Mobile advertising with ZestADZ

15 06 2010

This blog post is part of a series in which we’ll be reviewing self-service mobile advertising networks, to give you an insight on how they work, what they offer and to save you some time when setting up your campaigns. We’ll be analyzing the results using Bango Analytics.

ZestADZ in a nutshell:

Can I set up my campaign online? Yes

How long does it take to set up? Less than 2 minutes.

Payment options: Wire transfer or PayPal

Minimum payment: $100

Bidding options: CPC and CPM

Targeting:

You can target By Countries and Carriers, Device manufacturers and devices and also Required capabilities. This option lets you choose platforms (J2ME MIDP 1.0, J2ME MIDP 2.0, Symbian, Windows, RIM, OSX and Linux) and properties (MP3, Video and Midi)

You can also bid by Channel and the minimum bid is $ 0.05. The channels are: Entertainment, Downloads, Sports, News and Information, Communities, Portals, Ringtones, Videos, Photos, Careers, Technology, Education, Business, Health, Games, Finance, Weather, TV and Others.

Tricky bit: The URL I used to track the campaign was too long. I had to put a shorter URL and send them an email so they could amend the URL for me to include the tracking parameters. They did that immediately, so do not hesitate to contact them if you have the same problem!

Annoying bit: If you decide to do a wire transfer, and you are a non US resident, you need to send the money to a different account from the one shown on their website. Get in contact with them because otherwise you’ll be wasting time. It took me nearly a month to get the money into my advertiser account.

Great stuff: They have good APIs for publishers to get more information about the performance of the ads. Unfortunately they don’t have any for advertisers, so it is highly recommended to use an alternative analytics tool.

Analytics provided:

Impressions, Clicks, CTR and average CPC.

What I chose:

I chose US, UK, Spain and South Africa to check what kind of inventory they have in these countries and All devices and all capabilities. For channels I chose entertainment, downloads, sports, news and Information, communities, portals, videos, photos and health and my bid was $0.33 with a total budget of $100.

My observations:

To check the campaign results I used Bango Analytics. For country targeting I got 91% US traffic, 5% Spanish traffic, 1.5% UK traffic and 0.08% South African traffic. Almost 2% of traffic came from countries that I hadn’t targeted like India, China or Tanzania.

Even though I targeted my campaign by channel and devices, I didn’t get results on ZestADZ based on these choices. Using Bango Analytics I could filter the results by devices and it showed me that 83 % of my unique visitors where using an iPhone or an iTouch. From the rest of devices there was a high percentage of smartphones. I even saw a couple of iPads!

I exported a traffic report on Bango Analytics to check the referrals to get an idea of where the ads where being displayed. But when it came to the iPhone traffic, all I could see was the Mobclix platform, which doesn’t give me any info of where the ads have been displayed. For the rest of devices I saw a good variety of mobile sites.

Data to look at to fine tune this campaign:

Analysing conversion rates

The call to action for this campaign was to download a free application, a BMI calculator. Using Bango Analytics I was able to check the conversions, by tracking how many visitors clicked on the download button. The conversion rate for this campaign was 16%, which is very good!

But after seeing the conversation rate, which in this case was very high, surely it’d be good to get more detailed data. So I used the filters in Bango Analytics and I was able to see the

Bango Analytics Unique Visitor by Hour

performance across different days, here’s what I found:

  • I filtered by Unique Visitors and saw that Friday performed much better than Saturday, Sunday and Monday, with more than 3000 unique visitors.

  • I was curious to see what caused the spike in traffic on Friday so I then filtered visitors by time of the day, and was able to see that the campaign ran for 12 hours on Friday and 3 and a half hours on Saturday and Sunday. It would be interesting to see if by running the campaign for the same amount of active hours each day, we would get the same amount of traffic

  • 18% of US traffic converted, 13.7% for and only 5% of the traffic in Spain. Maybe it is worth to put our budget on US traffic when using this ad network; even though the UK converted well the amount of traffic coming from this country wasn’t so high.

