Search, PPC, Performance, Social, Turkey

PPC campaign automation and bid management: Software solutions and Google AdWords features

Wed, 01 Jun 2011 10:30:00 +0300

Automated bid management tools: will they improve our campaign performance?

There are a lot of third party vendors on this market including Marin Software, Kenshoo, Acquisio, Apex-BidMax, Adobe Search Center (pformerly Omniture) and many more, and recently Google also quietly launched it’s own tool built into AdWords. Let’s take a look.

First: why need to automate campaign management or optimisation?

As an inmediate answer I would say there is no need. Every action a software helps us in can be done manually – we need a person with professional Excel and analytical skills and with lots of time available. We’d make up our mind when we start considering human resources as a part of costs in the ROI function. Simply, we can automate the most time consuming parts of the campaign optimisation by using software solutions.

Manual, semiautomatic and automatic optimisation?

Manual optimisation: obviously changes made manually on bids and other parts of the campaign.
Semiautomatic: when we create rules in the bidding engine but we don’t allow the rule to run automatically – so that we check each and every change the rule would make before letting them go live.
Automatic optimisation: There are two levels of this. First is when we set up rules and let them run from time to time – without checking that each and every time they run. The second level is when we do not use rules but we let algorythms to optimise the campaigns towards the goals. This is the less transparent solution but can work well.

Which provider to choose?

First, there is a built in Google AdWords solution for Google search campaigns and that one is free to use. As Google’s automatic CPA optimisation tool that lacked setting options has died in lack of use and sometimes bad results – they had to come up with a solution. The ‘Automate’ function has just been released a few weeks ago (in Turkey) with no hype created around it – Google just simply copied the solutions of third party vendors. Easy choice.
There are quite a few third party vendors in this arena: their pricing model is different ranging from 1.5 to 3% on spendings, their functions and usability is also quite different.

My only advice is to check and try them all. As these solutions are designed for making our professional life easier I’d choose the one that’s the easiest to use.

The basic functions of bidding engines

Campaign editor
This is a basic feaure although most software solutions deal with this very differently. Some vendors have the option to copy the whole campaign structure (including keywords, ads etc.) to another channel, others don’t. Some include a fully featured editor, some require to make some changes in the engine. (Just like in case of Google AdWords Editor, there are some campaign settings you can only make in the engine.)

Cross platform optimisation
Simply said: managing not just search but display too. This already is a significant feautre on matured markets and also gained momentum as Facebook and its advertising solution started to evolve. This allow advertisers / agencies to manage campaigns using the same metrics and goals whether they should be text/image/video ads or combinations of these – on any channel these campaigns run on.

Cross channel management & optimisation
Simply said: managing different search channels. This is a feature that was developed focusing on markets that are not Google dominated. Same metrics, same goals, unified reporting and easy managing of multiple search channels. This usually include Bing, Yahoo, some software also manages the Chinese Baidu! and Russian Yandex and of course Facebook campaigns

Automated reporting
A very useful feature – pulls all data together and reports using the same template. Excel does that too – but filling in the cells from many different sources can become a nightmare. Some software offers customisable charts too – bright presentations to the Client or to the Management by the click of a button.

Bid rule examples
Examples always come handy when learning how to manage bids and/or when experimenting new approaches on campaigns. As there are unlimited bid rule setting options are available, sometimes it’s easier to choose a good one from the templates.

Data integration
First, automatic integration with engines and media through APIs. This is a basic and obvious feature.
Second, conversion tracking and/or web analytics systems. This is not as obvious as the previous so it’s important to learn which tracking systems are connected with the bidding engine.
Third, data and list vendors. It’s enough to think about remarketing and we see why this part is important.

Effect simulation – pending changes
I think this feature is the most important of all. We set up rules but before running them we need to see what keywords / placements / etc. will this have an effect on and what will be the final outcome of running the rule. This allows managers to finetune rules and change them before it goes live. Unfortunately this feature still not works well in the Google AdWords interface but I hope they’ll include this.

Automatic optimisation based on algorythms
Some vendors created specific algorithms that contain multiple rules and deeper relations between data to automatically optimise campaigns.

To sum up if time is important and you count that as a cost factor in calculating your campaign ROI then you should choose a software to automate some parts of your campaigns or the whole process. Start with extensive testing of the different solutions – functions are not too much different but user experience and ease-of-use is.

Best practices on getting more Facebook Likes (with Turkish market insights)

Wed, 30 Mar 2011 13:41:00 +0300

This post is to provide you with some insights to hundreds of Facebook campaigns and some Social Media projects and aims to answer the question: which channel is the most effective driving new Likes to your Facebook page?

How much is your Cost per Like? – The Turkish market

Back in the times when Facebook advertising was still new to the Turkish market (around the beginning of 2010) my colleague took a look at the Facebook campaign management interface and asked, where is the Cost per Fan (Cost per Like) figure? At that time this question seemed clever, but a little bit awkward. Then we had to face that this metric became one of the most important ones regarding Facebook Fan Engagement campaigns.

