-commerce merchants and retailers are quietly getting on with the business of adding social media to their toolkit.

Looking at the question from a different perspective, we have seen that catalyzing your following at the time of purchase by enticing buyers to become advocates has increased social engagement, as measured by the total number of Likes, by 23% in an average of six weeks. That’s a stunning result, and we humbly believe only partly due to our product and our awesome support team; the rest is due to simply leveraging the opportunity at the right time, when the visitor is at the point of purchase and feeling good about their decision.

Also captivating is that in early results we’re seeing that more than half of the visitors will Like or Follow, but then not purchase… yet. It’s too early to show stats on how many of those social advocacy actions turn in to purchases at a later date, but we can see that it’s not zero. And looking at the infographic we released today you can see that 54% feel more likely to purchase after Liking. So rather than simply losing the momentum of that visitor after the non-converting visit, you have preserved on average 54% of the time and money you spent attracting that visitor. And of course there’s the indirect effect, again too early to quote stats but it’s not zero either; customers are seeing a clear increase in sales from social media channels.

It’s about Link Building, Not Link Buying

Posted on May 24, 2012 by Tim_Huntley

We haven’t spent a lot of time on Search Engine Optimization (SEO) here at the Spring Metrics blog; however SEO can be an incredibly important component of your inbound marketing program – that is if you don’t take shortcuts.

Some companies without the in house expertise may need to engage with an SEO firm to help improve their ranking for a few relevant keywords. Is this you?

If you need to bring in outside talent, PLEASE pay attention to the terms and parameters of the engagement. Why? If the SEO chooses to take the easy way out by using “black hat” methods of link building, you may reap some additional traffic in the short run, but you risk incurring substantial damage to your reputation.

For example, consider the case of Dun & Bradstreet Credibility Corporation (as reported by Josh Davis). According to the story, DBCC engaged an SEO firm that employed a “link buying” strategy instead of working hard to find legitimate ways to gets links. This is a HUGE no-no with Google and a quick way to get slapped with a penalty (no more ranking for your top search keywords). But the Google penalty is only the obvious part of the problem.

Perhaps the bigger problem is the damage to your reputation. Danny Sullivan at Search Engine Land has written a great post on the topic, including how both JC Penny and Overstock shot themselves in the foot last year by engaging in this shady practice. And pleading ignorance of what the SEO firm was doing on your behalf isn’t going to do anything to repair the damage.

Be smart and make sure you know what is happening with the people who represent your business. You have worked too hard to earn a great reputation with your customers.

If you want to build links the right way, a great starting point is with the resources available at SEOmoz – Please check out their recent post SEO Isn’t Magic, So Stop Doing SEO Tricks.

photo credit: vaXzine

What is Real-Time?

Posted on May 22, 2012 by Peter

The Wikipedia entry for Real-time talks about several different categories of real-time: computing, media, communications… (it also mentions Bill Maher but we can skip that for now). The computing entry describes “real-time response times are understood to be in the order of milliseconds and sometimes microseconds,” which is pretty much what I used to think of when I was a software engineer, and in my mind is the origins of the phrase.

what is real-time In contrast, when I ran sales, out here in the human world, one of my guys used to say “we’re going to do this one real-time, partner” and that meant either this minute or this hour or this day depending on what exactly it was that we would be tackling, but in any case not much else was going to get done until the opportunity was nailed. (In most cases it meant “saddle up, I’m putting you on the phone in minutes and you better be loaded for bear!”).

Now in SaaS, especially in our business of serving up content to website visitors in real-time, we’re exactly at the interactive epicenter between the human and computer. If our system or the other systems it interacts with miss a beat, a customer may miss a conversion. And sometimes that’s microseconds or milliseconds.

Today in the office we talked about this. Our engineers have done a ton of work to tighten all the screws down and eliminate the slippage wherever possible. But still there’s the “fast bounce” visitor (who truthfully was very unlikely to be enticed to convert anyway, regardless of how fast or how targeted the content is); and there’s the inevitable delays between our Amazon servers and wherever the site is being hosted and the CRM or data sources we’re accessing to present the visitor with relevant content. But the more important aspects are the human factors aspects of real-time. More like my sales friend. We’ve spent time building in the right behavior to induce the desired results. Having something pop up instantaneously, it turns out, doesn’t have the best effect. A very slight delay creates a better contrast effect.

