Slow and Steady Wins the Online Marketing Race

Back in June I wrote a piece called, "Pssst, What Is Your Secret to Online Marketing?"  In that post I laid out the case that analytics were the most critical component to an online marketing strategy.  However, months later I continue to get people who ask me for the silver bullet to online marketing.  I guess that post wasn't too convincing so this month I am going to lay out another overarching principle in good online marketing - being slow and steady.

Am I really suggesting that you should move slowly in online marketing?  No, move slowly in the Internet business and you will get left behind. However, success in online marketing often comes from paying attention to a million and one details, and you cannot pay attention to details if you are moving too quickly.  Even those companies that really break through and create a nice buzz often do so by doing many little things right.  Customers see the company over and over in a positive light and build a great impression of the company.

What sort of details matter in online marketing?  Quite frankly because all these pieces build on each other, the answer is pretty much everything does.  Here are examples of details that I often see companies overlook:

• SEO - are you creating good title tags, creating alt tags for your images, writing good page copy, etc.?
• Website hygiene - do you have dead links, old content and other major mistakes on your site?
• Retargeting - do you have a re-targeting program that puts customers into all the appropriate segments and hits them with the right follow-up messages?
• Triggered emails - are your targeted messages crisp, well designed and regularly revised based on data?
• Paid search - are you constantly reviewing your campaigns, allocating more budget to the good ones and tweaking or killing off the dogs?
• A/B testing - are you A/B testing, well, everything?  No, I really mean everything. 
• Cross marketing - are you actively looking for good cross-marketing partners?  Do you regularly see who already sends you a lot of traffic and then reach out to them?
• Relationships with key customers - do you look to build deeper relationships with key customers?  Do you thank them in any way for their support?

This isn't an exhaustive list by any means but it should give you a taste of all the little details that any online marketing teams needs to pay careful attention to. 

What do I mean when I say these things build on each other?  If you can double the number of people coming to the front door of your site through a lot of little things and then double the conversion rate to trial and then double the conversion from trial to paid, you are talking about an 8x improvement in revenue!  Now we are talking about some big results.

Folks, it is time to stop looking for silver bullets and focus on the blocking and tackling.  Put enough blocking and tackling together and you will see big results.

Where to Start with A/B Testing

I am astonished by how often I run into Internet entrepreneurs who have an Internet business but have yet to build A/B testing into their culture.  By using A/B tests, you can quickly improve how parts of your business operate.  You can A/B test the web site, emails, and many other facets of your business.  Each time you are measuring the effectiveness of one approach versus another. 

The benefits can be huge.  At TeamSnap we made one relatively minor change to our home page that immediately increased trial signups by over 30%.  That test required just a couple hours of a designer's time plus an A/B test tool which cost us about $300/month at the time.  If we tried to buy 30% more trials through customer acquisition methods, it would have cost us a fortune.

The first place to start with A/B testing is to agree on what you want to optimize. Yes, this sounds obvious but sometimes not everyone on the team might agree.  For a typical SaaS business, you might look at things like trial sign-up rate and conversion from trial to paid plans.  For an e-commerce business you probably want to focus on revenue per visit or a similar metric.

Next you need to put in place a testing system.  We use Optimizely because it is super easy to use and allows designers and marketing professionals to launch tests with little to no involvement by development. That allows us to iterate very quickly.

From there you need to decide where to start testing.  There are a couple approaches you might consider:

• High traffic pages are the most obvious choices.  If you focus on pages that are seen by lots of people, you get two benefits.  First, your tests will complete more quickly because you collect a lot of data in a short period of time.  Second, as a general rule, high traffic pages will impact your key metrics more than low traffic pages.
• "Broken" pages are another logical place to start.  Everyone has a few pages on their site that they know they would like to blow up and re-do.  A/B testing can help guide you through that process.  Run an A/B test with a new design that you have been itching to roll out and see if it improves performance.
• High "page value" pages are a more scientific way of approaching things.  Google Analytics computes a page value for each page on your site using a formula that links that page to transactions and goals on your site.  This is a more systematic way of identifying pages that are likely to impact your key metrics (assuming you are using Google Analytics goals and advanced e-commerce capabilities).

One final thing to consider is that you need to have some basic understanding of the math behind A/B testing before you start drawing too many conclusions from them.  Remember what a confidence interval is from your college probability class?  If not, time to brush up.  Here is a very common mistake.  Let's say that your test shows that variant A was better than variant B at a 95% confidence interval.  A on average was 10% better at whatever goal you defined.  The test is saying that you can be 95% certain that A is better than B, but you cannot be 95% certain that the improvement is actually 10%.  Brush up on your math and you'll understand why.

The most important thing with A/B testing is to build it into the culture.  Every time you approach a problem and every time you consider a change to your site, ask whether you can use A/B testing to help guide you.

Not Using RLSA with AdWords? Shame on You

A couple months back Google quietly rolled out a new feature that completely changes how advertisers think about paid search.  The new feature is called RLSA. If you are running AdWords campaigns, and you don’t have RLSA enabled, stop what you are doing and go check it out now.

