I really had no idea what to expect going into this – so now, at the end of the event, I’m publishing the full results here for anyone thinking of running a similar campaign.

1. The Giveaway

Let’s start with the Giveaway Page – as you can see, the entire Giveaway content is set up in Gleam and only embedded on this page:

The main way to enter the giveaway is of course email – that counts as 1 entry. I have set up 3 alternative ways to increase your chances of winning:

• Follow Baeldung on Twitter (@baeldung)
• +1 the Baeldung page on Google Plus (G+ page)
• Refer Friends (via Twitter, G+, etc)

So – pretty straightforward all in all – took me about an hour to set up and off it went.

2. How I Promoted the Giveaway

Promotion of the Giveaway is of course the most important thing. Here’s what I did:

• I emailed my list (~10.000 subscribers)
• I tweeted the giveaway page out (~2500 followers)
• I posted on G+ (in the Java community)
• I posted on Facebook

3. The Numbers

Here’s what the final Gleam stats show:

• 2315 impressions
• 1195 actions (~52% conversion rate)
• 750 users (~32% conversion rate)

And here’s the breakdown of the actual results that came out of the campaign:

• 239 – Google+ Followers
• 179 – Twitter Followers
• 258 – Email Subscribers

Finally – here’s the full traffic, actions and users – graphed for the entire duration of the event:

4. What You Can Do Better

First off – Gleam is a nice system for running a Giveaway; it’s not free, but it is reasonably priced at 39$.

Now that the event has finished, the one thing I would recommend if you’re aiming to go beyond the results I had here is to up the ante with the giveaway. Basically – give away something more valuable to your audience. The Spring Security Video Course is (usually) priced at 30$, so the value of a copy isn’t huge. Going with a higher ticket item – but still relevant to your audience – will increase the engagement and the virality of the campaign.

For my next event, that’s what I’m planning to do – pick something that is both highly relevant for my readers, but also just higher value overall.

5. Conclusion

Hopefully that gives you some insight into what kinds of results you can expect out of a Giveaway/Contest like this. The numbers aren’t huge, but they’re definitely not negligible either.

The post Running a Giveaway with Gleam first appeared on Meta Baeldung.

Last month I performed a SEO experiment with the internal linking structure of my site. I picked one of my newer tutorials – one that wasn’t getting any traffic from Google, and I linked to it from 12 individual articles. These are relevant articles on the same topic – basically all of my articles in that particular category.

The main tutorial page is: http://www.baeldung.com/httpclient-guide

And the individual articles are all of the 12 articles on that page.

The link I used is very simple and natural:

If you want to dig deeper and learn other cool things you can do with the HttpClient – head on over to the main HttpClient tutorial.

The anchor text of the link is exactly the same for all of the 12 links I added.

2. Traffic Results

When I started the experiment, the page was already a little more than a month old – it existed and was indexed since the 21st of March – but, until the 26th of April, it was getting 2-3 visits per week, so almost no traffic.

I added the internal links in the last week of April; the result was quite obvious – the traffic spiked on the 29th to ~100 visits / week and again more recently to ~150 visits / week:

3. Conclusion

Good internal linking is a powerful technique – it’s entirely possible to get a page ranking using only these types of links as long as you have a few relevant pages with some juice behind them.

So – overall, I’d say the experiment was a quick win and a learning experience.

The post Rank a Page using Only Internal Links – an Experiment first appeared on Meta Baeldung.

I started this second blog to chronicle my work and the experiments I am running on my main blog. And, for the last six months, I have been doing just that – I documented in detail how I grew my email list from 3 new subscribers/month to around 500. The focus was the email list – and partly, it still is (stay tuned for my next experiment on that – as the data is looking to be very promising).

Now – I want to tackle another challenge – and that’s going to be launching an infoproduct – a Video Course to be exact. I have already created one last year with Packt Publishing – and the grueling 6 month process resulted in my Spring Security Video Course.

The Spring Security Video Course

It was quite the learning experience to go through this process – from building the actual code (that was the easy part), going over each section with colleagues to make sure someone else can pick up and understand all the concepts that I was introducing, all the way to scripting and slowly recording the material. I’m not quite happy with the first few sections of that course – by the time I reached section 8 I had learned so much that I wanted to re-do sections 1 through 3 (I did manage to re-do small parts, but not completely).

Now, having done all that, I’m planning to do it again – this time for myself. So – the plan is simple – go live with one or two Video Courses in 2014.

The obvious question is – what should the course be about? It’s not that I’m in need of ideas – just the opposite – to many options look good right – at least on paper. My blog is mostly focused on Java, Spring, Spring Security, Persistence and a host of other technologies. And in the last few years, I’ve been working with Machine Learning algorithms – mainly Recommendation Systems, Classifiers, clustering logic and the like – so that’s a topic near and dear to me as well.

Hence – the plan.