This reminds me of professional cycling. Some time back, the game was all about making the components lighter and lighter. Several years ago, we crossed the technology threshold and now the cycling regulations specify a minimum weight for the bikes. The technology outpaced the needs of the sport. And so it is with real-time: it’s about the people, your customers.

The Power of Audience Segmentation

Posted on May 15, 2012 by Peter

Its a story as old as humankind: specialization. As a species, we’ve been becoming increasingly specialized for about 10,000 years since the advent of agriculture, which allowed us to remain in one place for longer.

A decade or so in to the e-commerce evolution, we’re now seeing an increased focus on delivering specialized content to your audience segments. This great post today from HubSpot focuses on the value of segmentation in email marketing (which Durham startup Windsor Circle is leading). Clearly, targeting contacts for whom you have preference information with content suited to them will have higher conversion rates, and less opt-out responses or other brand-diminishing results. And everybody knows “email converts,” which makes sense when you consider that by the time you have somebody’s email address you already know a thing or two about them.

Segmentation also applies to PPC campaigns – when you set up the campaign, you’re segmenting the target audiences and channeling them to segment-specific messaging on a relevant landing page.

In the brave new world of content, as pointed out in the HubSpot post, segmenting the audiences is only valuable if you have something differentiated to say to each segment. The great side effect of this forced behavior is that it requires you to think about who your customers are and what their need is.

Developing on that, and taking the concept further up the conversion funnel, the most effective way to widen your sales funnel is to take your Audience Segmentation existing site traffic and say something meaningful to them from the very first moment. But how do you say something meaningful to somebody you’ve never “met?” Well, you observe their behavior and find out what you can about them before you speak. “Those are nice boots. Are you from Texas?”

As an online merchant, such hail-Mary potshots aren’t necessary because by the time somebody is on your site, you already know a few audience segmentation facts about them such as where they came from, where they are geographically located, whether they’re on a mobile device, whether they have visited before… And with those simple facts, you can start a conversation that will lead to a conversion.

The Picture Perfect Ideal Customer

Posted on May 9, 2012 by John Jantsch

When it comes to attracting your ideal customer you should be able create a picture in your mind as you describe them. Using images of a real life customers can prove an effective way to help everyone in your organization narrowly focus on and communicate in ways that more directly appeal to your specific ideal customer.

A detailed profile, one that includes photos and stories of real customers, should be part of your marketing action plan documents. You may never share this type of document publicly, but it can be one of the most important internal training documents you ever create.

In order to create your profile you need to understand as much about your ideal customer as possible. Remember the key phrase here is ideal. I suggest looking long and hard at the characteristics of your most profitable customers that also refer business to you, that’s the model of an ideal customer.

Once you dig deep and profile the common characteristics you should also start asking yourself some questions about these folks.

What brings them joy?
What are they worried about?
What challenges do they face?
What do they hope to gain from us?
What goals are they striving to attain?
What experience thrills them?
Where do they get their information?
Who do they trust most?

The answers to the types of questions above are not always available, but pondering them in relationship to your ideal customer may allow you to more fully address their wants and needs in every interaction and communication.

Complete the profile, add a real photo, and hang it up in your office for all to see. Simply hanging photos of your customers around the office, may be the reminder that everyone in the office needs to connect with what your business is really about, what their work is really about, and who really pays everyone’s salary.

Courtesy of John Jantsch, Duct Tape Marketing. Republished with permission.

Social Media Campaigns: Valuing Tweets and Likes

Posted on May 2, 2012 by Peter

Social media campaigns are one of the hottest ways for companies to promote their products and services to potential customers. It is becoming the hallmark of many branding efforts. Everyone agrees that social media marketing is important.

But, what nobody is sure about is exactly how important it is, and what parts of a social media campaign deliver the most value. Data is beginning to emerge, but there has yet to be any consensus as to the real value of a tweet, a Facebook “like” or a Facebook “share.”

