Články o online podnikání - Webtrh.cz

Muzete mit temer vse jen to stoji dost prace

Fri, 10 Feb 2012 21:50:53 +0000

Nejhorsi je, ze skutecne vsechno co si prejete se muze splnit. Ma to vsak hacek. Nic neni nikdy zadarmo. Takze nefunguje: “Chci byt bohaty, ale bez namahy”. Vzdy to stoji praci. Cim tezsi, narocnejsi cil, tim muzete pocitat s tim, ze to bude vice prace nebo odrikani. A v tom je to zklamani. Hodne lidi [...]

Martin Snížek na WAWu: Jak zařídit, aby vaše A/B testy přinášely výsledky

Fri, 10 Feb 2012 07:33:19 +0000

Druhým spíkrem o A/B testování na nadcházejícím Web Analytics Wednesday bude: Martin Snížek - spoluzakladatel Optimics. Martin je garantem optimalizace výkonnosti webu v Optimics, pracoval již pro desítky velkých klientů z finanční i retail sféry. Od roku 2004 se věnuje uživatelskému testování a UX, později rozšířil svůj pohled díky webové analytice a zákaznickému výzkumu do komplexního pochopení [...]

PS. Zapasite s webovou analytikou? Chcete lepe merit vase aktivity v online marketingu? Poslete mi svuj dotaz, rad vam poradim.
Pokud chcete sledovat komentare k tomuto clanku nebo pridat svuj vlastni, kliknete na Martin Snížek na WAWu: Jak zařídit, aby vaše A/B testy přinášely výsledky.

Související články:

Why I Don’t Host My Own Blog Anymore

Thu, 09 Feb 2012 17:18:43 +0000

I moved my blog over to WPEngine recently. Why? Read on. I started blogging about 375,000 words ago (about three full-length novels… crikey). At first, I was on a subdomain of WordPress.com, mostly because a) it was free and b) I had no intention of ever writing anything more significant than a few observations I [...]

Two Giant Steps Forward For Entrepreneurs

Thu, 09 Feb 2012 16:19:59 +0000

While entrepreneurship is in the news fairly regularly, I seldom make news myself. Today, however there are two important updates for entrepreneurs everywhere. Let me be brief… The “Startup Owner’s Manual” goes On Press Tuesday 2/14 Two years in the making and literally ten years in development, I’m proud to announce that my new book, [...]

Mžik

Thu, 09 Feb 2012 07:50:18 +0000

Zrovna čtu knihu Mžik: Jak myslet bez přemýšlení od Malcolma Gladwella (Blink: The power of thinking...

The wow effect

Gabriel Weinberg — Tue, 07 Feb 2012 12:06:23 +0000

There is a startup proverb that says you have to be 10x better than your competition to get people to switch. A related one is that you should concentrate on being a painkiller as opposed to a vitamin.

Certainly if you're 10x better or are a real painkiller, your job of getting people to switch will be easier. Nevertheless, I've come to think there is an easier way to get people to really try out your product. I call it the wow effect.

The wow effect is when you show someone something and they truly go "wow!" or have a substantially similar response. It is genuine amazement because you've gone above and beyond in some way. You've done something new or cool or that someone really cares about regardless if it is core to the product or related to a pain point.

At DuckDuckGo our privacy policy and various goodies often produce the wow effect. Some other ones I've personally experienced recently are using foursquare + AMEX for the first time, Hipster images on Android, and seeing the resolution of the Path app on Android (though they just unfortunately killed it in their latest update).

A good proxy for the wow effect: is it something you want to immediately share on Twitter or with friends? If so, then you're on to something.

Getting to 10x better or being a painkiller maybe difficult or impossible in some markets. Yet generating the wow effect in a variety of ways and situations and for different user demographics is something you can shoot for and achieve more easily.

However, it's just a hook. It's a signal that you're onto something awesome. It gets people to try you out and then you have to do other things to keep people coming back.

Představujeme spíkry na WAW: Josef Hos, AVAST Software

Tue, 07 Feb 2012 08:59:05 +0000

Únorové setkání Web Analytics Wednesday na téma A/B testování se nezadržitelně blíží a my vám nyní můžeme představit prvního přednášejícího: Josef Hos - eCommerce Director AVAST Software. Josef má více jak 10 let zkušeností v oblasti IT a online trhu se silným analytickým zaměřením na design a optimalizaci webu pro zvýšení jeho prodejní úspěšnosti. Od roku 2009 [...]

PS. Zapasite s webovou analytikou? Chcete lepe merit vase aktivity v online marketingu? Poslete mi svuj dotaz, rad vam poradim.
Pokud chcete sledovat komentare k tomuto clanku nebo pridat svuj vlastni, kliknete na Představujeme spíkry na WAW: Josef Hos, AVAST Software.

Související články:

The Future of Networking

Mon, 06 Feb 2012 15:54:41 +0000

Major technology platforms tend to last about 25 to 30 years. This gives them time to gather sufficient developer momentum, enable a set of transformational ideas, build those ideas, and form a large industry around it. The platforms then sustain for an additional 5-10 years due to inertia and lock-in. Finally, when the shortcomings of…

