Thursday, March 6, 2014

Targeting - Types of targeting in online advertising

Targeting

Targeting is defined as selection of an object of attention or attack. Same is what it means in online advertising as we target our specific users (attack) as per the requirement of the campaign.
Targeting is about catching the right user at the right time at right place so that the output coming out of it should be maximum.
Targeted Advertisement – an advertisement that is shown only to users
exhibiting specific attributes or in a specific context or at a particular time
of day.
In online industry targeting is a big world and is divided into sub parts :
Contextual Targeting : Contextual targeting is a form of targeting that the ad servers use to target a user for showing ad units based on the context of the page they are viewing.
Semantic Targeting : A type of contextual targeting that also incorporates semantic techniques to understand page meaning and/or sentiment.
Behavioral Targeting : Behavioral targeting is a form of targeting that ad server and ad networks use for targeting users based on their online behavior.
Audience Targeting : A method that enables advertisers to show an ad specifically to visitors based on their shared behavioral, demographic, geographic and/or technographic attributes.  Audience targeting uses anonymous, non-PII data.
Geographic Targeting : The targeting of users as per zip code, area code, city, DMA, state, and/or country is called as geographic targeting .
Creative Retargeting : A method that enables advertisers to show an ad specifically to visitors that previously were exposed to or interacted with the advertisers‟ creative.
Site Retargeting : This targeting enables advertisers to show an ad specifically to previous site visitors.
Time-based Targeting : Time- based parting can be divided into parts : time parting – this parting enables the ad to appear on certain time in a day while another is Day parting – this parting enables the ad to appear on certain days of the week.

Demographic Targeting : The targeting of users on the basis of demographic information such as age, gender and income etc is called as demographic targeting .

Tuesday, January 21, 2014

Basic SEO Question/ SEO Interview Questions

Basic SEO Question
Q1: What’s your SEO experience?
When you are asked this question, if you are a fresher, you must answer the theoretical knowledge you have. Say that you know about keyword research, on page and off page SEO, link building best practices and so on from a theoretical point of view.
If you are experienced, tell them the projects you have worked on, some of the rankings you have achieved and the time you took to accomplish this. It also helps to talk about technical aspects of SEO you have handled as well.
Q2: What do you think about the latest algorithm update from Google?
You must follow industry blogs and be updated about the SEO trends. So when you are asked about the latest algorithm update, you should talk about your views on what kind of sites/ search queries got affected, how it compares to the previous update which was rolled out and so on.
Here is a scenario. Assume the latest update rolled out was the Penguin update and the one before that was panda. You could say that Penguin update was mostly targeted at low quality content and not about links as the Panda update was. It hit heavily on content farms and thin content pages.
Q3: Are you technically savvy?
Not many SEO interviews focus on your technical skills. But those who really know what SEO is all about will really put your technical skills to test.
Most of the time, you’ll have to dig deep into a site code to find issues. If you are not at all technically inclined, doing SEO might be quite challenging. No matter how many tools you purchase, you still need some basic level of technical knowledge to succeed in SEO.
Q4: Do you take on freelance projects?
There are two things that an interviewer can find out from the answer you give.
If you take freelance projects, it means you are a go getter and that you are comfortable with interacting with clients and doing business development. This is a huge plus for any employee.
On the other hand, freelance consultants have a tendency to get clients for themselves instead of the company if it’s a service company.
If you are going to work in a company and their policy states that you should not take on other projects, you shouldn’t do it. It’s not only unethical; you could also lose your job and face prosecution for violating employment agreement.
Even if your company allows you to take freelance project on the side, it’s suggested that you don’t do it. Instead channel that lead to your employer. Not only are you showing commitment in your current role, but you are also contributing to the growth of the company that is paying your bills.
Q5: Do you use any SEO tools?
If you have been in the SEO industry for some time, you’d have come across many tools. If you have used tools like SEOmozbuzz stream, Wordtracker, Google keyword tool or any other tool, mention it.
The tools you used also gives the employer an overview of what kind of work and projects you have been associated with. If you say you are good at SEOmoz tools, then the employer will know that you are a professional because it’s a paid tool and not many amateurs can afford it.
Q6: Mention some SEO myths you know and what the actual facts are
This question helps the interviewer understand your knowledge of SEO. Instead of throwing some myths at you and trying to make you answer, this broad question will help them understand your depth of knowledge. Here are some myths and answers for reference
  • Meta description is used for ranking- Meta descriptions are not used for rankings by Google and only used as a short snippet of text which users see along with search results
  • Google will always use my meta-data- No. Google will not always use your meta title or meta description. Based on the query, Google may change your meta data to suit the searchers query
  • More links the better- False. Even one high quality relevant editorial links can trump 1 million spam links you might get via blog comments, forum posting, site wide links and so on (Not to mention your chance to get a penalty)
Q7: Which industry blogs and sites do you follow?
This question is asked so that the interviewer can judge how well you are updated with the current SEO trends. Some of the good ones to follow are
  1. Searchengineland.com
  2. SEOmoz.org
  3. SEObook.com
  4. Seo-theory.com
  5. Distilled.net
Q8: Do you do black hat SEO?
If you do black hat SEO, admit it. If you don’t, tell the interviewer why you do not follow such practices. If you are ready to stop blackhat and go completely white hat (which you should anyways), mention that as well.
Some of the reasons you do not want to use black hat are
  1. Risk of getting penalty
  2. Unethical
  3. Not a long term strategy
Q9: Can you tell me the difference between on page and off page SEO?
On page SEO refers to all SEO activities done on the website that needs to be ranked. They include
  1. Meta optimization
  2. Content creation
  3. Internal linking
  4. Fixing technical issues like page discovery, duplicate content, implementing canonical tags and so on
  5. Optimizing usage of keywords
Off page refers to any activity we do to get links to point to our site. These may include
  1. Guest posting
  2. Link outreach
  3. Viral link building
  4. Submission to niche sites
  5. PR outreach
Q10: What are your thoughts on buying links, article marketing, article spinning, and directory submissions?
Buying links: Don’t do it. Not only does it violate Google guidelines, it’s also a costly method which will backfire when caught.
Article marketing: Submitting high quality articles to article directories to get traffic and some link juice is fine. However, with the latest Google updates, most article directories have been rendered useless from a link acquisition point of view. Submitting articles to 1000’s of article directories is not a good idea.
Article spinning: It’s better to syndicate content rather than creating spun content. Article spinning rarely works and most of the spun content can be easily detected by advanced algorithms Google employs.
Directory submissions: It was valuable maybe 4-5 years back. But now it’s worthless. Links from directories other than maybe dmoz.org, botw.com etc is worthless.
Good luck!