In the past we couldn’t incorporate what we knew about customers into what we showed them when they were searching on Google.  In other words, we largely were serving up the same ads to everyone on paid search.  Sure, we could vary the ads by region or device or language, but those are fairly blunt targeting objects.

RLSA essentially combines re-marketing with paid search.  In other words, when someone visits your website, you can tag them and put them into customer segments.  You then show them different search results in Google depending on what they have done on your website. 

One of the easiest ways to get started with RLSA is to tune your branded campaign.  For example, you can decide that you don’t need to pay for branded ads for existing customers.  With RLSA you tag existing customers (e.g., visitors who come to your website and log in).  In AdWords you then specify that visitors with that tag don’t see your branded ads. You can greatly reduce your branded search spend with this technique.  Why burn your marketing budget on paid ads for existing users who can just as easily log in by clicking on your natural search results.

Once you have tuned your branded search, it is time to think about visitors who are in key stages of the conversion funnel.  For example, let’s say a customer has put something in their shopping basket.  You might be willing to bid higher for those folks if you see them searching again.  Alternately, you might want to show them a different ad.  With RLSA, you segment out the customers who have something in their shopping cart and then run different campaigns for them.

Those are just two simple examples of how RLSA can be used.  It truly is a rethinking of what paid search means.

Pssst, What Is Your Secret (to Online Marketing)?

I get several calls a day from people who want to know my secret formula for online marketing.  These are business acquaintances who look at how we are doing at TeamSnap and want to replicate our growth at their company.   So far this month we have added five times as many customers as last June, and last June was a great month in terms of customer acquisition.

We have a great product that people find easy to use and incredibly handy.  In addition, our customer service is excellent.  As a result, the service is extremely viral.  Having said that, we definitely have greatly increased our usage through online marketing.

When I tell these acquaintances our online marketing "secret" they blink and say, "Really?  That is it?"  My answer is so, well, pedestrian that they are always surprised.  They are expecting some super secret Google Ad Words trick or a SEO tool that no one knows about.  I have a few of those too, but I require a good bribe before giving those out.  (Hint, I love sweets.) At the same time, when I ask these acquaintances if they are following my "secret," they almost always say no.

So what is this "secret"?  It is analytics, analytics, analytics.  As simple as it sounds, you need to figure out what action you want to increase and find an easy way to measure it.  Are you trying to increase trial sign-ups?  Then measure what the sources are of trial sign-ups.  Are you trying to increase leads?  Then measure what sites, ads, emails, keywords, geographies, devices, etc. send you the most leads.  While I am partial to e-commerce tracking in Google Analytics (even for companies that are not in e-commerce), there are many ways to measure these things.

The key is to set up a regular process for reviewing the sources that drive this action that you are trying to increase.  On my marketing team we religiously review trials by source at our weekly marketing meeting.  It is the core report that determines where we focus our time.  If a source delivers a lot of trials, we put more time and effort into it.  If it doesn't deliver many new trials, we cut back or turn the program off.  It is that simple.  It is a simple but highly effective time management exercise.

Once you have this process down, it is critical that you run lots of experiments.  When I am evaluating a new marketing source (e.g., a new advertising venue or even a business development partner) I try to start with the smallest commitment that will allow me to effectively test that channel. If it is a test of a new advertising channel, I often find that spending $10,000 or even far less is plenty to prove whether that vehicle works or not. 

Don't overthink it - test, measure, tweak, and repeat.

The Big Mobile Story of 2012... That You Almost Missed

Last week there was arguably the biggest announcement of the year for the mobile market.  Yet, many people missed the announcement amidst all the clatter about Hurricane Sandy and the presidential election.  What was this announcement?  The iPad Mini?  Something about the iPhone 5?  An update to the Samsung Galaxy S3?  Nope, none of those.  It was Google's announcement of the Google Nexus 4 (and Nexus 10), building on the success they have had with the Nexus 7 tablet which is now selling a million units a month.

Huh?  What is the big news about the Nexus 4?  The Nexus devices are Google's flagship, reference phones and tablets that are meant to set the standard for all hardware vendors in the Android market.  On the surface the Nexus 4 is certainly a highly quality device.  It has a stunning 4.7" 1280 x 768 display, a smoking fast processor, a large battery and a ton of features all packed into a lovely package. It also runs Android 4.2, the latest iteration of the Android operating system.

On the surface the phone can go toe-to-toe with the iPhone 5.  Compared to the iPhone 5, the Nexus' screen resolution is higher, it has about 50% larger battery, it has twice the processor cores, twice the RAM, etc.  Android 4.2 includes location-specific information before you even ask for it and other features that are not available yet on iOS.  On the other hand, the iPhone 5 is a beautiful phone with gorgeous packaging.  While its OS has taken some hits lately (e.g., the whole map debacle and the over-promises of Siri), it still offers a great user experience overall.  Said another way, calling a winner between the Nexus 4 and the iPhone 5 is like trying to call a winner in tomorrow's presidential election - you could try but it is like splitting hairs.