The problem is the methodologies being used to determine the value of a social media connection. Companies are looking at the value for their own business – but each business is different. For example, if your company is selling toys that cost $2.99 each, the value of a like or retweet that results in a purchase will be lower than if you are selling $75,000 sports cars.

There has yet to be any data that is valuable across various industries. And, it would seem that the ROI in a given social media campaign is still something that has yet to be quantified over the retail market as a whole.

For example – imagine a gas station that simply puts up a sign telling people to follow them on Twitter or Facebook. This costs them nearly nothing. If they sell two tanks of gas from that effort, they have made a profit. Now, imagine that a car company spends $10,000 on Facebook ads to direct people to like their Facebook page. Which company is going to get the better ROI? This has yet to be studied.

In addition, the methodologies used to generate data are focused on the primary marketing goal – making an immediate sale. But every retailer really has three goals from a social media campaign. The top priority is making a sale, the secondary priority is getting contact information, and the third priority is to at least get a like or a follower so that you can use future marketing efforts to accomplish one of the first two priorities. Until data has been analyzed showing how social media affects all three of these goals, then the data is not giving companies a complete picture.

At Spring Metrics, we are gathering our own data to measure the effectiveness of tweets, likes and Facebook shares over the next couple of months. Our goal will be to demonstrate exactly what kind of return on investment companies can expect from different social media actions – and how to measure the effectiveness of different campaigns.

We are doing this because we believe that social media is an incredibly valuable marketing tool. After all, since the beginning of time nearly all commerce has been social in one way or another. Long before the existence of social media or even the Internet, people have been much more likely to purchase products that were popular or liked by their friends.

Social media therefore simply streamlines the ever-present social aspect of commerce – meaning that its effects will not be a fad. It is here to stay, and must be understood.

When our results are completed, we will share them on this blog. In the meantime, we thought it would be helpful to talk about some of the data that is out there so that companies have a baseline to work with when thinking about social media campaigns.

The ticket sales company Eventbrite did their own internal study on the value they got from people sharing links to their events on both Facebook and Twitter. What they found was that each Facebook “share” resulted in $2.52 in ticket sales. That was a far better result than the 43 cents worth of ticket sales that were generated for each retweet on Twitter. However, that gap is closing, very interesting data here.

There are a few thoughts out there as to why this may be the case. People who share an event on Facebook are more likely to be actually going to the event which makes it more likely that their friends will want to but tickets to go with them. People on Twitter, on the other hand, are less personally involved with the people they “follow” and less likely to know the person in the real world. This means that a retweet is less likely to spur a ticket sale.

This study is, of course, very specific to the event industry. It is highly probable that companies in other industries would see different overall results.

A similar study of their own metrics was recently released by ChompOn – an online coupon company that competes with Groupon. Their study also showed that Facebook shares seem to have more value than retweets but at a much larger scale. There are infographics from the study, but a more complete look is here.

ChompOn said their data showed that a Facebook share had a value of $14 while a retweet only had a value of $5. Their metrics went further than Eventbrite. They determined that the value of a Facebook “like” is roughly $8 and a Twitter “follow” is roughly $2. I wonder if that gap is closing since the study as well?

Keep your eye on this space while we continue to examine this emerging science. We are dedicated to helping our customers get the most out of their social media efforts, and will have the data companies need to make the best decisions about where to concentrate their campaigns.

Improve Page Load Times and Improve Conversion Rates

Posted on April 26, 2012 by Tim_Huntley

I hope everyone is having a great week – me, I spent the past couple of days working on my website to dramatically improve page load times.

Why? Because when people come to a website and it takes more than a few seconds to load, they leave. And just like Peter said in his post last week, this is a huge waste of your marketing dollars. Don’t make the supreme effort to get someone to your site only to have them leave because it was unnecessarily slow.

Take a peek at this chart from KISSmetrics (click here for the entire infographic):

As you can see, slow load times lead to cart abandonment and low conversion rates.

What I Did to Improve Page Load Times

My website at My Athletic Life is set up using WordPress ; however if you are using a different environment, the concepts will still apply.