  • Regarding devices, Apple iPhone and Apple iTouch were by far the devices that downloaded the app the most. Even though the app we were advertising was not a native iPhone app, these users seemed more prone to click the download button. Interesting!


Would you like to see the result of this campaign on our demo account? Just follow these easy steps:

  1. Go to the Bango Analytics platform by visiting the bango demo
  2. Go to “Analytics reports” and change the date range to start from 4th June 2010 and end date 7th of June
  3. Click on “Campaign and goals” on the left hand side and then click “Campaign parameters”
  4. Filter the report by clicking on “Create a new filter” at the top of the report and drag the “Campaign parameter” tab across. By selecting the relevant “Source” you will be able to see the results and also compare all the other mobile advertising campaigns we ran at the same time. For the ZestADZ campaign choose Zestadz_BMI
  5. To see how many people clicked on the download button, simply edit the filter you have previously set up and drag the “Page title” tab across. Select BMI Calculator DOWNLOADS and apply the filter.

We invite you to visit our demo account and check other filters and play with the different reports. If you like what you see and you would like to try Bango Analytics, sign up now for a free trial.

Share: add to del.icio.us Add to Blinkslist add to furl Digg it add to ma.gnolia Stumble It! add to simpy seed the vine TailRank post to facebook





Device analysis and call to actions

21 05 2010

The arrival of the iPad to the market has brought new marketing opportunities and new discussions around whether it should be consider a mobile device or not. Whichever way we go, we should really consider the proliferation of different devices such as Netbooks, Kindles, iPads and alike. They are becoming more popular and they are connected to the internet either via WiFi or 3G. Bear in mind that even though these devices are portable and internet connected, they are not communication devices.

But whether they are mobile or communication devices is not the only factor to consider when planning your marketing strategy. Most importantly, you should be considering if these devices are in fact relevant to your business, think about how you are going to adapt you messages and call to actions to these devices. And you can only do do this by analyzing the traffic to your site.

iPad scottevest

Scottevest's big pocket for big gadgets

  • Click-to-call won’t work
  • Voucher redemption won’t be ideal, they are portable devices but they won’t fit in the pocket of your jeans!
  • Video ads will become the holy grail, so think how you can make them different and how to get your users involved in the video experience.

My colleague Andy Bovingdon has published a very interesting blog post  The iPad…so is it mobile? in which he analyzes whether we should consider this device as a mobile device, and gives us some figures on the size of the Netbooks market.

Another recommended reading by the same author but from a more strategic point of view can be found on mSearchGrove blog Should iPad Be High On Your Radar? Or Low On The List?

I hope you enjoy these readings and get a better understanding of how crucial is to accurately identify what devices are important for your product or service proposition, and how to adapt your strategy.

Share: add to del.icio.us Add to Blinkslist add to furl Digg it add to ma.gnolia Stumble It! add to simpy seed the vine TailRank post to facebook





Lost in the mobile maze, free whitepaper

14 01 2010

If you feel like you’re living in an episode of Star Trek when people in the industry talks about transcoders, latency, cookies, gateways, tracking codes, markup languages and postbacks,Lost in the Mobile Maze today is your lucky day!

Today, my colleagues at Bango have published a new white paper entitled “Lost in the mobile maze”, explaining Why mobile complexity continues to challenge the biggest and best in the industry.

This white paper sets out the challenges that need to be understood to succeed in developing mobile solutions and explains all the bits that need to be considered when taking your brand to mobile.

Now you´ll be able to understand the conversations and make better decisions about your mobile strategy. Download your free copy today and “geek up”!

Share: add to del.icio.us Add to Blinkslist add to furl Digg it add to ma.gnolia Stumble It! add to simpy seed the vine TailRank post to facebook





Would you like a piece of the BlackBerry pie?