A Peek Inside the Erlang Compiler

James Hague — Mon, 06 Feb 2012 06:00:00 +0000

Erlang is a complex system, and I can't do its inner workings justice in a short article, but I wanted to give some insight into what goes on when a module is compiled and loaded. As with most compilers, the first step is to convert the textual source to an abstract syntax tree, but that's unremarkable. What is interesting is that the code goes through three major representations, and you can look at each of them. Erlang is unique among functional languages in its casual scope rules. You introduce variables as you go, without fanfare, and there's no creeping indentation caused by explicit scopes. Behind the scenes that's too quirky, so the syntax tree is converted into Core Erlang. Core Erlang looks a lot like Haskell or ML with all variables carefully referenced in "let" statements. You can see the Core Erlang representation of a module with this command from the shell: c(example, to_core). The human-readable Core Erlang for the example module is written to example.core. The next big transformation is from Core Erlang to code for the register-based BEAM virtual machine. BEAM is poorly documented, but it's a lot like the Warren Abstract Machine developed for Prolog (but without the need for backtracking). BEAM isn't terribly hard to figure out if you write short modules and examine them with: c(example, 'S'). The disassembled BEAM code for the example module is written to example.S. The key to understanding BEAM is that there are two sets of registers: one for passing parameters ("x" registers) and one for use as locals within functions ("y" registers). Virtual BEAM code is the final output of the compiler, but it's still not what gets executed by the system. If you look at the source for the Erlang runtime, you'll see that beam_load.c is over six thousand lines of code. Six thousand lines to load a module? That's because the beam loader is doing more than its name lets on. There's an optimization pass on the virtual machine instructions, specializing some for certain situations and combining others into superinstructions. To check if a value is a tuple of three elements is accomplished with a pair of BEAM operations: is_tuple and is_arity. The BEAM loader turns these into one superinstruction: is_tuple_of_arity. You can see this condensed representation of BEAM code with: erts_debug:df(example). The disassembled code is written to example.dis. (Note that the module must be loaded, so compile it before giving the above command.) The loader also turns the BEAM bytecode into threaded code: a list of addresses that get jumped to in sequence. There's no "Now what do I do with this opcode?" step, just fetch and jump, fetch and jump. If you want to to know more about threaded code, look to the Forth world. Threaded code takes advantage of the labels as values extension of gcc. If you build the BEAM emulator with another compiler like Visual C++, it falls back on using a giant switch statement for instruction dispatch and there's a significant performance hit. (If you liked this, you might enjoy A Ramble Through Erlang IO Lists.)

Four Levels of Idea Theft

James Hague — Sun, 05 Feb 2012 06:00:00 +0000

Imagine you've just seen a tremendously exciting piece of software--a mobile app, a web app, a game--and your immediate reaction is "Why didn't I think of that?!" With your mind full of new possibilities, you start on a project, a project enabled by exposure to the exciting software. What happens next is up to you. How far do you let your newfound motivation take you? Borrowing specific features. You like the way the controls work. The sign-in process. Something specific. Sliminess Factor: None. This is how progress happens. General inspiration. If web-based photo sharing had never occurred to you, and then you saw Flickr, that opens the door for thinking about the entire problem space. Some of those options may be Flickr-ish, some aren't. Sliminess Factor: Low. It's a common reaction to be excited and inspired by something new, and it inevitably affects your thinking. Using the existing product as a template. Now you're not simply thinking about photo sharing, but having groups and contacts and favorites and tags and daily rankings. You're not still writing a full-on Flickr clone--there are lots of things to be changed for the better--but it's pretty clear what your model is. Sliminess Factor: Medium. While there's nothing illegal going on, you won't be able to dodge the comparisons, and you'll look silly if you get defensive. Any claims of innovation or original thinking will be dismissed as marketing-speak. Wholesale borrowing of the design. All pretense of anything other than recreating an existing product have gone out the window. Your photo sharing site is called "Phlickr" and uses the same page layouts as the original. Sliminess Factor: High. This is the only level that legitimately deserves to be called theft. (If you liked this, you might enjoy Accidental Innovation.)

Focus on The Business Model

Sun, 05 Feb 2012 00:43:23 +0000

Google's Take on Search Plus Your World

A few weeks ago Google announced the launch of Search Plus Your World, which deeply integrates social sites (especially Google+) into the Google search experience to make it more personalized.

While Google claimed that the socialization was rather broad-based, the lack of inclusion of Facebook & Twitter along with the excessive promotion of Google+ raised eyebrows. While the launch was claimed to be social for personalizing results, the Google+ promotions appeared on queries where they were clearly not the most relevant result even when users are not logged into a Google account.

Google+ Over-promotion

A couple weeks ago when Google announced Google Search Plus Your World competitors collectively complained about Google over-promoting their own affiliated websites.

Twitter was perhaps the loudest complainer, highlighting how Google basically eats all the above-the-fold real estate with self promotion on this @WWE search.

It is no surprise that folks like Ben Edelman, Scott Cleland & Fair Search chimed in with complaints, as this is just a continuation of Google's path. But the complaints came from a far wider cast of characters on this move: the mainstream press like CNN, free market evangalists like the Economist, Google worshipers indoctrinated in their culture who wrote a book on Google & even ex-Googlers now call into question Google's transparently self serving nature:

I think Google as an organization has moved on; they’re focussed now on market position, not making the world better. Which makes me sad.

Google is too powerful, too arrogant, too entrenched to be worth our love. Let them defend themselves, I'd rather devote my emotional energy to the upstarts and startups. They deserve our passion.

The FTC's Google antitrust probe is to expand to include a review of Google+ integration in the search results.

Facebook & Twitter launched a don't be evil plugin named Focus On The User, which replaces Google+ promotion with promotion of profiles from Facebook& Twitter.

For the top tier broad social networks framing the idea of integrating promotion of their networks directly in the search results is a natural & desirable conclusion, but is that just a convenient answer to the wrong question?

Whether Google ranks any particular organic result above the corresponding Bing ranking in Google's now below-the-fold organic results is a bit irrelevant when the above the fold results are almost entirely Google.com. But is the core problem that we are under-representing social media in the search results? According to Compete.com, Facebook & YouTube combine to capture about 16% of all downstream Google clicks. Do we really need to increase that number until the web has a total of 5 websites on it? What benefit do we get out of a web that is just a couple big walled gardens?
If Facebook is already getting something like 20% of US pageviews & users are still looking for information elsewhere, doesn't that indicate that they probably desire something else? Absolutely Facebook should rank for Facebook navigational queries, but given all their notes spam, I don't like seeing them in the search results much more than seeing a site like eHow.
The he said / she said data deals are also highly irrelevant. What is really needed is further context. Before Google inserted Google+ in their search results the Google+ social network was far less successful than MySpace (which recently sold for only $35 million). If social media is added as an annotation to other 3rd party listings then I think that has the opportunity to add valuable context, but where a thin "me too" styled social media post replaces the publisher content it lowers the utility of the search results & wastes searcher's time. Further, when those social media results are little more than human-powered content scrapers it also destroys the business models of legitimate online publishers.

Over-promotion vs "Search Spam"

At any point Google can promote one of their new verticals in a prominent location in the search results & if they are anywhere near as good as the market leader eventually they can beat them out of nothing more than the combination of superior search placement, monopoly search marketshare, account bundling & user laziness. What's more, they can make paid products free and/or partner with competitors 2 through x in an attempt to destroy the business model of anyone they couldn't acquire (talk to Groupon).

Amit Singhal is obviously a brilliant guy, but I thought some of the answers he gave during a recent interview by Danny Sullivan were quite evasive & perhaps a bit inauthentic. In particular, ...