Love brands get it cheaper

Webtrends recently created a free report on the same topic and they compared industries in terms of CTR and CPC. Their conclusion is simple: the more ‘socialized’ the brand is, the bigger the CTR and hence lower the CPC. The range is wide from USD 0.12 to USD 1.27 per click. Unfortunately the study does not go into further detail on Like conversions but we can see that the US average is USD 1.07 getting a Like.

Source: Webtrends (http://f.cl.ly/items/2m1y0K2A062x0e2k442l/facebook-advertising-performance.pdf)

We also summarized dozens of our local (Turkey) Facebook engagement campaigns and found that well known love brands tends to have lower Cost Per Like numbers ranging between 0,08-0,4 TL per Like, the less known but still big strong have a CPL between 0,5-1 TL and niche or startup brands can get fans for 1-1,5 Liras.
How much is yours?

Which format or placement to choose?

If we talk about campaign performance and set up the campaign metrics perfectly - every channel and placement becomes comparable. If we talk about video ads, we can compare Cost Per View numbers in Facebook Video Engagement ads, Google Display Network video ads and of course all local websites.

If we talk about getting Likes the two obvious channels to compare is Facebook Homepage ads on a CPM buy and Facebook Fan Engagement ads on a PPC buy. After testing the two channels against each other we could say that for some specific Clients the Engagement ads bought on a PPC basis performed significantly better. It means that the Cost Per Like figure was 3-4 times lower than using the homepage ads. Performance based advertising pays off.

What to expect in 2011?

Increase in prices – mostly in CPCs. During 2010 in Turkey we experienced an increase in average CPCs of about 100% and this trends seemingly continues throughout 2011. We expect that the slightly deccreasing average Google AdWords Search CPCs and radically increasing Facbook Ads CPCs will meet this year.

What’s next?

As we see that Facebook PPC ads converting Likes significantly better than other channels we most probably will see this trend continuing with another 100-150% increase in Facebook CPCs this year.
Besides this, new advertising formats have started to roar in, e.g. Facebook Sponsored stories. They are perfectly using the social engagement of our friends so we expect that even though CPCs are increasing, we can get Likes for less money than before. Of course the limitation is our current Fan base as we can only rely on our current Fans to bring in their friends. We do have to mention that Facebook introduced this advertising format together with the rollout of their new News Feed management rules. Based on the new algorithms Facebook page messages only appear in Top News, not in the News Feed any more. This means that even though a brand has already built up a community it has to spend a great deal of money on Sponsored Stories to keep visibility up. That’s actually another story…

(Disclaimer: We aggregated the above results from hundreds of campaigns in Facebook and Google AdWords in Turkey – in agency account - but of course results are depending on how different sectors are represented in our Client portfolio – so it might be different agency by agency.)

How about Facebook Applications?

We saw many Facebook applications this year from almost all types of brands. In general, the cost of planning, designing, maintaining applications, putting incentives and launch campaigns on top makes Facebook Applications an expensive experiment. If we break down all related costs to the number of Likes we may see a number sky high.

Of course, users Liking our brand page from an application are more engaged having spent some time with the brand and having a good experience in the application. This part of course is very hard to measure and compare and that’s why a Cost Per Like is not the only metric we should use for Facebook apps.
Designing a very successful Facebook Application is very similar to creating a viral campaign. If we found the basic motivation why people should share, they will share. If you have already created viral campaigns you might know how risky it is. Needs creativity, a very good method for selecting ideas and seamless execution on the agency side and risk taking, trial and error approach and a deep wallet on the Client side.

There are and there will be very well performing Facebook brand Apps but I’d advise brand managers to go for the Fanpage Engagement ads too – to be on the safe (and cost effective) side.

Seeding – is that really obvious?

It’s an abvious choice – and the strong local blogger and social media community seemingly take the job. Seeding can work better if we plan incentives, support team and of course a strong message. Sometimes we can see that they spontaneously create a small ‘blogger’ community around the brand and support it. As the reach of blogs is quite limited, this project either cannot be judged upon Cost per Likes only – or if we do that, we end up in saying that those few thousand Fans we gathered were too expensive. We’d rather use different KPIs for seeding like number of video views or post views besides the number of Fans collected. (Have to add that the number of Fans coming from seeding projects can only be estimated as referrals are not shown correctly in Facebook Insights and there is no other way to measure this. I think we’ll see some new things coming from Facebook on this field soon.)

The power of owned media

I should have put this first. Do you have a one click Like button or Like box available on your homepage and throughout the site? Have you put that to the most visible place? If not, this is the time to do it. Obviously the bigger traffic your site has the more effective you can drive Likes to your Facebook Page. Costs not to be associated to this but surely it’s a must.