A few weeks ago I decided that it was time to deal with a high bounce rate on my site. I thought the problem was a result of a cheap hosting solution, so I decided to shell out a few more dollars and improve the database portion of my site. Performance was better, but not to the degree I expected.

To help with my analysis, I found a free tool called WebPageTest which loads a specific page and maps out where time is being spent. As an example, this page had been loading in a tortouise-like pace of about 8 seconds. However, by making a few tweaks and without spending additional money to upgrade my hosting, the average load time is now about 3.5 seconds. And this is for the first time the page loads. Subsequent loading of the page by the same user now takes under 2 seconds!

Here are the changes I made:

1) I installed a WordPress plugin called W3 Total Cache – by turning on the Page and Browser caching features, I made it much more likely that subsequent visits by the same user would be faster.

2) The W3 plugin also helped to speed things up by compressing much of the information that is sent between webserver and browser. I did have to make an easy configuration change to my webserver (enabling mod_deflate in apache), but this single change resulted in about 50% of the performance improvement.

3) Next I set up another plugin called Better WP Minify – This plugin was able to take a bunch of individual javascript and css files that needed to be loaded for a page and turned them into a single file. So rather than making 20 requests back and forth between browser and server, it now takes a single request.

4) I also installed a plugin called WP Smush.it to compress all of the image files on my site. I would estimate that this reduced the average file size for each image by about 45%. Huge savings.

5) Lastly, I de-cluttered my sidebar. Over the past few months, I had added a few images and features to the sidebar that didn’t add a lot of value and were slow to load. Goodbye.

I may look at some additional improvements like adding a Content Delivery Network (CDN); however it looks like I have already gotten a huge benefit from just a few hours of effort.

Let me know if you have other ideas about low cost ways to improve website performance.

5 Ways to Make the Most of Real-time Data

Posted on April 24, 2012 by John Jantsch

Every businessperson knows that good data makes you smarter. Well, a truer statement might be that an accurate analysis of the right data can make a company smarter, but lack of any data can be fatal.

Today’s reliance on the web and social media requires real time tracking and analytics that are delivered in real time. This isn’t just a way to feed the obsession over traffic and other sometimes meaningless stats, this is a valuable way to view and react to business opportunities as swiftly as they arise.

Speed may indeed be the most important competitive business advantage.

With real time data tools you can:

React to mentions of you brand more fully – you’ll know the minute your business has been mentioned by a blogger or other online publication and the implication that mention may have in terms of traffic and potential reciprocal reactions. You may choose to go comment on every post that mentions your brand, but you may really make it a priority to comment on a post that is sending buckets of traffic your way.

Interact with site visitors live – you’ll get a view of who is visiting your site in a live mode and witness where they came from, what paths they take and where they exit. Some real time stats packages actually allow you to launch a browser chat session with a specific visitor that, say, clicks on several links like they are looking for something.

Watch your test in real time – You can track A/B split tests and see immediately not only what page is getting conversions, but what visitors are doing and not doing on each test page. I think this might allow you to make your testing adjustments much faster and more accurately.

Test your PPC campaigns in real time – Google AdWords data is often hours old and, in my experience, not terribly accurate. By monitoring a new ad campaign you real time you can know immediately which ad version or which landing page version is a winner in a matter of moments.

Understand the impact of your blog posts – I’m still puzzled sometimes by the reaction to my blog posts. I find it very hard to predict when something I’ve written is going to catch fire. With real time stats I know within minutes of publishing if I’ve happened to spark a reaction with readers and that gives me the cue to get proactive about promoting the post in bookmarking sites and social networks much more aggressively. I don’t like to promote everything I publish in this manner, so this gives me the filter to know when to amplify big right away.

There are by my guess dozens of packages available that can help you get web site data in real time. Some are free and some come with monthly fees.

For some, the idea of a monthly fee for the benefit of stats in real time may feel excessive, but in the increasingly competitive web world, the businesses that react the fastest win.

Spring Metrics Blog

Website Analytics, Web Marketing, and Conversion Optimization For SMBs