"The overall takeaway that I have in my mind is that people are judging a product and an overall direction that we have in the first two weeks of a launch, where we are producing a product for the long term." If the product wasn't ready for prime time you were not required to mix it directly into the organic search results right off the bat. It could have been placed at the bottom of the search results, like the "Ask on Google" links were. Bing has been working on social search for 18 months & describes their moves as "being very conservative."
"The user feedback we have been getting has been almost the other side of the reaction we’ve seen in the blogosphere." Of course publishers who see their content getting scraped & see the scraped copy outranking the original have a financial incentive to care about a free & automated scraper site displacing their work. They don't get those pageviews, they don't get that referrer data, and they don't get those ad impressions. Google's PR team is anything but impressed when another company dares do that to Google.
"The users who have seen this in the wild are liking it, and our initial data analysis is showing the same." Much like the Google Webmaster Tools shows that pages with a +1 in the search results get a higher CTR, this Google+ social stuff also suffers from the same type of sampling bias & giving the listings a larger and more graphical stand out further help them pull in much more clicks. Any form of visual highlighting & listing differentiation can lift CTR. I might be likely to click on some of my own results more, but when I do so you might just be grabbing a slice of navigational searches I was going to do anyway where I was looking for something else I posted on Google+ or my Google+ account or the account of a friend & so on. Further, aggregate data hides many data points that are counter to the general trend. I have seen instances of branded searches where the #1 organic site was getting a CTR above 70% (it even had organic sitelinks, further indicating it was a navigational search) and for such a search in some cases there were 2 Adwords ads above the organic results & then the Google+ page for a brand outranked the associated brand in the SERPs for those who followed it! That is a terrible user experience, particularly since the + page hasn't even had any activity for months.
"Every time a real user is getting those results, they really are delighted. Given how personal this product is, you can only judge it based on personal experiences or by aggregate numbers you can observe through click-through." First, publishers are not fake users. Secondly, as mentioned above, there is a sampling bias & the + listings stand out with larger & more graphical listings. If they didn't get a higher CTR that would mean they were *really* irrelevant.
"out of the gate, whereas we had limited users to train this system with, I’m actually very happy with the outcome of the personal results." They could have been placed at the bottom of the search results or off to the side or some such until there was greater confidence in the training set.
"People are coming to a conclusion about the product today, within the first two weeks, and they’re not fully seeing the potential where we can build this product around real identities and real relationships." If a publisher promotes a site to the top of the search results & then says something like 'we will improve quality later' they are branded as spammers. In the past Google has justified penalizing a site based on its old content that no longer exists on the site. Investing in depth, quality & volume is a cycle. If others get prohibited from evolving through the cycles due to algorithms like Panda then it becomes quite hard to compete as a new start up when Google can just insert whatever it wants right near the top & then work on quality after the fact.
"We don’t think of this as a promotional unit now. This is a place that you would find people with real identities who would be interesting for your queries." If this is the case then why does it only promote Google+?
"We’re very open to incorporating information from other services, but that needs to be done on terms that wouldn’t change in a short period of time and make our products vanish." The problem is, if a company builds a reputation as a secretive one that clones the work of its partners & customers then people don't want to do open-ended transparent relationships. Naive folks might need to see the blood and tears 3 or 4 times to pick up on the trend, but even the slowest of the slow notice it after a dozen such moves.
"I’m just very wary of building a product where the terms can be changed." Considering Google's lack of transparency & self-promotional bias on the social networking front, would you be fully transparent and open with Google? If so, then aren't the search algorithms complex enough that it would make sense to make those transparent as well? How can you ask other social networks to increase transparency at the same time Google is locking down their search data on claims of protecting user privacy?
"It’s not just about content. It’s about identity, and when you start talking about these things and what it takes to build this, the data needed is much more than we can publicly crawl." This is where being trustworthy is so crucial. Past interactions with Yelp, TripAdvisor & Groupon likely make future potential partners more risk adverse & cautious. Outrageous "accidents" like those that happened with Mocality & Open Street Map from playing fast and loose further erode credibility. And even when Google hosts the media & has full access to user data they still rank inferior stuff sometimes (like the recent Santorum YouTube cartoon fiasco), even on widely searched core/head keywords.

The big issue is that if people feel the game is rigged they won't have much incentive to share on Google+. I largely only share stuff that is irrelevant to tangentially relevant to our business interests & won't share stuff that is directly relevant, because I don't want to be forced to compete against an inferior version of my own work when the deck is stacked so the inferior version wins simply because it is hosted on Google.

As we move into the information age a lot of physical stores are shutting down. Borders went bust last year. Sears announced the closure of many stores. And many of the people shopping in the physical stores that remain are using cell phones for price comparisons. Given Google's mobile OS share this is another area where they can build trust or burn it. A friend today mentioned how their online prices on Google Product search almost always show a lower price near the header than the lowest price available in the list - sometimes by a substantial margin.

Identity vs Anonymous Contractors

In the past we have mentioned that transparency is often a self-serving & hypocritical policy by those atop power systems who want to limit the power of those whom they aim to control.

When Google was caught promoting illegal drug ads there was no individual who took the blame for it. When the Mocality scraping & the Open Street Map vandalism issues happened, all that we were told was that Google "was mortified" and it was "a contractor." If people who did hit jobs could just place all the blame on "the contractor" then the world would be a pretty crappy place!

Eric Schmidt warned that "If you have something that you don't want anyone to know, maybe you shouldn't be doing it in the first place." That sage advice came from the same Eric Schmidt that blackballed cNet for positing personal information about him. Around the same time Eric offered the above quote, Google was engaged in secret & illegal backdoor deals with direct competitors to harm their own employees.

What happened to Google recruiters who dared to go against the illegal pact? They were fired on the hour:

"Can you get this stopped and let me know why this is happening?" Schmidt wrote.

Google's staffing director responded that the employee who contacted the Apple engineer "will be terminated within the hour."

When Google+ launched they demanded that you use your real name or don't use the product. They later claimed that you can use a nickname on your account as well, but there is a difference between a nickname and pseudonyms.

