Web 2.0 productivity

One of the most inspiring sessions at Web 2.0 Expo last week was “Mastering the Low-Information Diet” by entrepreneur Tim Ferriss. Tim is author of the book The 4-Hour Work Week. (Yes, it’s true, he spends only 4 hours per week running his 2 businesses.) The session was part of the “Ignite” evening of lightning-round 5-minute sessions. Tim’s presentation was voted by the audience (using cell phone voting via SMS) as one of the best sessions of the night and was thus included in the keynote on Day 3 of the conference. So I got to enjoy Tim’s 5-minute “drinking from a firehose” talk — twice! It’s amazing what a speaking pro can do with a mere 5 minutes! Here are my notes from the session:

Tip #1: “Selective ignorance”
We’re in a world of infinite interruption and infinite minutia. Practice “selective ignorance” — you don’t need to know and follow everything.

Tip #2: “Batching”
Batching is performing similar tasks at set times. You only do these tasks at specific times and in the meantime you let them accumulate.
For example, check your email only twice a day. Use an auto-responder to let urgent issues get picked up sooner. Example of auto-response message: “Dear esteemed colleagues, In order to get more done, I’m checking email only twice per day — once in the late morning and once in the late afternoon. If you require a response sooner than that, please call my cell phone at 555-555-5555.”

Tip #3: Pareto Principle
Focus on the “critical few,” not the “trivial many.”
You may ask “What if i miss something important?” Tim responds that he’s never missed anything that cost him more than $500.00. Whereas, by practicing this, he has gained millions of dollars in additional booked revenue.
“Pareto Principle” is the 80/20 rule. For example, 20% of the people in your business life will consume 80% of your time. Not all customers are created equal. 5% of your customers may contribute 95% of your profit. Figure out which customers are not profitable and “fire” them. Tim fired the worst offenders and put remaining lower-profitability customers “on auto-pilot” — never proactively contacting them or thinking about them.

Tip #4: Outsource your life
Tim has between 20 and 40 MBAs around the world that he outsources various aspects of his business and personal life to, such as: database creation / prospect list creation, etc… Even online dating!
Find people to outsource your life to on GetFriday.com (7 day trial), Elance.com, etc.
If you can pay someone half or less of what you earn per hour and they can do a reasonable job of it, outsource it!

Tip #5: Schedule life in advance
It’s not about “work-life balance,” it’s about “work-life SEPARATION”.
If you have a void, you’ll fill it with work. So fill your schedule with personal activities too, not just business activities.

Read Tim’s blog for more of his wisdom.

Watch his talk at Ignite

Open APIs: Free content for SEO, just add water

Having a large content-rich website makes life a lot easier when it comes to SEO. Particularly if you’re trying to rank well in Google for a range of different keywords. APIs (Application Programming Interfaces) to the rescue! “Mash up” multiple APIs together and voila! — unique content. HousingMaps was one of the first and most notable mash-ups. What can you mix-and-match to make into a deep site full of keyword-rich useful content? Start your journey by consulting ProgrammableWeb’s APIs list. This list is famous among programmers.

Anybody who’s into APIs will know the site Programmable Web. I had the pleasure of attending the session about APIs presented by John at Programmable Web last week at the Web 2.0 Expo.

Here are my notes from John’s session:

ProgrammableWeb.com
Site launched in Summmer 2005.
APIs, mashups and tools directory
News, community, reference materials
400 APIs and 1800 mashups

Why bother with APIs?
Make money: eBay over 45% of all products get listed via their APIs. 3 billion API calls per month.
Save money: Smugmug saves $500k year with Amazon S3 storage.
Build brand: Google Maps 300% growth vs 20% Mapquest
Build empires: Salesforce.com – over 50% of all transactions via their API. 500 apps so far.
Innovate: 100′s of Flickr apps. Is Twitter next?