Content is king – you already know it

In many terms on Facebook we can see the past trends to come back again. As history repeats itself we can see that e.g. greeting cards live their second life. Posting nice greeting cards on events (Valentine’s Day, Christmas etc.) and asking people to tag themselves on will have a nice viral effect. And people will tag themselves. Sending one greeting card to all by two clicks is definitely easier then sending them one-by-one. Effectiveness, effectiveness!
Same is the story with simple gift supported questions. Ask the people to tag themselves, to look for a tiny bit of info on the product package or to answer a simple question and give away a small gift. Of course as answers and interactions appear on their feed will trigger more fans to come. Not to talk about the overhyped feedback rate of the post…

Summary: Facebook engagement ads on a PPC basis

So as a summary it seems that nothing works better in getting more fans than PPC Facebook engagement ads – supposing that you’ve already made the other necessary steps, have updated content and a well managed community on your Facebook page. As Facebook PPC prices are on the rise, we’ll see other ways becoming more interesting in reaching out to new fans, but surely the next few months are still about this tool.

(Sidenote: This article is a good example why digital performance marketing and social media is strongly interconnected. We as marketers have to come to the conclusion that we need to implement performance metrics to our social media activities to the extent technology allows today.)

Infographic of Social Media Equivalents in Turkey

Wed, 03 Nov 2010 15:54:13 +0200

Tracking down the origins of this chart I found that was published for a lot of coutries including China, Japan, Korea, Vietnam and even for Hungary. So I created one for Turkey too.

I changed one section of it as I think there should be a space for the rising private shopping sites: Limango, Trendyol and Markafoni. Maybe the ‘Deal of the day’ section should contain these but I put the group buying sites (ŞehirFırsatı - Groupon, Yakala.co, grupanya! and grupfoni) there instead as they seemingly work with shorter deadlines and with a focus on group deals.

Easy to notice that YouTube has a lot of local metoo sites but as YouTube ban was lifted their traffic has already begun to decrease. Sad news for the local social network Yonca and the local check-in site Gezenzi that their international competitors also have proven to be too strong - they are winning the game.

Also interesting that among the local websites (not including Google or Facebook) the second largest is a blog engine Blogcu.

Facebook ads: the missing features

Tue, 21 Sep 2010 13:38:15 +0300

We marketers all know Facebook ads well already. As the online ad spending in Turkey is still growing with 50+% a year, ads are really cheap either if they are priced on a CPM or CPC base.

This is the main reason why the globally set CPMs for Facebook proved to be way too expensive here forcing all advertisers to use PPC ads. And they were suprising for many of the advertisers with really low prices.

Some reasons why I miss some features from Facebook ads. Some new features:

can make agencies’ life much easier,
can make much more money to Facebook,
can help advertisers to target their ads more precisely,
can help users to get rid of annoying unrelevant ads.

All parties in. Some of the features missing are so basic I can hardly believe they don’t have a priority. OK, let’s see them:

Agency work:

bulk uploader for ads
campaigns separated by advertisers (it’s really annoying to crawl thru hundreds of campaigns to find one…)
bulk managing keywords (interests, sry)
a way better keyword finder
keyword targeting based on not just interests but wall posts and/or websites given in personal data
keyword reports
conversion costs for engagement ads
ad placement reports (like ad positions)
better charting (and of course no 2 day delay in charts…)
description of new features (This one bulletpoint would deserve a whole post as I often see some nice enhancements on features and/or usability - and find no information in help as the new feature is rolled out without updates in the help section. Pffff.)

More money to Facebook (in addition to the above):

better placements for PPC ads (as CPM is not going well)
country based pricing for CPM ads
broadening keyword targeting (pls. see above)
Frequency capping (I know Facebook tells that there is no frequency cap as they would like to push advertisers towards CPM ads - where frequency capping is available. But this initiative is about to fail anyway. Frequency capping helps higher CTRs so higher income from the same inventory for Facebook.)

Advertisers (in addition to the above):

Opportunity to change ad title for engagement ads
More options to see the possible inventory and estimated campaign outcome for campaigns.

Users: of course all the above helps them to receive more targeted, so more compelling ads.

And today we are not talking about that how many unnecessary features are in the Facebook ad system. A lot…:-)

Social media marketing and MLM combined - a brilliant example

Wed, 18 Aug 2010 18:39:00 +0300

This year, the Central-Eastern Europe region will see a new e-business and e-marketing conference coming. Actually Webit is not new as this is their second year. As the conference is a bit far away (in most terms) from Central-Europe they had to come up with something special. And what they are doing is extraordinary.

By clicking this link you’ll arrive to a page that explains the concept: The Most Influential People game. And we people involved in online marketing are all part of this game IRL aren’t we? Influencing other people, sharing, publishing.

You’ll see why I mean this concept is brilliant. No forcing, no spamming, no hidden stuff, all is clear and simple. You share (if you want, of course) the news about an upcoming conference and you’re rewarded with a special honor that all of your influence is being tracked. Go and get in the game, even if you’re not interested in the conference but want to see a perfect example of social media marketing.