What is so outrageous about the claims for this need for real identities is that past studies have shown that pseudonymous comments are best & Bruce Schneier highlighted how we lose our individuality if we are under an ever-watchful eye:

Cardinal Richelieu understood the value of surveillance when he famously said, "If one would give me six lines written by the hand of the most honest man, I would find something in them to have him hanged." Watch someone long enough, and you'll find something to arrest -- or just blackmail -- with. Privacy is important because without it, surveillance information will be abused: to peep, to sell to marketers and to spy on political enemies -- whoever they happen to be at the time.

Privacy protects us from abuses by those in power, even if we're doing nothing wrong at the time of surveillance.

In many markets ads and content are blended in a way that is hard to distingush between them. Whenever Google wants to enter they can demand greater transparency to participate (and then use the standard formatted data from that transparency to create a meta-competitor in the market.)

Increasingly Google is placing more of their search data & their webmaster-related functions behind a registration wall. If you are rich & powerful they will sell you the data. If you are the wrong type of webmaster that aggregate data can be used in *exceptionally* personal ways.

User Privacy

Ahead of Google updating their privacy policy Google has directed a large portion of their ad budget toward ads about how they protect users online.

What better way to ensure user privacy than to allow them to register their accounts under psydonyms? The real name policy on Google+ was part of what made Google want to stop providing referrer data for logged in users who search on Google. This has had a knock on effect where other social sites are framing everything, requiring registration to read more of public user generated content & sending outbound traffic through redirects.

Google's new privacy policy allows them to blend your user data from one service into refining the experience (and ads) on another:

If you’re signed into Google, we can do things like suggest search queries – or tailor your search results – based on the interests you’ve expressed in Google+, Gmail, and YouTube. We’ll better understand which version of Pink or Jaguar you’re searching for and get you those results faster.

Google & Facebook's war (against) user privacy is catching media and governmental attention. Microsoft highlighted some of Google's issues in their "putting people first" ad campaign & the blowback has caused Google not only to publish PR-spin "get the facts" styled blog posts, but to launch yet another ad campaign.

EU regulators have asked Google to pause their privacy policy changes.

Bogus Testimonials & Social Payola

Is social media a cleaner signal than links? If search engines put the same weight on social media that they put on links it would get spammed to bits. It won't be long until a firm like Ad.ly offers sponsored Google+ posts.

Some have suggested that you won't be able to buy Google+ followers however Google already includes user pictures on AdWords ads (even when they desire not to be & even when they didn't endorse the product that Google suggests they endorsed). In due time I expect Google will indeed sell followers & other user interactions as ad units (just like Twitter & Facebook do).

Further, celebrities sell Tweets to advertisers. When they are hot their rates go up:

When Ad.ly introduced self-destructing Charlie Sheen to Twitter, he was paid about $50,000 per tweet. It was worth it. Sheen’s tweet for Internships.com generated 95,333 clicks in the first hour and 450,000 clicks in 48 hours, created a worldwide trending topic out of #tigerbloodintern, attracted 82,148 internship applications from 181 countries, and added 1 million additional visits to Internships.com.

Search engines might consider these to be clean signals if those same search engines were not busy buying the manipulation of said "relevancy" signals.

Attention is purchased to create demand. It isn't comfortable to put it this way, but we are trained to obey authority & to like what others like:

The average Facebook user has 130 friends, which equates with four degrees of separation to thousands of people, Mr. Fischer said. Metrics like that led him to believe that if Facebook could figure out a way to capitalize on "social endorsements," it would be like creating a word-of-mouth campaign that could reach millions of people simultaneously. Since the campaigns would come from a friend, they would theoretically be taken more seriously than, say, a TV commercial, he said.

On an individual basis reviews and ratings get faked everywhere. Even stodgy old slow-moving institutions like colleges game their ranking systems.

There recently was a question raised about how Google's rating systems skewed high on the underlying data. Surely Overstock (the same Overstock Google penalized earlier this year) wouldn't promote Google's trusted stores aggressively on their own site if it made their business appear worse than it actually is, thus a positive bias must be baked in to the system.

Entire categories of demand are created by those tied in with power cost shifting to create bubbles. The federal reserve helped spark a real estate bubble with low interest rates. FBI warnings of mortgage fraud were ignored. Consumers were constantly fed propaganda about "real estate only goes up." Then when that bubble popped, the US government bailed out those who caused it & burned trillions of Dollars propping up home prices. The government even bailed out a company that is now shorting the housing market (when that company was about to get bailed out the secretary of treasury leaked that material non-public information to some of his criminal investor buddies).

Does all the above sound circular, conflicting, corrupt & confusing? It should, because that is how power works & comes off as seeming semi-legitimate when acting in illigitimate ways. The perception of reality is warped to create profitable opportunties that are monetized on the way up and the way down.

Millions of kids take drugs that address the symptoms of being a child full of energy, imagination & entusiasm. In some cases they may need them, but in most cases they probably don't. The solution with the highest economic return gets the largest ad budget, even if it only treats symptoms.

Web Scrape Plus+ (Now With More Scraping)

When the +1 button & Google+ launched, Google highlighted how they would use the + button usage as a "relevancy" signal. Google recently started inserting + pages directly into the search results for brands & right from the very start they were using it as a scraper website that would outrank the original content source.

Google used the buy in from their promised relevancy signal to create a badge-based incentivized system which acts as a glorified PageRank funnel to further juice the rankings of these new pages on a domain name that already had a PageRank 10.

I recently read a blog post about how anyone could do the above & the opportunity is open to everyone. But the truth is, I can't state that something will become a relevancy signal that manipulates the search results in order to get buy in. Or, if I did something which actually had the same net effect, Google would likely chop my legs off for promoting a link scheme.

Recently the topic of Google+ as a scraper site came up yet again via Read Write Web & on Hacker News a Googler stated that it was "childish" to place any of the blame on Google!!!!!!

Google determines how much information is shown near each listing & can create "relevancy" signals in ways that things tied to Google get over-represented (look at the +1 count here). When they do that & it destroys other business models *of course* Google deserves 100% of the blame.

Thin Content & Scraper Sites

Remember the whole justification for Panda was that thin content was a poor user experience?

In spite of sites like eHow getting hit, Google is still pre-paying them to upload content to Youtube.

Now that the (non-Google hosted) thin content has been disappeared (and the % of downstream traffic from Google to Youtube has more then doubled in the past year) it is time for Google to take another slice of the search traffic stream with Search Plus Your World:

The Google vs Facebook locked down walled garden contest will retard innovation. As the corporate internet silos grow larger the independent web withers. Them going after each other may leave room for Twitter, but it doesn't leave lots of room is left for others, as the economics of publishing have to work or the publishers die.