8 09 2009

We have been talking lately about the increase of smartphones accessing the internet via Wifi connections, how iPhone is gaining positions on our device ranking and today Bango published a free whitepaper Increasing your slice of the BlackBerry pie, to help us understand and overcome the challenges of how can we accurately identify BlackBerry users for tracking, CRM and marketing campaign purposes; and also how is it possible to enable these customers to make transactional payments on their mobile phone bill, just like any other phone.

Smartphones have become the standard handset of choice in the mobile phone market – even during global recession the growth in smartphoneBlackberry WP banner sales continues to soar. Gartner reported a global smartphone sales growth of 27% year-on-year in Q2 2009 compared to a 6% drop in cell phone sales during the same period. And the numbers in the USA were even more dramatic during the second quarter. The research firm NPD Group reported US smartphone sales grew by 47% and accounted for 28% of all handset sales.

Get your copy of the white paper and find out:

•Why BlackBerry has become a popular consumer device

• How BlackBerry users consume data services

• The challenge of capturing identity and delivering payment to Blackberry users

• Optimizing your mobile services to overcome these challenges

Share:  add to del.icio.us Add to Blinkslist add to furl Digg it add to ma.gnolia Stumble It! add to simpy seed the vine TailRank post to facebook





Mobile Analytics shows iPhone gaining positions on device ranking

28 08 2009

With all the hype around the iPhone, we couldn’t help but wonder how many iPhones are accessing our mobile site. So we checked our Bango analytics package and had a look to the most popular devices accessing our mobile site since February. The data reveals the following:

Month iPhone position
February 5th
March 7th
April 5th
May 4th
June 4th
July 26th *
August (so far) 2nd

*we run a mobile campaign in June targeting everything but iPhones.

More and more mobile advertising networks have now specific targeting options aiming at iPhones. There are specific publisher groups so ads can be displayed on iPhone apps.

Taking all of this into consideration we have decided to create our own  iMobislim. Yes, you read it right. Our own iPhone specific site is on its way. Our wap master , Tom Thurston from Mobile Impossible is developing the site and this is what he has to say:

Bango analytics told us that iphone users were spending less time on pages, viewing less pages and were becoming an increasingly large proportion of our users. Tests found that due to O2 image transcoding, our iPhone users were seeing poor page presentation. So we decided to give our iPhone users cool ajax based slidey screens, some glossy iPhone icons, and we used png format instead of gifs and jpeg, which escape the dreaded transcoder. The site looks a lot cooler on iPhones now, almost like a native app – we’re hoping that our next analytics run will show an improvement in browsing stats for our iPhone users.
We also found out that even iPhones are going up in popularity the percentage of traffic that represent is not so high in absolute values. So don’t stop advertising to the rest of the handsets, because you will have not such a big market to engage with. Just fine tune your targeting capabilities and the creatives when advertising to iPhones. You can find out all the information about handsets by accessing our live Bango Analytics package and clicking on devices. To get more in depth data, filter the reports.

Millenial Media’s July SMART shows this same trend on their device highlights:

  • iPhone continued to dominate the top devices category with a 12.21% share, a 3.62% increase over June.
  • iPhone/iPod Touch impressions increased 29% across the network in July.
  • Apple continued to close the gap; however, Samsung remained the top device manufacturer in July.
  • Two new mobile devices entered the top 20 in July –the LG enV2 and the Samsung SCH-U410.

You want to have a preview of iMobislim? Check this out! Browse the whole site on http://iphone.mobislim.com/

iMobislim homepage iMobislim radish diet iMobislim registration

Share: add to del.icio.us Add to Blinkslist add to furl Digg it add to ma.gnolia Stumble It! add to simpy seed the vine TailRank post to facebook






Setting up a paid subscription option for Mobislim BMI Java App

6 01 2009

Extending the basic Java application using Bango’s web services.

 

Following on from my first post, my second objective was to develop a way to make it so that my application will only work if the user has a valid Bango subscription with some time remaining on it. This started with quite a lot of research into how I was going to do it.

 

The first obvious problem was how to get the application to find out if the user has a valid and active subscription.  I played around with a few different techniques and approaches, the one that appeared to be the most successful was to have the application connect to a web server which would then send a web service request to Bango. The web server then returns some data to the phone (Illustrated right).