Actually, the share-and-get-rewarded concept is nothing new. Back in the year 2000, I visited an online marketing conference in New York when I was rewarded with a free ticket being one of the top25 influencers worldwide. That was a long time ago, the only tool that helped me spreading the word was e-mail. There was no Twitter, no Facebook, no sharing, no blogs, Google was almost ‘invisible’. What is new in this thing is how they combined the channels. It’s brilliant. Congratulations, guys!

Webit is the biggest e-Business & e-Marketing digital industry event for Central and Eastern Europe

Twitter doubled within a year - top Twitter countries - June 2010

Fri, 13 Aug 2010 13:32:00 +0300

I found these figures in a recent press release from ComScore which came out two days ago. Worldwide growth was 109% from June 2009 to June 2010 which also means that Twitter is very close (maybe they already passed since June) the big barrier of 100 million users.

(Interesting fact that DoubleClick Adplanner shows 190 million users wordwide. The difference is too big to just say this is because of the different methods.)

Asia Pacific tops the chart with 25 million users while Latin America is closing very quickly, the number of Twitter users tripled over there in this period.

Turkey is the 13th on the list with 11% penetration and that is fairly close to the 3,5 million DoubleClick adplanner shows for the coutry.

So Twitter is marching towards reaching hundreds of millions - and although a lot of specialists (including me :-)) said this surge is ending soon - it does not seem like.

Media expenditure skyrockets in Turkey

Fri, 06 Aug 2010 18:11:00 +0300

The below chart does not need to much explanation. In the recent years the Turkish media also suffered from the effects of the downturn but 2010 first half showed a dramatic change. The growth rate shows that the whole market (not including search) will be bigger by one third this year. Wow.

TV is the old-new hero besides the all-time-best-performer internet in terms of growth rate. What a surprise. Also is that even newspapers are on the growth.

Search (actually anything text based) advertising is not included in the report. There are not so many sources for estimating the size of the search market but one thing is sure: it’s already a lot bigger than display. As of my guess the search market should be around at least EUR 80 million this year.

The figures were put together by the Advertising Association of Turkey, you can find the original chart here.

City tourism marketing strategy idea - the second runner-up on Call4Istanbul competition

Sat, 10 Jul 2010 21:01:00 +0300

The digital advertising competition, Call4Istanbul was announced by the Turkish Foundation of Advertising for employees of digital agencies throughout Europe in cooperation with IAB Europe and Istanbul 2010 European Capital of Culture organization.

The brief was short, applicants were required to come up with an interactive campaign idea for ‘the young Europe” target group. “It is up to the applicant whatever, whoever to choose. Yet a strong, coherent backbone strategy is a must and a number of creative applications are required.”

Our group, Gergely Nyikos designer and me came up with a strategy based on my experience of a colorful, vivid and big city showing a variety of faces to the tourists. Although I’ve been living in Istanbul for a few months I don’t think I can describe the city. So we chose the group of people who can do it the best instead of us.

The application could only contain 300 words of strategy, 300 words of communication ideas and 3 small artwork images. We have to carefully select each and every words as the 300 words limit was very low. Also we had to play a bit with the emphases in the text as no text formatting was allowed.

I got approval for posting the strategy document only, so here you go:

Start a MOVEMENT: INSPIRE people to SHARE how they love Istanbul!

INSIGHT
- Travelers’ needs: Accommodation, food, sightseeing, shopping and souvenirs. False. They need something that TOUCHES them.
- Goals: See all museums and monuments. False. YOUNG EUROPEANS like to go beyond basic sightseeing. Visit places frequented by LOCALS. Find ‘HIDDEN’ treasures. MEET local people.
- Istanbul’s strengths: Istanbul is the city of all senses. Colourful. Vivid. Living. Multicultural. BIG. Friendly and helpful people. Neighborhoods have different “tastes”. TRUE.

Istanbul is too colorful to describe with few ads.

GOAL
Create desire: visit Istanbul!

STRATEGY
Add something personal and local. Show the many FACES of REAL city LIFE.
Website, rich content created in Istanbul. By young, Istanbul people: college, university students (and artists).
Content: LOT OF ‘amateur’ VIDEOS. 2 minute ‘clips’, multi-language. Personal VIDEO CLIPS created by the STUDENTS (who reacted to the CALL) in different LANGUAGES they learn.

TASK
Show something local you like and talk about it. (Favorite band, bar or restaurant. Shop or building or city panorama. Interview favourite actor, artist. Sporting, partying, shopping, anything would apply that you love.) All in your living neighborhood or near university (so all different neighborhoods be represented). Shoot a maximum 5” clip.

Clips selected, edited by a professional team to 2 minute (add music, logo, ending screen, tagging). Music and cut scenes come from ARTISTS volunteering in the program. ADVERTISING PLATFORM is created.
Clips uploaded: YouTube, website & seeded. COMMUNITY is created.

CALLING STUDENTS
Students from ALL universities apply via local university AMBASSADOR PROGRAM (similar: Nokia’s and Red Bull’s). Ambassadors: communications, lending semi-professional cameras, basic education on camera usage & clip making.