Start ups that were on a successful trajectory were killed by Panda:

The startup had been on a roll up until last February when Google altered its ranking algorithm with the release of “Panda.” The changes decimated TeachStreet’s traffic, and the company never quite recovered.

“We lost a lot of our traffic, and overnight we started talking to partners for biz dev, not for acquisition,” he said. However, many of the potential partners wanted to know about an outright acquisition.

About.com was also smoked by Google:

The biggest worry, though, is that the decline of About.com itself may be irreversible. Fewer people are clicking on About ads placed by Google and the site’s own display ads have dropped in value.

The company has attributed this decline in value to Google’s decision last year to downgrade About pages in its search results. With more than 80% of traffic coming from search, the Google denigration was indeed a blow but About’s problems may be rooted in something deeper.

Keep in mind that the reason these websites were hit was that they were claimed to be thin & thus a poor user experience. When the NYT bought About.com one of the top competing bidders was Google!

Now that the "thin content" has been demoted in the search results Google can integrate deep content silos from Google+, like this one:

That is an 8-word Google+ post about how short another blog post is. I like Todd & do like to read his writings, but here Google is clearly favoring the same sort of content they would have torched if it was done on an independent webmaster's website.

How Google has raters view other websites that redirect traffic is based upon those sites having a substantial value add. Clearly in the above example there was nothing added to the interaction beyond sharing a bookmark with a punchy tagline.

If Google wants to use the + notation to pull up that other referenced page then perhaps that can make sense, but to list an 8-word Google+ page in the search results nearly a year after the Panda algorithm is outrageous. This sort of casual mention integration in the search results occurs on expensive keywords as well. Not only do they list your own Google+ posts...

...but they also list them from anyone you follow...

In addition to information pollution, the other big issue here is time. Google wants to make forms more standardized to make filling them out faster & they give regular sermons on the importance of fast search results. Yet when I do a navigational search, Google delivers two AdWords ads, a huge Google+ promotion, and then the navigational search result barely above the fold.*

*Since I thought the above was obnoxious, I renamed our Google+ company page to S_E_O Book to help Google fix their relevancy problems.

Can anyone explain how Google's speed bias is aligned with putting plus junk right at the top, even on brand searches? Yahoo! has been pretty aggressive with putting shopping ads in the search results, but their implementation is still a better user experience than what Google did above.

And Bing offers an even cleaner experience than that.

Due to how Google integrates Google+ in such a parasitic way I see no incentive for participating on their network except when I have something that is outside of my domain of expertise, something that I am not targeting commercially, something that is thin, or something irrelevant to say! That incentive structure combined with Google's photo meme feature will ensure that content marketers will help plenty of people see Star Wars stuff ranking for mortgage loan search queries.

When you own search/navigation you own language. that position can easily be extended into any other direction/market in a way a social graph can not:

"The only technology I’d rather own than Windows would be English," McNealy said. "All of those who use English would have to pay me a couple hundred dollars a year just for the right to speak English. And then I can charge you upgrades when I add new alphabet characters like ‘n’ and ‘t.’ It would be a wonderful business."

Further, Google can chose at any point to respond to or ignore market regulations in accordance with whatever makes them the most money. They can also fund 3rd parties doing the same (like undermining copyright) to force others to strike an official deal with Google to be "open."

A lot of businesses live on small profit margins, so Google's ability to insert itself & fund criminal 3rd parties aligned with Google's internal longterm interests is a big big big deal. Companies will learn that you either work with Google on Google's terms or you die.

When a public relations issue brews they can quickly change their approach and again position themselves as the white knight.

Brand Equity & Forcing the Brand Buy

Yahoo! put out a research paper highlighting activity bias, stating that the efficacy of online advertising is often over-stated because people who see ads about a topic were already more closely tied in with that particular network & that particular topic before they even saw the ad. As an example, any person who sees an AdWords ad for hemorrhoid treatment was already searching for hemorrhoid-related topics before they saw your ad (thus they were in the subset of individuals that might have came across your site in some way if you were in the search ad ecosystem or not).

This sort of activity bias-driven selection bias (homophily) exists on social networks online & offline.

Google did research on incrementality of ads & they came to the opposite conclusion as Yahoo! did. Google suggested you should buy, buy, buy, even on your own branded keywords. They suggested that testing was expensive (no mention that the only reason it is expensive is because Google chooses not to make such tools easily accessible to advertisers) & that the clicks were so cheap on branded keywords that you should buy, buy, buy. Many advertisers who mix brand & non-brand keywords together don't realize that they are using the "returns" from bidding on their own brand to subsidize over-paying for other keywords.

Google Analytics is the leading & most widely used web analytics program. They can share whatever metrics help them sell more ads (defaulting to crediting the last click for conversions, even if it was on a navigational search to your site) & pull back on features that are not aligned with their business interests (SEO referral data anyone?)

This goes back to Scott McNealy's quote: "The only technology I’d rather own than Windows would be English. All of those who use English would have to pay me a couple hundred dollars a year just for the right to speak English. And then I can charge you upgrades when I add new alphabet characters like ‘n’ and ‘t.’ It would be a wonderful business."

Analysts didn't understand why Google CPC rates were down 8% last quarter while overall search clicks were up 34%. The biggest single reason was likely more clicks on adlinks on branded AdWords ads. While a brand buying its own keyword typically pays far less per click than what some of the biggest keywords go for, the branded keywords typically have an exceptionally high CTR. Those additional clicks dragged down Google's average CPC, but the extra revenue they offered was a big par of the reason why Google was about to grow at 25% even though their display network only grew at 15%.

That slow growth of display is in spite of Youtube now serving over 4 billion video streams per day & Google adding display ads to log out pages.

Online views are not the same as TV views. A comScore study found that 31% of display ads are never seen. In spite of that, US online advertising will reach nearly $40 billion this year.

Google wants to insert itself as a needed cost of business in the same way credit card companies have.

On Google Maps they put an ad inside your location box.

Even if most people don't participate on Google+, Google can still force advertiser buy in through over-promotion of the network in the search results. On your branded keywords they may drive your organic listing below the fold & put Google+ front & center.

Facebook earnings are still growing much faster than Google's & Facebook encourages advertisers to advertise their Facebook pages, so even when you pay for the click Facebook still keeps the user. Facebook is adding apps to the timeline & is trying to win VEVO music video hosting from YouTube.