FAQs in Online Advertising

What are companion ads?
Companion ads are the ones which are shown around the video ad. It can be rich media, image, text or skin too. These ads come in number of sizes. It is called companion because it runs as a companion to the video ad,
e.g. video ad of Levis apparel is running, the ad running alongside or surrounding the video ad will be a companion ad showing the sale or any new apparel of Levis brand.
What are the two ad impression measurement approaches?
Two methods are currently used to deliver ad content to the user are SERVER-INITIATED and CLIENT-INITIATED.
Server initiated ad counting uses the site's web content server for making requests, formatting and re-directing content while client (users’ browser) initiated ad counting relies on the user's browser to perform these activities.
Client side measurement involves a recognition on the client side of the successful delivery of an Ad impressions, and this information is communicated to a counting server for reporting while Server side measurement involves the recognition on the ad server side that an Ad Impression has been delivered, but with no assurance that the ad was actually rendered on the user’s browser.
Client side approach is responsible for accurate reporting of ad impressions as it is counted only when the user is served with the Ad
What do you mean by In-Unit clicks?
In unit clicks are those clicks which do require user initiation or action on advertisements but does not result in a transfer from the publisher site which generally happens in normal click or click-through which takes an advertiser site from a publisher site.
What does (Aborted) mean in HttpWatch?
Aborted is one of three status codes that are used in HttpWatch to display information about HTTP requests that did not receive a status code from the server: To be more precise, the browser started to process the request for a URL like downloading of an image (or 1×1), but then decides to cancel the operation, then the status code in Http watch is shown Aborted!!
What is $RND$ or any random number macro which is used in third party tag?
Normally Third party tags will contain placeholder [RANDOM NUMBER] which needs to be replaced by a random number or cache buster .The Ad server automatically converts the macro to a 8 digit random number when a Ad is served.
The main use of this random number is to differentiate the user, basically to count the unique users, if that random number is already and matched it is not counted.
What is CAPP?
CAPP is client alternative pricing plan, it allows companies or clients to choose that how they would like to pay for advertising, like if the campaign is CPM based it means the client will be paying as per impression, if the campaign is CPC, the client will be paying as per the number of clicks, similarly for CPA, CPL etc.
What is CDN?
CDN is content delivery network i.e. a global network of cloud servers that actually house the raw creative graphics to fetch the actual ad. It is where the creative are stored.
What is Date and Time Stamp?
The third party tags will contain a placeholder such as [TimeStamp] that you need to replace with a timestamp. Simply replace their placeholder with%%TIMESTAMP%% and the ad server will convert it to a timestamp in YYYYMMDDHHMMSS format when serving the Ad.
What is GateFile.html and for what purpose it is used?
In order to enable Safecount Link/Layer invites to work from within an of an ad tag or a publisher’s advertising placement, a file, “GateFile.html,” has to be placed in the root directory of each domain of the site where Safecount tags will be issuing survey invitations for research purposes.
This is a one-time request. Once the GateFile is there, all Safecount tags will be able to use Link/Layer invites through the inclusion of simple JavaScript tags on the desired HTML pages or ad served banners.
In short, this gatefile is to be placed in the root of every publisher, which is a quite tedious job, but it’s just one time work!
Example: The publisher is dictionary.com, we need to place a gatefile.html on its root directory,
The idea location will be like :
“//dictionary.com/safecount/GateFile.html”.
What is RESPAWNING?
Re spawning is a process by which the tracker which is deleted can be repopulated. If one storage location is cleared, the tracker can repopulate it with the same value from the other storage location. This can be achieved by storing the same identifier at different storage locations (HTML5, Flash storage, Etags etc.)
Reverse re spawning is also quite possible i.e. repopulating of local storage from cookies.
What is the basic difference between first party cookie and third party cookie?