The Plan

The plan is simple – instead of me choosing what this course will be about – I’ll let my readers pick the subject. Or rather, I’ll pick based on the data, and the data is email signups (I haven’t forgotten email).

What I’ll do is put up a landing page with 6 course ideas – and then an individual page for each one. People will sign up for the courses they like and ignore the rest – and the first one that gets to 250 signups – that’s the one I’ll build (first).

See? Simple.

The Execution

I’m in the process of building the individual landing pages for each course myself (I’m using leadpages), and I’m working with a designer for the main landing page. All of these will live on a subdomain of my main blog (like this site does) – products.baeldung.com (I’ll update this article with links to these as soon as I have them up and running).

The Conclusion

This is the first in what’s going to be a series of articles documenting the entire process of creating and launching the video courses. The next steps will cover:

• Creating the product pages with leadpages
• Driving traffic to the product pages from baeldung
• A/B tests on baeldung – test out different ways to link to the product pages
• Setting up Analytics and tracking how each traffic source converts
• Traffic Sources – Paid Traffic?
• Running A/B tests on the individual product pages
• Keeping the email lists warm
• Getting to 250 signups
• Probably a lot more

The next few months are going to be exciting.

The post Creating an Infoproduct – Case Study first appeared on Meta Baeldung.

All my articles have a simple Call To Action at the end – a way to convert a casual reader into a more engaged one actually following the content I put out.

A while back, that copy was:

If you liked this article, you should follow me on Twitter.

Followed by a link to my twitter profile.

This worked reasonably well and so I left it on for a long time without thinking twice about it.

When I started to pay attention to my conversions, I noticed that the message was a bit vague and general – not really aligned with the article itself in any relevant way.

If someone is reading an article about say – Java, shouldn’t the message also say something about Java? If they’re reading about Conversion Optimization – like you are now – shouldn’t the message also mention that they can further learn about conversion optimization?

With that in mind – I quickly set up my experiment.

2. How to do it in WordPress

To unlock this kind of custom message capability  – we’re going to use what WordPress itself provides – Shorcodes – via a any WordPress plugin that supports them (I’m using Shortcoder but there are several).

We will first define a new shortcode for each category of the site: [category_end] – for example, for Java articles, we will define [java_end].

Next, we will go into each of the articles in that category and manually add the shortcode at the very end of the article. My site has around 100 articles, so this was not super quick but definitely doable.

Now – with the shortcode empty – this doesn’t do anything – but now you have a place to add your copy and can actually tailor it to the topic the article is about.

So, for an article about Java – you may write:

If you want to read more when I write about Java – you can follow me on Twitter.

Or (my current copy):

I usually post about Java stuff on Google+ – you can follow me there.

3. Results – Before and After

Now – on to the results of this experiment.

I haven’t found a good way to monitor the “Follow” event from the Google+ Badge I’m using, but I am receiving emails for each new person that follows me there. So – for the 2 weeks before and the 2 weeks after I made the change, I kept track of these emails:

• 2 Weeks Before: 29
• 2 Weeks After: 43

The ~50% increase has remained relatively stable ever since – so it’s good to put numbers behind the assertion that – a focused, contextual message can significant increase on your conversions.

The post How I contextualized my “End Of Article” Calls To Action first appeared on Meta Baeldung.

It’s a crying shame that I didn’t collect emails from the readers of baeldung.com up until a few months ago – email is a powerful tool, and one that I’m starting to appreciate more and more.

Once I did start to collect them, I figured that the logical course of action was to set up an RSS to Email campaign in Mailchimp – and I was done.

I didn’t pay to much attention to the fact that, most of the emails I was receiving from various blogs were not a word for word reproduction of the article. No, instead these were custom written emails talking about the article, giving the reader some context and only linking to the full article, not reproducing it.

What’s more – these emails were part of a narrative – they had a regular schedule / cadence – and most importantly – they were linking to evergreen content, not stuff written 4 hours ago.

So – I decided to learn how to do it intelligently, and I stopped my RSS to Email campaign for good.

2. Results of RSS to Email

First – let’s look over some early results of my RSS to Email campaign running for about 3 months:

• Average Open Rate: 25 %
• Average Click Rate: 6 %

This is the raw data for exactly 30 emails campaigns.

3. The New Way = The Old Fashioned Way

A few weeks ago I decided to experiment – put aside the mental unease of writing these custom emails and jump straight in.

I’ll share the numbers here about how my last few email campaigns are converting – how my open rates and more importantly my click rate has improved:

• Average Open Rate: 40 %
• Average Click Rate: 13.5 %

This is still early on – so I will keep updating this article as data comes in, but based on these last few campaigns – this is HUGE – the jump I’m seeing in clicks alone is 125%!

4. Conclusion

Growing your email list is the first step of better engaging with your readers.