While Google is primarily known as a search company, it is getting harder to get off of Google though any channel other than a toll booth. Google keeps driving the organic search results downward, while Google verticals fill up many of the organic results that remain. Many companies already buy Google ads on their own YouTube content. Some buy ads on Google to drive them to their Youtube videos & then buy ads on their own Youtube video to promote their websites. Soon Google will try to push you to buy them on your Google+ page as well. Google is becoming a walled garden:

Google wants to control more elements of your social world now. They don’t just want to be a search engine.

Is that so bad? Maybe not. It’s certainly no different from how other companies, from AOL, to Microsoft, to Apple, to Disney, to Facebook, have viewed the world — as ideally a walled garden, an all-consuming platform that most people use for pretty much every form of entertainment and social interaction.

A lot of people thought that Google was somehow different. They were, of course, wrong.
...
To move forward either as the old Google or Google+, Google needs to be capable of making fair deals with the partner ecosystem. It needs to curb its instinct to kill competing media companies that were actually producing great content that Google helped you find.

I suspect there will be plenty of bloodshed before Google figures that one out.

"This is the path we’re headed down – a single unified, ‘beautiful’ product across everything. If you don’t get that, then you should probably work somewhere else." - Larry Page

Google no longer believes in the concept of the open web. Blame it on Larry Page becoming the CEO, blame it on him talking to Steve Jobs & Steve telling him to make fewer and tighter products, blame it on Google funding eHow, or blame it on basically anything. But if you go back far enough, much of the stuff that is going on now was clearly envisioned a decade ago:

I was lucky enough to chat with Larry one-to- one about his expectations for Google back in 2002. He laid out far-reaching views that had nothing to do with short-term revenue goals, but raised questions about how Google would anticipate the day sensors and memory became so cheap that individuals would record every moment of their lives. He wondered how Google could become like a better version of the RIAA - not just a mediator of digital music licensing - but a marketplace for fair distribution of all forms of digitized content. I left that meeting with a sense that Larry was thinking far more deeply about the future than I was, and I was convinced he would play a large role in shaping it. I would rather jump on board that bullet train than ride a local that never missed a revenue stop but never." - Douglas Edwards

What happens when the Google+ version of your content outranks the version on your own site? And what happens when your branded channel and/or your fans become a vertical ad silo Google sells to your competitors?

I tested submitting a couple posts to Google+ with a Wordtracker top keywords list & valuable keywords (on a cpc*traffic) basis in posts about top keywords. Those posts rank #2 or #3 in Google for many people that follows me. No harm to me since those posts were irrelevant to this site, but if they were about my theme & topic I just would have out-competed myself. When Google outranks you (even with a copy of your content) they get to taste the data again and sell off the attention another time. You only get a slice of that monetization, even when it is your work that is being monetized. Maybe it is great for stuff that is somewhat less relevant and/or keywords that are so competitive that you otherwise wouldn't score for them, but we have to be really careful we don't out-compete ourselves. Though if Googke keeps this up they won't be the only ones monetizing it. Give it a few months and celebrities will be selling sponsored Google+ posts based on some metric created by multiplying search volume, CPC & how many followers they have.

Is Bing Better? Will Enough People Ask That Question to Matter?

For years Google built their reputation as being the search engine that offered the cleanest & fastest search results. They were known for monetizing less aggressively than the competition. But over the past couple years Google has dialed up their ads to where they now send a greater ratio of ad traffic than organic search traffic. One Google engineer recently described the ability to rank highly in Google without buying their ads as being a bug that was getting fixed!

Google's big risk in their coupling of aggressive monetization, aggressive self-promotion & changing how users feel about user privacy is that they can create the perception that users should go elsewhere for for an honest or trustworthy search. This not only builds momentum for smaller search services like DuckDuckGo & Blekko, but has also won praise for Bing from Gizmodo, Dave Winer & The Next Web.

Z emailu: Otazka k obchodu s CN

Fri, 03 Feb 2012 18:21:09 +0000

Prislo do emailu: Chtel bych se ctenaru zeptat na to jak zacinali (v malem) obchodovat s neznamymi firmami v Cine. Jde mi o uplne zacatky, kdyz vubec o firmach nic netusite (v CN kupujeme, ale u firem kde z jinych zdroju neco vime). Dam priklad, potrebuji koupit nejake displaye (jako soucastku pro dalsi vyrobu). Takovou [...]

Kill The Bugs!

Fri, 03 Feb 2012 15:47:54 +0000

You can learn a lot more about what Google really thinks by reading what their new hires say. They are not yet skilled in the arts of public relations & make major gaffs like this one:

Instead of being able to SEO the entire Internet, businesses can now only affect the search results for a tiny percentage of users. That's a good thing because SEO can't scale, and SEO isn't good for users or the Internet at large.

If you look at the Google experience from the standpoint of customers, it's pretty good. Users get relevant search results and ads. Advertisers get their content on top of everything else. It's a good compromise between advertising and usability, and it works really well. It's a bug that you could rank highly in Google without buying ads, and Google is trying to fix the bug. Manipulating Google results shouldn't be something you feel entitled to be able to do. If you want to rank highly in Google, be relevant for the user currently searching. Engage him in social media or email, provide relevant information about what you're selling, and, generally, be a "good match" for what the user wants. - Googley Jon Rockway

Would love to hear someone more senior confirm this as the official Google company position, however they are too skilled at public relations to make that blunder (at least outside of foreign AdWords ads that tell you to "forget SEO").

The Inside Story of a Small Startup Acquisition (Part 2)

Thu, 02 Feb 2012 16:00:52 +0000

Photo by psiaki Why I Bought My Next Startup (Instead of Building It) This is part 2 in a series covering my acquisition of HitTail; part 1 went to the top of Hacker News last week and I have a slew of questions from that discussion that I will answer next week. But first I want to address [...]

Jak budovat databázi e-mailových adres: Vyzkoušejte službu PadiAct.com

Wed, 01 Feb 2012 16:36:34 +0000

E-mailing může být zajímavý a levný způsob, jak zvýšit prodej a budovat dlouhodobě vztah se stávajícími nebo potenciálními zákazníky. Bohužel firmy často nemají dostatečně velkou databázi e-mailů a spoléhají se jen na rozesílky na stávající zákazníky. Jednou ze základních možností, jak rozšířit databázi e-mailů, je registrační formulář na webových stránkách. Zkuste se podívat do Google Analytics, [...]