What makes an API successful?
Rule #1, the underlying service: or, a good API on a bad service is like lipstick on a pig
A plan and a business model
Simple, open, easily adopted
Providing choices
Good developer support

Number of API providers in each of the top 10 categories:
Geo/mapping 38%
Reference 26%
Internet 21%
Search 18%
Shopping 18%
Music 17%
Photos 14%
Messaging 14%
Bookmarks 10 %
Widgets 10%

Unexpected sources of APIs:
HOTorNOT
U.S. postal service
Weatherbug
Betfair
Theholybible
Secondlife
theworkforyou
Sunlight

API Business Models

Who	What	How
Amazon ECS	retail	affiliate model
Amazon.com	infrastructure	pay-as-you-go
eBay	auctions	APIs for listings
Salesforce.com	CRM	“seats”
Exact Target	bulk email	tiered pricing (they do millions of API transactions per quarter)
Cypress Golf	ASP for reserving tee times	integration
Rhapsody	music	partnerships

Best practice is to bake your business model into your API

API pricing & licensing models

Revenue sharing	via affiliates; advertising (see new AdSense API)
Subscription	per time period; classic Software as a Service
Tiered	pricing in volume buckets
Per-call metering	fee per API call (see strikeiron)
Utility	API is free, but fee for resources (cpu, storage, bandwidth)
Pay per conversion or sign-up	commissions and referrals (yahoo mail)
Ad-supported is free, ad-free is premium	see ad and branding-free premium mapping
Units	different calls priced at different levels (google AdWords API)

APIs as bizdev (business development) enabler
decentralized business development
an API can lower barriers to working with you
consumer/hacker –> business/partner, e.g. Zazzle, Englaze, Qoop, Moo (we love to print)

Design
design for people & computers

Technology: the big debates
protocols – REST vs SOAP

Top 5 API protocols & styles
1. rest 68%
2. soap 40%
3. javascript 7%
4. xml-rpc 6%
5. APP (atom publishing protocol), Google gdata 3%
specialized or proprietary (xmpp, sms, saml, skype, oscar)
(20% support more than 1 of these protocols)

REST vs SOAP – pros vs. cons

Protocol	Pros	Cons
SOAP	established standard, wide tool support	complexity, interoperability issues
REST	simplicity, built on how the web works (URIs and core HTTP methods)	lack of standards, tool support challenges

Comparing protocol choices
77 APIs from 6 providers:

	rest	soap	xml-rpc	javascript	other
google	11	3	1	4	3
amazon	8	10
microsoft	1	6	1	3	1
ebay	2	2			1
yahoo	19	1		3	2
aol	3	1	2	2	4

favor simple designs
use more than one to broaden your base
be as consistent as possible

Data formats supported: 2 dozen
xml, json, rss, atom, yaml, icalendar, csv, serialized php, html, pn, georss, vcard, text, rdf, opml, vml, tv-anytime, hcalendar, xspf, sql, gml, cdf
favor simplicity and standards
use more than one to broaden your base
provide domain-appropriate alternatives
25% of new apis listed at Programmable Web in 2007 support json
8% of all apis listed before 2007 supported json

xml: well established, general purpose; greater parsing overhead; well suited for documents; wide tool support
json: objects and simple types; fast parsing, good fit for client side scripting; bindings for 20+ languages

Yahoo Mail added json support in 2 days

John’s favorite APIs are from Flickr

#1 api provider issue is adoption
How many steps from zero-to-code? Easy sign up and self service is key. Offer a free level of API use.
How well do you treat your developers? Developer network and evangelism
How’s your pre-code support? Docs, gallery, forums, mailing lists, and lots of code. Media: videos, screencasts, podcasts
Any post-code support? Certification, marketing, events. System status (trust.salesforce.com shows availability of all their API servers), usage reporting

Know the tools ecosystem:
Data mashup tools – yahoo pipes, grazr, rssbus, data mashups
Scraping tools – kapow, dapper
Visual development tools – jackbe, qedwiki, proto, teqlo, bungee labs
DIY consumer tools – platial, google my maps, map builder

Risks and issues
Dependencies: business and technical. Strategize accordingly – go in eyes-open. Mashups: know switching costs (consider tools like Mapstraction)
Lack of service level agreements. Consider commercial license models (ex google maps enterprise). Mashups: should also disclaim warranty (“trickle-down terms”)
Commercial vs. non-commercial use. Most API terms of service differentiate these, but vaguely. If you place ads on your mashup, is that “commercial use”?
Copyright law issues. What data can be remixed and how? See: Flickr API integrates Creative Commons license.