However – that is really a first step – nothing more, and without crafting and thinking through your message and how you can help your regular readers – there’s little point to having an email list in the first place.

As I learn how to do email better – I can clearly see how the big jumps in conversions – the low hanging fruit (like this one) are going to be harder and harder to come by, but that doesn’t mean they’re not there!

The post Why you really shouldn’t do RSS to Email first appeared on Meta Baeldung.

As my email list has rapidly grown to over 1000 subscribers, I am working on keeping the list clean, engaged and full of subscribers that are actually getting value from the stuff I put out. The goal of the list is to help developers – pure and simple – however, the fact of the matter is that the stuff I send out is not for everyone – nor do I intend it to be. A big part of that is segmentation – sending the content only to the subscribers who is genuinely interested in that content – and part is setting expectations – being more explicit about why someone should take the time to be on the list in the first place. With all of that in mind, I am putting into practice a simple idea – with the goal of keeping my list clean and providing people that may not be getting the value they should from it – with a WAY OFF – you guessed it:

I am asking people to unsubscribe from my list!

2. Who to ask?

Let me be clear – asking someone to unsubscribe from your email list is not an email you want to send to your engaged subscribers. These are the readers that are actually reading your content regularly, replying to your emails and generally getting some level of value from the stuff you send them. What you want to do – and what I did in both instances I sent out this email – was to identify a segment of the list that is not engaging with your content. Since I am using Mailchimp – I did this by creating a low engagement segment of:

• readers who have under 3 stars (which – in Mailchimp – represents the level of engagement with your content – meaning opens and clicks)
• readers who have been on the list for at least one month

This represents about 250 subscribers – so about 25% of my list.

3. How to ask?

Let’s not look at how exactly to ask readers to unsubscribe – this can be a tricky proposition – you don’t want to be vague and undecided, but you don’t want to be to forceful either. What I did was I crafter two emails – one with softer language and another that used some harder language to get the point across. Email 1: subject = “Please unsubscribe!”

Do You Find My Articles Useful?

Don’t worry if you don’t – I’m writing to let you know is perfectly fine to unsubscribe (see the BIG LINK below)

As you may already know, I’m sending you programming tutorials – once or twice a week. I am mainly writing about Java, Spring, Security and REST, and general web development. If that’s not your cup of team – I understand. So I just want to let you know it’s okay to unsubscribe. If you’re having trouble keeping up with my emails, or you’re just sick of getting them, I’d actually prefer you to unsubscribe — which you can do simply by clicking here:

The Big Unsubscribe Here Link!

P.S. If you’d rather continue to get my articles – you can check out my more recent one – the Big Kahuna of Java 8 Resources (I like to call it that). I am going to continue writing about these topics in 2014 – if you want to make a suggestion about how to improve my content, go ahead and reply to this email. Cheers, Eugen.

As you can see – the language of this first email is softer – while still having the clear purpose of providing readers with an easy one-client way to unsubscribe from my email list. Email 2: subject = “Yes, I’m really asking you to UNSUBSCRIBE!”

Do You Find My Articles Useful – if not you should probably UNSUBSCRIBE!

If you’re like me, and you get to many emails you don’t really care about, then you’ll find this email usefull. YES I’M REALLY ASKING YOU TO UNSUBSCRIBE!

As you may already know, I’m sending you programming tutorials – about twice a week. I am mainly writing about Java, Spring, Security and REST, and general web development. If that’s not your cup of team – I fully understand. So I just want to let you know it’s perfectly OK to unsubscribe. If you’re having trouble keeping up with my emails, or you’re just sick of getting them, I’d actually prefer you to unsubscribe — which you can do simply by clicking here:

The Big Unsubscribe Here Link!

P.S. If you’d rather continue to get my articles – you can check out my more recent one – the Big Kahuna of Java 8 Resources (I like to call it that). I am going to continue writing about these topics in 2014 – if you want to make a suggestion about how to improve my content, go ahead and reply to this email. Cheers, Eugen.

This second email, sent about 2 weeks after the first to the same segment of unengaged readers – is more forceful, and follows the same goal as the first email – getting people who are not interested in my list off of it.

4. Results – Conclusion

Keep in mind that I sent these emails to a targeted segment of unengaged subscribers – who have not opened or clicked through my emails, and who weren’t getting real value out of my content. However, with that in mind – the results were not surprising – the open rates were low for these emails, just as for the rest. First Email:

• open rate: 12.8%
• unsubscribes: 1

Second Email:

• open rate: 12.0%
• unsubscribers: 3

So – the results are not encouraging, but I am not giving up. In the next “Please Unsubscribe” email I will send, I will ask for feedback about what kind of content they would like to receive. On the upside, I did receive some personal (and positive) replies to these emails from readers who don’t want to unsubscribe from the list.

The post Asking you to UNSUBSCRIBE first appeared on Meta Baeldung.