AHREFS Review: An In-Depth Look at a New Link Research Tool

Wed, 01 Feb 2012 01:28:48 +0000

Ahrefs is the newest entry into the link research tool space. They use their own bot and their own index (which they state is based on information from a trillion website connections).

They claim their index is updated every 30 minutes and the fresh data is available to their users within 30 minutes of the actual index refresh.

Ahrefs also has a ranking database of roughly 45 million keywords from 9 different countries (US GB FR RU DE ES IT AU BR). The tools within their membership are:

Site Explorer
SERPs Analysis
Reports
Labs/Tools

Their pricing is very straight-forward and only increases or decreases based on volume of data you have access to. You can check out the easy to understand pricing on their pricing page (and they offer SEO Book readers a 50% discount on the first month).

Site Explorer

Ahref's Site Explorer functions in a similar way to Majestic's Site Explorer and SeoMoz's Site Explorer. You can choose a specific URL, the domain without subdomains, or domain with all its subdomains:

If we look at the Site Explorer results, you'll see an overview of the last 45 days or so from Ahref's crawl history:

On the left you can see some interesting stats like the total number of backlinks, different referring IP's and subnets (class c blocks and such), unique domains, and the types of backlinks the site has (text, image, redirects, and so on).

In addition to the overview report, you have other research options to chose from:

New Links
Lost Links (great opportunity for you to swoop in and alert the linker + sell them on linking to you and your resources)
Anchor Text Profile
Pages Crawled on the Site
Referring Domain Breakdown
SERP Positions (organic ranking report)
Raw Export of the Data (up to your limits based on your pricing plan)

New Links

In the New Link tab you can go back to a previous month, or work inside the current month, and find newly discovered links by the day. Here is what that looks like:

Click on whatever day you want and you'll get a list of linking urls, the target link page, and the anchor text used for the link:

This report can help you reverse engineer, down to the day, a link building campaign that your competitor is running (always good to be out in front of a big link push by a competitor) and can also help you evaluate your own link campaign or even help you spot a link growth issue that may have resulted in some kind of penalty or over-optimization filter.

Now keep in mind that, based on their stated crawling guidelines, the stronger links from stronger sites tend to get crawled more frequently so the spammiest of the spammy link approaches might not get picked up on. For that level of deep research a historical report from Majestic SEO and a link status checker, like Advanced Link Manager, is likely a better bet.

You can export this report to Excel or .CSV format.

Lost Links

The Lost Links tab has the same interface as the New Links report does. For your own domain you might want to consider tracking your own links in something like Raven or Buzzstream but this tool does report dropped links down to the day. Combine that with their crawling preferences (better links = quicker attention) and you can spot drops of substance quickly.

You can use this report to find links that a competitor has lost, off of which you can contact the webmaster and see if you can't promote your site or similar content to earn the link your competitor was previously getting.

You can export this report to Excel or .CSV format.

Anchor Text

The anchor text report is exactly what you expect it to be. It lists the anchor texts of external links, the number of occurrences, as well as an expandable dropdown menu to see the pages being linked from and the pages being linked to on the site you are researching.

You can export to Excel or .CSV and choose to export everything, up to your limit, or just the current page.

Crawled Pages

This report will show you all the pages crawled by Ahrefs with the following stats:

Page URL and Title
Crawl Date
Page Size
Internal Links
External Links

I would likely use this report (on competitors) for checking some of their more popular internally linked-to pages as well as checking out how they structure their site. You can also jump right to a site explorer report for any of the URL's listed on that report as well as check the SERP positions for any of them.

Referring Domains

One thing I like about Ahrefs is that it's straight and to the point. It's very easy to get in, get your data, and get out. Each report does pretty much what you expect it to. This report shows the referring domains + number links coming from that domain. You can access the links from each domain by clicking the Expand button next to the referring domain:

SERP Positioning

Similar to SemRush, Ahref's provides estimated ranking data for keyword sets on both Google and Bing/Yahoo in multiple countries (US, UK, AU, DE, FR, ES, IT, BR, RU). The tool shows the:

Position
Keyword
CPC
Estimated cost
Ranking url
Global search volume
Advertiser competition
Last date checked
Rating (estimated visitors per month based on assumed traffic distribution)

The other cool thing about this report is that it will tell you the change from the last time they checked the ranking.

SERPs Analysis

This is similar to the SERP positioning report. Essentially, you enter a URL and you get the Google + Bing & Yahoo ranking data with those same metrics as stated above:

Position
Keyword
CPC
Estimated cost
Ranking url
Global search volume
Advertiser competition
Last date checked
Rating (estimated visitors per month based on assumed traffic distribution)

In addition to that, you also have the following reports:

Daily Stats
History of Changes
Ads Analysis

Daily Stats

This report shows you, on a daily basis, the following data points:

New Keywords
Lost Keywords
Total Keywords that moved up
Total Keywords that moved down
Total Positions up
Total Positions down
Rating Change (estimated percentage of traffic gained or lost)
Cost Change (rating change * CPC)

There are graphical charts for:

Search Engine Traffic (shown above)
Keyword Trend (total keywords ranking)
Traffic Cost
Bar Graph for New and Lost Keywords
Estimated Traffic Changes
Estimated Traffic Cost Changes

History of Changes

This report breaks down the keyword changes by day and how much the specific keyword moved up/down (and the corresponding page that is ranking).

You can look at a daily report, a 7 day report, 30 days, or a custom range.

Ads Analysis

Ahrefs also incorporates Google (and Bing/Yahoo but I had a hard time getting figures for Bing/Yahoo) PPC data. You can pull in the ranking of the ad, the ad text, volume & CPC data, as well as last updated date & competition levels.

You can look at just the keyword/ranking data or choose from their other 2 reports; keywords/ranking + ad text (Table + Ads) or just the PPC ad text itself (Ads Preview).

Reports

You can create reports for your own domain for free or a any other site as a part of your subscription. Each domain counts as a separate report, so you can enter as many as you are entitled to in this interface but they do count against your monthly allowance.

The report overview looks like this:

Each tab represents a data point you can review. In any tab you can choose to export the visible page or the entire report.