A cookie set under the domain of the website embedding a tracker is site-owned cookie i.e. first party cookie e.g. Facebook‘s cookie is a first party cookie when the user visits facebook.com but it becomes third party cookie when someone use “Like” button of Facebook on a different website.
Similarly the third party cookies are the one when any other website tracker is placed on another website code.
What is the need of parameter “jump” in the code or Tag?
A command, passed to the ad servers that is used for counting clicks. This command is substituted dynamically with the click-through URL of the creative.
What is the purpose of “ord” parameter in the code or Tag?
The ord value is used for cache-busting; to ensure that a fresh call is made to the ad server every time a page loads. Without the ord value properly implemented, you may experience impression counting discrepancies.
There needs to be a piece of code in the top of tag that generates a random number to populate the ord value on every tag.
Reference: var ord = Math.random();
What is “ncu” in Eye Blaster (Media Mind) tag?
It is a parameter defined by Media Mind which is normally used by the Ad server to track clicks and also can be used for custom logging URL’s that is we can pass the required Custom URL’s using this parameter.
For your reference :
<script src=”https://bs.serving-sys.com/BurstingPipe/adServer.bs?cn=rsb&amp;c=XX&pli=XXXXXXX&PluID=X&w=XXX&h=XXX&
ncu=$$%%CLICK_URL_UNESC%%$$&ord=%%CACHEBUSTER%%“>
Why Click macro shouldn’t be embedded before src (source) line but after href line?
If you put the click macro before the “script src” the tags will count with every impression as the tag appears when the script fires so when you have inserted the click macro it will fire together which will result in counting of clicks, and when you put the macro after “href”, the clicks are counted only when the landing page opens which will open only when someone clicks on the creative.
Why is “ abr=!ie “ used in DART tags under Jscript part of the whole tag code ?
abr = !ie helps in giving the command to the Ad server for not serving the Jscript tag when the browser used by the user is Internet explorer . The !ie denotes “NO INTERNET EXPLORER” , now why is it used is because all the browsers are supposed to read iframe tags first if it supports and if not Jscript is served so at times on IE both becomes possible to be served due to some obsolete versions of Internet Explorer which reads both iframe and jscript and which will lead to serve two tags for one dart tag , one iframe and the other jscript which is wrong anyway so just to avoid this kind of scenario abr = !ie is used.
Do the banner size changes as per the screen resolution of the mobile phone? If yes then How?
Yes the banner size automatically changes with the help of view port settings. The banner size automatically scales to fit the screen using the view-port settings built into the apps, generally app web view objects use view port settings for density independent pixels by which say a 320×50 banner will scale depending upon the resolution of the phone screen. The scaling problem can be removed by only changing the settings in the app where you are showing the ad so test it properly on the same app to see is it scaling properly or not.
How to serve rich media ads ob Blackberry 10?
Every one is concerned about showing banner on iOS or android but what about the users using blackberry? The main RM providers like celtra, goldfish also don't support for BB which is strange but yes it is so. One way by which this problem can be sound is serving the ad directly from server and not by using any SDK. UnicornMedia is the one who is capable for doing it easily.
Reference link: www.unicornmedia.com
Some Best Mobile Networks for In-App Ads?
Advertising.com (more successful is serving mobile brand campaigns with quite high ecpms)
Millenial
Lifestreet
Buzz City
Vserv
xAd
One97
CPX
Mocean Mobile
How to make the ad load only in “In view” area?
Disable BTF though reducing inventroy but this will always help in serving the ad on in view area
Enable ATF
Pre-load the ads so that it doesn't delay in serving on the page
Avoid serving ads on sites having high bounce rate, mainly photo gallery sites
Prefer JavaScript tag than Iframe tags when the objective is to serve view able impressions as solving view ability with Iframe is too tough and is still in experiment to solve the issue about it, basically challenge with iframes is that iframes that come from a different domain delivering ads, is that the view able impression code within the iframe is not able to access the publisher page due to the “Same Origin Policy.