 

So the first stage was to develop a script on the web server to get the correct information to the device. The information I needed to get to the device was really only the time remaining on the subscription. I did however actually decide to send a few extra pieces of data to the phone too though for some further ideas I had.

 

To get the subscription status, we have to go through three major stages:

  1. Get the users Bango user ID
  2. Get the users active subscriptions on a Bango number
  3. Get the time remaining on the subscription

 

To do this I used PHP as my server-side scripting language, and this is what I came up with.

 

<?php

 

$username = “myusername”; // Bango web services username

$password = “mypassword”; // Bango web services password

$bango    = “000111222333”; // Bango number with subscriptions

$script   = “http://domain.com/userid.php&#8221;; // URL of this script

 

if(isset($_GET[‘b’])){

 

       $client = new SoapClient(“https://webservices.bango.com/subscriptions/service.asmx?wsdl&#8221;);

 

       $SubscriptionID = $client->GetSubscriptionID(array(“request” =>

       array( “username”    => $username,

              “password”    => $password,

              “bango”       => $bango,

              “userId”      => $_GET[‘u’])

       ));

      

       $SubscriptionInfo = $client->GetSubscriptionInfo(array(“request” =>

       array( “username”        => $username,

              “password”        => $password,

              “subscriptionId”  => $SubscriptionID->GetSubscriptionIDResult->subscriptionId)

       ));

 

// Redirect to same script, but only with the variable we need.

// We then scrape the variable in the app.

       header(“HTTP/1.1 301 Moved Permanently”);

       header(“Location: ” . $script . “?time=” . $SubscriptionInfo->GetSubscriptionInfoResult->accessMinutesRemaining);

 

}

 

?>

 

 

I did actually have quite a few problems with PHP and SOAP. There are quite a few ways of using SOAP; I used the PHP SOAP module (http://uk.php.net/manual/en/book.soap.php).

 

I then setup a Bango number with the identifier function enabled which redirects to the script. I was now set to work on two more things:

  1. Getting the data into the application
  2. Using the data to make the application expire

 

String scripturl = “http://domain.com/userid.php&#8221;; // The same url you entered in the php script

String bangoid = “444333222111”; //the bango number with identifier function turned on

 

       HttpConnection conn = null;

 

       String url = “http://bango.net/id/?bango=&#8221; + bangoid + “&register=n”;

        try {

           while (true) {

               conn = (HttpConnection) Connector.open( url );

               conn.setRequestProperty( “User-Agent”, “Mozilla/4.0” );

 

               rc = conn.getResponseCode();

 

               if ((rc == HttpConnection.HTTP_TEMP_REDIRECT) ||

               (rc == HttpConnection.HTTP_MOVED_TEMP) ||

               (rc == HttpConnection.HTTP_MOVED_PERM)) {

                   // Get the new location and close the connection

                   url = conn.getHeaderField(“location”);

                   conn.close();

 

               } else {

 

       String timeremaining = replaceAll(conn.getURL(),scripturl + “?time=”,””);

 

                    conn.close();

                    break;

 

               }

 

            }

 

        }catch( IOException e ){

           // handle the error here

        }

 

This also requires a small method called replaceAll, below.

 

private String replaceAll(String text, String searchString, String replacementString){

    StringBuffer sBuffer = new StringBuffer();

    int pos = 0;

    while((pos = text.indexOf(searchString)) != -1){

        sBuffer.append(text.substring(0, pos) + replacementString);

        text = text.substring(pos + searchString.length());

    }

    sBuffer.append(text);

        return sBuffer.toString();

}

 

I then just compared the variable timeremaining against 0. If the number is greater than 0, the application will start. If not we tell the user that the subscription doesn’t exist or has expired.

 

if(Integer.parseInt(timeremaining) > 0){

       // Code if not expired

}else{

       // Code if expired

}

 

What could be easier?