Students’ motivation:
- _Eager_ to do something for Istanbul, Turkey.
- International contacts. Students become local ‘scouts’, can be personally contacted.
- Competition, prizes.

In competition we also came up with a bonus idea:

To Istanbul visitors: VIDEO GUESTBOOK KIOSK on Istanbul airport.
Passengers leaving the country can tell a video saygoodbye to Istanbul in an automatic kiosk.
Funniest/craziest message wins the price of the flight ticket.
Videos automatically edited and uploaded to YouTube & website.
Flyers to hand out at the international departure entry doors about the kiosk.

And here’s the prize we won:

Please note that this document is just a part of the application. Together with the things described in the communication strategy document it would provide the city with a long term communication platform and sustained worldwide attention.

We were glad to participate in the competition, and all I can say the city tours and the final ceremony was awesome.

Fresh Facebook stats - Turkey

Thu, 24 Jun 2010 21:31:57 +0300

Now it’s time to refresh the stats I posted nearly 5 months ago as there are some changes: e.g. Turkey is no longer the third largest country on Facebook. Indonesia took over in the recent couple of months although Turkey added more than 4 million users to this count in the period and now having 22 500 000 Facebook users.

This means 24% of growth within 5 months which clearly shows not just the power of Facebook but the opportunity in Turkey’s relatively low internet penetration also.

This growth was mostly produced by the 30-49 age group which should not be a surprise as Facebook’s 75+% penetration among internet users is very close to 100% among the young. I put the growth rates onto the below chart by red, based on the data from my previous blogpost.

And again, the Turkish sport clubs. They did a good job in expanding their fan base.Both Fenerbahce and Besiktas outgrew the Facebook population in the country, Fenerbahce actually added more than 400 000 fans from one day to another after a famous football match.

And of course I have to mention that Galatasaray is actually the largest sport club on Facebook, also listed among the top 50 largest Facebook pages. (The official sports club pages has less users than this as on the above chart I added up all fan clubs).

Turkey bans Google - myths and truths

Thu, 10 Jun 2010 00:47:41 +0300

The Google ban in Turkey was headlined on most of the IT portals at the end of last week, the beginning of this week. Some of the news sources, like Mashable tried to crowdsource the info collecting process on their FB page (“If you’re from Turkey, we’d appreciate any updates on the blockade!”), so they could build up their coverage on that quite well.

Some other portals said that even Google, the search engine is being blocked or they added Facebook to the list.

Being in Turkey this is what I experienced (although some other users could have different experiences):

First Google Translate went off, then on the next day, some other Google services like Docs became unavailable. (Google, Gmail, Google AdWords, Facebook worked well, but Google Analytics became much slower.)

Meanwhile, YouTube - which has been banned since 2007 also worked well (as it was a pure DNS based ban, changing to Google DNS or Open DNS easily solved the problem). YouTube had around 6 million unique visitors per month from Turkey, seemingly all of them used the same trick as even the prime minister told so in front of the public.

Techcrunch came out with the info that the ban was not becuse of a decision made by one of the the government offices, but as they put the original YouTube ban into a higher ‘level’ (onto IP ranges), they just eventually stopped a lot of other services. We found the original news from the goverment officials saying almost nothing.

Access to services started to become normal again today. And yes, YouTube is absolutely off, only a VPN could help.

Actually, I have to add that life without Google Translate or having access to my Google Docs is hard. Not unimaginably, but I had to spend a lot of time unnecessarily on translations or recollecting my documents. I was definitely not happy with that.

Meanwhile, some of my Clients had to remove Google Analytics codes from their slowed-down websites and some of them removed Google AdWords tracking codes as well just in case. It’s definitely a loss to the whole industry, not just Google.

Another myth that the ban is because of Islam. No, it has nothing to do with that. The original YouTube ban was put up because of some ‘insulting’ videos of Atatürk, the founder of the republic. (As I heard the video played disco music toghether with Atatürk and that’s an insult as of the Turkish government ‘standards’.

Another rumour says that after Google offered to limit access to certain ‘offending’ videos in Turkey in return to forget the ban. The goverment said that it is not enough, because as Turkish people go abroad and there they still can be insulted with these. So, they said Google should delete them, no compromise.

Some other sources say that this dramatic change in lifting the bans of YouTube is because of the Gaza Flottilla videos. “On the 3rd of June three most viewed videos on YouTube contained documented footage of the aforementioned events, invalidating Turkish allegations. On the 4th of June, YouTube site was banned in Turkey.”

Hmm. I can imagine that.

Turkish Digital Agencies are protesting against the bans by putting a protesting text onto their homepages. Another 6500 websites are now being banned in Turkey, this site has a huge list of them, the reasons vary.

The real battle between Google and Facebook is for what?

Mon, 31 May 2010 17:36:32 +0300

Some parts of a recent blogpost from a Hungarian online marketing blogger created buzz on the web and another Hungarian search aficionado just reacted on this saying ‘We can’t make our living from just Liking’. The ominous saying from the first blogger just looked like this: ‘Within 1-2 years Facebook will launch an advertising system for the content network (…) that will place ads based on the clicking and Like-ing behaviour of the users.’