However, there is one element of the Nexus 4 that really broke new ground - the price.  The Nexus 4 is priced at $299 unlocked.  Unlocked means that there is no contract required.  T-Mobile gives you roughly a $20 / month break on your bill if you bring your own unlocked phone.  So with the Nexus 4 you pay $100 more than most high end smart phones on a 2 year contract, but you save $20 / month by having an unlocked phone.  That nets out to a $380 savings over a 2 year period.  For reference, it is believed that Apple is selling the iPhone 5 to carriers at $649, more than double the price of the Nexus 4.

Price does matter!  The U.S. mobile phone market is now roughly 50% smart-phones.  The remaining 50% of the market are people who are less technically savvy and generally more price sensitive.  Being able to buy a top-of-the-line smartphone for roughly half the cost is tremendously attractive.  The same can be said outside the U.S. where much of the growth of smart-phones will likely come from countries like India and China where consumers are likely to be even more price sensitive.  The pricing of the Nexus 4 is likely to tap into broad new swaths of customers.

The price of the Nexus 4 also has the potential to impact technology savvy customers.  Most U.S. customers are on 2 year replacement cycles for their smart-phones because they have 2 year contracts with their carriers.  However, with the Nexus 4 priced at less than half the cost of an iPhone or a Samsung Galaxy S3, you can conceivably move to a one year cycle with your phone at the same cost.  Compare the iPhone 5 or any current high end Android phone to the iPhones / Android phones of a year ago and you will be stunned by the differences.  It is like comparing a Ferrari to an economy car - the displays, the computing horsepower, the network speeds, etc. all have come so far in a year.  Who wouldn't want to turn their phone over every year (environment considerations aside)?

The mobile market has been in a serious state of flux since the iPhone was first launched.  With the release of the Nexus 4, Google is pushing everyone to change again.

Are You Really Testing Your Email Campaigns?

Ask any online marketing professional whether they test and measure their email campaigns and they will quickly respond, "You bet!"  However, in reality, most online marketing professionals are doing a poor job of testing, measuring and optimizing their email campaigns.

Wired Magazine recently did a great piece on how web companies are using A/B testing tools to measure the effectiveness of their websites and make thousands of small tweaks to optimize them for various metrics.  However, many marketers forget to bring this same mentality to email marketing.  Let's face it - email marketing takes a lot of work.  You have to get emails out on a tight schedule, and it is always a battle to get the content done in time.  Segmenting your list and making sure you are sending the right content to the right list takes a lot of work.  I have a whole folder of emails that were major screw-ups from big-time Internet companies. These are cases where companies sent the wrong message to the wrong customers, often causing dire results like crashing their site or their call center.  As a result, as a profession we sometimes run out of steam to really tweak and optimize our emails.

Here is a quick test you can take to see if you are doing a good job measuring and optimizing your email campaigns:

• Do you A/B test subject lines on your newsletters to optimize open rates?  (OK, this is an easy one - most folks will answer yes.)
• Do you measure open rates by day and time?  (This is old news.)
• Do you A/B test subject lines on your transactional emails?
• Do you do A/B tests on content in both your transactional emails and newsletters to compare click through rates of different variations?  For example, do you change up images in your emails or key text sections to compare results?  (Now we are getting into the harder stuff that most people forget to do.)
• Do you segment your welcome emails to test the long term revenue impact of different variations of the welcome emails?  Do you measure the long term revenue impact of one welcome email vs. a series of welcome emails?

The last couple of questions require fairly sophisticated email and segmentation tools, but can really drive tremendous value to the business if you can unlock what works for your customers.  Don't let the day-to-day burden of sending emails bog you down to the point where you miss the strategic opportunities that come from doing more sophisticated tests and optimizations.

Automating Google Analytics Reports

If you followed my advice in my last post, you now have your web site traffic segmented perfectly in Google Analytics using custom segments.  While this gives you tremendous insight into where your customers are coming from, it can take some work to spit out meaningful reports each week.  While Google does have built-in reporting tools, you can only look at a few segments at once and frankly, the output of the reports can be a bit limited.

My preferred approach for analyzing Google Analytics data is to use the tool that so many office workers rely on every day - Excel.  Excellent Analytics offers an excellent tool for pulling Google Analytics data into Excel, and last time I checked, it was free.  What I will show you how to do is to use Excellent Analytics to build a spreadsheet that automatically pulls your weekly Google Analytics customer traffic data by segment into Excel.

The first step is to install Excellent Analytics.  I won't cover that here as the software provides instructions for that - it is a plug-in for Excel so it is pretty straightforward. Once you get it installed, you should see an extra tab in Excel like this.  Note the extra tab called "Excellent Analytics" on the right.

Next you will click on New Query and you will need to sign in with your Google Analytics account information.  You now can pick the metric you want to report on.  Let's pick "visits"...

Now pick the segment you want to report on.  Use the custom segments that we set up last week.  You will have one query for each segment...

Finally pick the start and end dates. When you are ready, hit the execute button.  This will bring the data back into Excel.  You basically will repeat this process for each segment you want to track.  You can summarize the data in a table or in a chart.  Each time you need to refresh the data, just hit "update query" for each segment, change the start and end dates if you like, and hit execute.  You will almost instantly have a powerful chart showing where your traffic is coming from that you can show to your management.