There are quite a few filtering options here, as you can see below:

Your filtering options, report-wide, are:

URLs from - you can include or exclude based on user-defined data (exclude by word(s), domain extension, and so on)
Backlinks Type - you can choose to show, specifically, different backlink types (nofollow, image, frame, redirect, form, deleted)
Pages - show links only to a specific page
Subdomain - show links only to a specific subdomain
Countries - show links from specific countries
Anchors - show or exclude specific anchor text links
Referring Domains - show links from a specific domain, or set of domains only
IP - show links from a specific IP or range of IP's only
Subnets - show backlinks only from specific subnets
TLD - show links from specific TLD's only
Date - show links based on specific crawl period

The cool thing here is that you can layer on the filters as you wish. The following screenshot shows all filters selected and available:

The reporting is really quite powerful and provides numerous ways to quickly filter out junk links so you can focus on the good stuff.

Labs

There are 3 additional tools in their Labs section.

Ahrefs Top - Top 1 million domains by number of backlinks, completely searchable
Domain Comparison - compare up to 5 domains for different link metrics (see below)
Batch Domains - (see below) dump in a bunch of URLs and get a total count of backlinks, referring domains, and IP's. Unsure of the limit here but I did about 25 with no problem.

Here is a screenshot of the domain comparison feature:

The Batch Domains feature looks like this (and is completely exportable!):

Ahrefs is Worth a Spin

I was impressed with the speed of this tool, the exportability of the data, and the report filtering capabilities. It hardly hurts to have another link database to pull from, especially one that is updated every 30 minutes.

The tool is quite easy to use and it does pretty much what you expect it to. If you are into link research you should give this tool a try. The database appears to be a fairly good size for a new database and the ability to slice and dice that data from right within the web interface is a solid feature. If you do try it out, let us know what you think! We are also adding their link data to SEO for Firefox & the SEO Toolbar today.

Goldilocks SEO

Tue, 31 Jan 2012 19:56:17 +0000

Sharing is caring!

Please share :)

Embed code is here.

Why Has Andreessen Horowitz Raised $2.7B in 3 Years?

Tue, 31 Jan 2012 15:04:27 +0000

Since Marc and I founded Andreessen Horowitz three years ago, we have raised $2.7 billion. That statement begs a few questions. The two most obvious are: Why did such a new venture capital firm raise so much money? How did such a new venture capital firm raise so much money? To get to the answers,…

Pretend This Optimization Doesn't Exist

James Hague — Tue, 31 Jan 2012 06:00:00 +0000

In any modern discussion of algorithms, there's mention of being cache-friendly, of organizing data in a way that's a good match for the memory architectures of CPUs. There's an inevitable attempt at making the concepts concrete with a benchmark manipulating huge--1000x1000--matrices. When rows are organized sequentially in memory, no worries, but switch to column-major order, and there's a very real slowdown. This is used to drive home the impressive gains to be had if you keep cache-friendliness in mind. Now forget all about that and get on with your projects. It's difficult to design code for non-trivial problems. Beautiful code quickly falls apart, and it takes effort to keep things both organized and correct. Now add in another constraint: that the solution needs to access memory in linear patterns and avoid chasing pointers to parts unknown. You'll go mad trying to write code that way. It's like writing a short story without using the letter "t." If you fixate on the inner workings of caches, fundamental and useful techniques suddenly turn horrible. Reading a single global byte loads an entire cache line. Think objects are better? Querying a byte-sized field is just as bad. Spreading the state of a program across objects scattered throughout memory is guaranteed to set off alarms when you run a hardware-level performance analyzer. Linked lists are a worst case, potentially jumping to a new cache line for each element. That's damning evidence against languages like Haskell, Clojure, and Erlang. Yet some naive developers insist on using Haskell, Clojure, and Erlang, and they cavalierly disregard the warnings of the hardware engineers and use lists as their primary data structure.... ...and they manage to write code where performance is not an issue. (If you liked this, you might enjoy Impressed by Slow Code.)

Changing the game

Gabriel Weinberg — Mon, 30 Jan 2012 14:38:02 +0000

I watched Moneyball last night and saw a lot of parallels to startup land, but tying them all together was the notion of competing effectively by changing the game. In the (true) story the Oakland A's competed against big money teams by leveraging undervalued players to achieve real world gains (focusing on runs instead of wins, though which of course correlate with wins).

The parallel is small startups competing with well-capitalized companies, especially ones with entrenched business and culture. The notion of where to find undervalued assets is where it gets interesting to me.

For example, one tactic is to look at customer acquisition channels too small for your competitors to care about. I did reddit ads when they first came out and since there was no one there you could get a huge ROI.

The saturation point was so small that for any major competitor it is a complete waste of time (now that Reddit is bigger, that may have changed). Nevertheless, you as a small startup can capture that ROI to get a beachhead. I put startup micro-opportunities in the same category.

However, neither of these tactics really change the game. They just get you into it.

The canonical way to change the game is famously outlined in The Innovators Dilemma where you focus on the future disruptive product line that the big business cannot focus on because it is so disruptive. Amazon Web Services is a great example of this strategy in the present.

Another way is to treat the bigger business(es) as commodities and layer on something valuable on-top. This is kind of like commoditizing the complement but you're treating your competitors as the complement.

Like the disruptive innovation, this strategy can make you seem like a toy at first, but if you end up closer to the customer you can capture real value. This is potentially risky if there is only one major competitor (which makes it hard to commoditize!), but often there is more than one place to get the same thing. And you could eventually get big enough to replicate the underlying asset(s) yourself. I put DuckDuckGo in this category.

A third way is to reframe the same problem by changing the way the customer interacts with the product, while keeping the same end goal/result. In Internet startup world this is often re-thinking the UI/UX completely. Mint was in this category.

You could also change the metrics that matter (akin runs vs wins). Badoo is doing this in the dating space by making base interactions free and focusing on meeting new people as opposed to strictly dating.

Here's one more game-changing strategy that comes to mind: shrinking the market. Craigslist is a famous example. Take a huge market and then just totally gut it, but find a profitable businesses in there for you (in this case, jobs).

Some of these strategies can overlap, and I'm sure there are others as well. If you think of any more, please let me know.

The other thing that really popped out at me in the movie was Billy Beane not watching games. It reminded me of vanity metrics. Not only did he get better metrics but he didn't waste time watching them.

It's all related. If you're a startup competing against huge companies, by definition you need leverage on something to compete effectively. Wasting time playing the same game is well, wasting time.