Take a look on the trends: Google launched Google Buzz. YouTube has recently launched a bunch of new social features. Google’s just recently launched its remarketing tool. What are these news about? They mean two focus points: connecting users with each other to create a personal network and a focus on content network to enhance effectiveness of campaigns.

But take an even broader view. Facebook has a unique network of personal relationships and profiles, but they don’t have an outlook on the web. Google has a unique database about the whole web but lacking user profiles and personal networks. (The first problem will be solved by a recently announced Like button, for the latter I mentioned some of Google’s efforts above.)

And let’s talk about a third phenomena, which is the world of statistitians and mathematicians: the recommendation systems. Recommendation systems are based on huge contextual databases of people - and their taste, choices and behaviour. For example if we have enough data about a movie: what kind of people like it and what kind of people dislike, a recommendation system can predict whether a person who has never seen that will like it or not.

We need to have two kinds of data for this: extended personal profiles with a lot of like (and dislike) data, plus a wide range of data about similar products.

You got the point? Google is very good in the second one and tries to collect the first, Facebook has the first and tries to collect the second.

But why? Of course for the advertising dollars. The company that succeeds in building up the largest database of these two can predict advertising effectiveness better. Meaning it can spend advertising dollars more effectively. As well meaning that the company will rule the online advertising world.

TechCrunch’s just recently published a guest article analyzing the same thing, the potential in Facebook as a gatekeeper of the world’s identity, even saying Google can become he next Microsoft. :-)

This battle has not started yet, as Facebook is yet having an advertising network product. But this will come.

If this theory is true it makes easy to predict some future happenings:

Facebook will need a dislike button. I’m pretty sure they will have it soon.
Facebook will buy a web analytics company. If I were them, I would go for Adobe’s Omniture.
Facebook will need an adserving company as well. Hard to break DoubleClick’s domination but why not try?

And additionally - or being absolutely inline with the above idea - Facebook’s now have a real chance to move the whole web (and no question to that Open Graph will add a measurably large chunk of the web) towards a more semantic world.

Finally here’s a nice video from Kate Ray - with a lot of interviews from the web’s masterminds about being semantic.

Web 3.0 from Kate Ray on Vimeo.

What is the real difference between Facebook Pages and Groups?

Sun, 16 May 2010 19:27:36 +0300

This is the most common question I face nowadays from brand managers, small business owners and startup companies.

Seemingly they already decided to create some kind of a Facebook presence - or even they started up with something. The next question is usually the above, sometimes followed by another one: “I started up with a personal profile, what should I do with the hundreds/thousands of friends I’ve already gathered?” And of course they are afraid of the most common answer: “shut it down”. Actually I heard rumours that Facebook is shutting down those profiles without notice - so the best way is to create a group or page for the brand, and notify friends about it urging them to change before it gets closed down. Unfortunately there’s no better option.

Back to the question firstly I would recommend to read this page from AllFacebook.com. The below chart from that page shows most of the differences quite clearly:

As for an international brand it’s no question that targeting languages (targeted stream posts) is an inevitable option - or they still can target users with two different language Pages, like Hungarian Airlines Malév does here and here. Or they can target all members with two language posts like Fania Music does.

For a lot of marketers the question is more simple: Which of the following features do you need more? Direct e-mail messages sent to the members’ e-mail address or the possibility of creating custom html pages?

For example a business or brand that manages a series of events should clearly stay with Groups (like I did with my local salsa addict brand). Of course the 5 000 members cap on this can be a serious issue for a lot of brands.

Big brands that spend money on building awareness surely can exploit the opportunities of creating custom HTML pages via Facebook’s FBML application. Of course there is another option open for big brands with Facebook pages to use e-mal addresses: they can create an application and ask for special permissions from the users. I still can’t see many brands doing this.

My personal note on this issue is that although I understand Facebook restricting Pages from using e-mail addresses - they are overwhelmed fighting spammers and they have a lot of privacy issues - but I still can’t see too many reasons for the feature differentiation of Pages and Groups. Hope that we’ll see barriers between these two basic Facebook features - Groups and Pages - disappearing or at least decreasing.

Does Facebook change the way we use the web?

Sat, 24 Apr 2010 17:51:00 +0300

After watching Facebook’s F8 conference keynote live on the web on April 21, I had the very same thoughts that I had when there were a lot of discussions about the Google IPO back in 2004.
Those thouths included two basic things: one was that this is the time to invest; the second that this is the time to apply for a job there.
(Unfortunately I did neither in 2004. :-))

New concept is coming
Facebook’s concept is nothing new. Instead of relying on shady algorhythms to classify the web it uses people to do that. With a simple click of a button.
Opening up the “Like” phenomena to the whole web offers a win-win business for website owners, users and Facebook. This win-win concept is the one Google did a very good job in - and life proved that they were right.
This gives FB a new horizon on the web.

Let’s have a quick overview on the new items that were just announced:

Like button for the whole web - any website can put the Facebook like button on - Likes will appear in the Facebook news feed.
Activity feed - this plugin will show the users their friends activity on that particular website.
Comments plugin - this will enable users to comment on any item on the website.
Live stream - show user feedbacks during a live event.
Recommendations plugin - It provides the users with recommended content on the given website.
Facebook Connect - this is not new, although it’s a powerful tool to replace subscription and login mechanisms on the website. Not to mention that Facebook’s just get rid of the rule that user data cannot be stored locally for more than 24 hours.

What’s missing from the FB concept?
I see a major logical fault in this concept. Having hundreds of Facebook friends I really miss the opportunity to classify them. On my news feed there’s been a recommendation system working in behind for a long time. If I don’t react to news coming from one of my friend, his news will tend to disappear from my feed.
This could be okay. But people keep on posting in very different topics - the system may miss some interesting pieces of news from that specific friend.
And on the other hand as Facebook’s Like concept now being open for the whole web news feed will contain much more items per day.
So I really appreciate Facebook’s efforts to build the semantic web and bring that closer by personal recommendations - but I think that not just the web needs classifying, but also friends.

Another big missing thing is being local. Google’s been working on local tools for a while and the mobile era of local web services are just coming. Wondering who will be the one buying Foursquare. :-)

End of the Google era?
The war has not ended yet, but quite a few battles has been won by FB, and some were lost by Google recently.
But definitely this is the time when we’ll see the war intensifying. IMHO the world would be a much better place if they could cooperate - Google seach on Facebook, Google’s ranking system supporting FB recommendations, Google Maps integrated, Google Places to be liked on FB…
This could be a very long list but makes no sense to continue as it won’t happen.
Google surely will grow as fast as it did before - AdWords can’t be beaten. And search can’t be beaten either.
On the contrary. As Facebook’s Open Graph concept will spread Google will have to somehow incorporate it to the search results.
The new “tagging” raid is on the way. The web will be more semantic than ever before - and search results will be more accurate using that.
Let’s see Google’s reaction.

Will this kill Twitter?
In some conversations with professionals I just occasionally said that opening up the Like buttons and the Activity Feed plugin toghether will kill Twitter.
Seeing the weird reactions made me think twice. And now I truly believe this will happen. Twitter is a channel that failed to integrate with the web. Outside the US it is still a playtool of the geeks. Twitter is about to fail on monetization as well. And money is needed gain more attention, more users.

Poor FriendFeed
Although I’m quite a beginner user of FF but I think that their concept has just been incorporated into Facebook. They will also have to come up with something very new and powerful to survive.

So I believe that recent announcements from Facebook will trigger a flow of new product features from competitors like Google or Twitter. Time will tell how much this concept will change the way we use the web but I think that this is a beginning of something powerful.
Facebook wants to sit on top of the web and they do it quite well. If they can go local - nothing will stop them.

PPC networks in Turkey - a comparative overview

Tue, 09 Mar 2010 18:32:00 +0200

I realized that it’s very hard in one hand to provide the readers with useful, hands-on information and on the other hand to remain neutral and even-handed. That’s why there are some cells in the table with questionnable/disputable content and I think that’s normal. Question that, dispute that.

The firsts are the ‘usual’ ones, Google AdWords Content Network and Facebook. I was really surprised seeing that a number of advertisers still yet go beyond Google Search hence the low CPCs on these networks. There are two-three possible reasons for that:

Google Search can eat up vast amount of money thanks to the size of the market compared to the maturity of the market and digital budgets
Surprisingly a lot of advertisers are “afraid” of the uncontrolled environment. This is a serious issue here and that’s why I put the row ‘negative news filters’ into the table.
The number of performance based digital campaigns is still really-really low. This market has a space to develop. :-)

On the other part, there are strong PPC networks on the market: LinkZ and AdKlik. They all have an impressive local and international Client portfolio thanks to the strong local sales teams they have. Most of their inventory is used for campaigns that agencies put up there to optimize reach and CPM of their campaigns.

Actually, the non-existent bidding, limited targeting and optimization options of these networks hardly allow the advertisers to go for real performance based campaign optimisation.

On the other hand, the Turkey PPC market is live and kicking, provides the advertisers with a very good alternative or complementary options to Google AdWords Content Network. All of them has a quality portfolio of websites - non of them have opened up their systems to small inventories yet.

And as I mentioned in the previous part - check out their perfect response for the local needs: all of them have a comprehensive negative content filtering algorhythm. And by comprehensive I mean better than the one that Google has.

I have to mention, that the latter two products belong to the same media owner, I had to separate them because their products are quite different.

*Addition: I forgot to add one more important line to all PPC providers whether keyword stats are available. This is important as if keyword stats are available the campaign can be optimised on a keyword level. That’s obvious. Actually the results are more than interesting. The only provider where keyword stats are not available is Facebook. (Have to admit that on a respondent level of the campaign stats some keywords can be seen but I’d rather not call this a keyword report. They don’t do that either.)

4 videos every media worker should watch

Thu, 11 Feb 2010 12:05:04 +0200

Of course you’ve seen the Social Media Revolution video, this is the first one in here, cannot be missed. Actually small the chance that you are not part of that 1,5 million who watched that but in that case, here it is.

The next one was watched by “only” one million in the recent months but I have to tell the truth this one was new for me as well. The way print circulation goes downwards is astonishing…

The last two are from TED. Sure these are as well were watched by millions.

What I really admire in Seth Godin’s speeches that watching him I always have the feeling that okay, this is absolutely nothing new to me. All things he’s talking about I already know, I am familiar with them. But at the end there’s someting that sticks to my mind as the context he changes of things. This time it’s all about communities he call Tribes (new buzzword we have) but the most sticky in this speech for me was what he said about change.

Clay Shirky, consultant, professor and author discusses the change we see in journalism and mass media. I like how he says that “…this is how you reached people. All of that is over.”

And finally there’s the quarterly new media trend presentation from isobar’s Dan Calladine. E.g. did you know that broadband internet access became a legal right in Finland recently? :-)

Next Generation Media Quarterly - January 2010 View more presentations from Dan Calladine.

Facebook stats - Turkey

Thu, 04 Feb 2010 17:57:00 +0200

As I mentioned in my previous post I am really amazed how social Turkey is. Actually that’s true IRL as well and we can see only it’s online counterpart on Facebook. This behaviour was the one that rocketed Turkey to the third position worldwide as of the number of Facebook users.

The big picture is changing rapidly - I would have never thought that the fourth country will be Indonesia - but here’s the current status of the TOP 10 countries on Facebook as of 3th February 2010. (Facebook penetration figures in the brackets represent the number Facebook users per the number of internet users, internet users data come from here.)

Turkey has been holding its third position for months now and it seems that it can keep it, as the Facebook population here reached the critical level when everybody feels the disadvantages not being there.

Let’s see the details. Turkey is a country with a median age of 28,8, and we see this suprisingly young figure quite well represented on Facebook as well.

Interesting that Facebook is more a male medium here as almost twice as many men (61,9%) use Facebook than women (32,4%). (The discrepancy is the number of users not having their age filled out.)

Not so surprising to see the breakdown by cities with 7 million Facebook users just in Istanbul. Actually it’s nice to see that other cities are doing that quite well too.

And some interesting stuff researching some keywords on Facebook: the size of football Fan Clubs:

They are not in one Facebook club but each sport club has its own official Facebook page which usually is as well the biggest.

As far as I know the largest Facebook page is Türk Bayragi (sorry for the missing Turkish chars) which stands for Turkish Flag and has 2 747 000 users as of now. They were the fastest growing fan page in the world for a few weeks in November last year.

Turkey: a place of opportunities in digital?

Tue, 08 Dec 2009 14:00:00 +0200

Experiencing the Turkish online market on a daily basis and having an experience with Central-Eastern European ones I always a little bit confused what’s going on here.
In the one hand I see lots of creativity and world-class online marketing solutions but in the other, sometimes it looks like there’s still a very-very long way to go for online businesses to grow and reach the service levels of the more developed markets. So confused I started looking for benchmarks and here’s what I got.

I calculated the per capita (per internet user) online spendings (search, ppc included), and put that up on a chart with per capita (total population) GDPs.
I know statisticians would run crying seeing these together but it just supports my experience with the markets.
I used 2008 local IAB spendings versus Internet World Stats combined with CIA World Factbook GDP per capita data. Wherever I could.

I took a look at countries that somehow do not fit in:

Norway. The highest per capita GDP, but only the third place in per capita spendings. This I call an opportunity for marketers.
Italy and Spain. Somehow a bit of a surprise for me. Although they are in the upper region of per capita GDP, the online spendings do not show the same. Opportunity as well.
Hungary (my home country) is way ahead a lot of countries as of the maturity of the online market (see the low GDP/capita vs. relatively high online spend/capita)
Russia, Turkey, Romania. All at the end of the chart as of per capita spendings and in per capita GDP as well. Please not that there are huge differences comparing just these three to each other.

I thought that I will get a clear evidedence confirming my assumption that there is a big opportunity in Turkey for marketers. Actually, the results are not so clear on that but I’m still confident that at the EUR 6,5 spent online / internet user rate Turkey is really a good place for investments in digital.

Not to mention that Turkey is the third largest country on Facebook.

Some other ideas that I had during posting this:
- Crisis leveled differences: growth of online ad spend almost stopped in the States, slowed down dramatically in UK and Germany, but it still grew double digit in Eastern Europe and I believe it grew a lot here in Turkey.

- No big jump. Although the less developed countries grow on a higher rate (as of spendings), they usually do not jump big. Reaching the EUR 20 current Hungary level per capita spending would mean tripling the current spendings in Turkey. So that will last for at least 2-3 years, surely.