The big issue: identity
Where are the ‘personal mashups’? high-value data locked behind IDs and password
Limited standardization & consensus. Even in new competing “authorization APIs”. Yahoo! bbauth, microsoft liveid, google web auth
Solutions: OpenID? Meta-identity networks? TBD

Privacy issues. Beware.
Data mining 101: e.g. finding subversives by mining Amazon wishlists for readers of “1984″

Trends
Battle of the platform players
Having an API as a checklist item. Got an API? Check.
Lightweight services & data. rest protocol & json data format
New tools. for developers and non-devs
Enterprise mashups (mashups behind a firewall)
New business models (even more business models will be forthcoming)

Recommendations
Have an API strategy and a plan
Keep it simple: protocols, formats, tools, adoption
Provide choices: in order to reach the widest base
Treat developers well: and they will treat you well in return

For Microsoft, there’s always the “nuclear option”

I just had a very enjoyable dinner with my colleague Alan Rimm-Kaufman, founder of paid search firm, Rimm-Kaufman Group. We always have such thought-provoking conversations whenever we meet up. This evening was no exception.

At one point our conversation turned to Microsoft. What can they do about Google and Yahoo eating their lunch? Microsoft has become an “also ran” in the paid search space. They are so far behind in paid search (organic search, too, for that matter), that it’s going to take a miracle for them to pull out from behind. Or not? Could they perhaps launch a devastating nuclear attack against Google, wiping out the entire paid search industry in the process? Believe it or not, the answer is “yes,” and the approach would be surprisingly simple and inexpensive for Microsoft to implement…

Alan described Paul Bryant’s “doomsday” scenario to me as follows (hat tip to Alan, btw)… Microsoft simply introduces a Google ad blocker to Windows (which works in IE and Firefox) and pushes it out through a Windows Update. And here’s the kicker — Microsoft turns it ON by default! Of course they will also have to block their own ads, so they don’t get nailed under anti-trust laws.

Scary.

Coolest new web analytics tool you’ve never heard of

One of the 5-minute “lightning round” sessions at Web 2.0 Expo on Sunday night was the folks from Nitobi demoing their brand new Robot Replay technology. All I have to say is… WOW!! Every one of you readers HAVE to go check this tool out. It is a free beta, so there is no excuse for not trying it out! Robot Replay allows you to track the mouse gestures of your visitors and play them back. Imagine tracking individual prospects and see how they are interacting with your site! Not just where they click, but whether they scroll, how they interact with your forms, etc.

Does your brand have what it takes to go viral on YouTube?

Success on YouTube is as much about effectively tapping the social network as it is about the content. Brand-building viral videos such as the ones I blogged about recently only happen if the conditions are right.

If you don’t follow MarketingProfs, you may not have seen that my article “How to Market on YouTube” came out this week. For the article I had interviewed the marketing director at Blendtec about their famous “Will It Blend?” videos. “Will It Blend?” is one of my favorite examples of viral marketing on YouTube. It didn’t take a big budget and an ad agency to dream it up, but boy did it work wonders for their brand!

So what are the secret ingredients to going viral on YouTube? I guess you’ll have to read my article to find out!

Badges, gadgets and widgets = Link bait!

I’m here at the Web 2.0 Expo. Sat in on the session today “An Overview of Badges and Widgets: The Fast Rise of Viral Web Parts”.

Before I get into the content of the session, I have just a couple of remarks. These come from me and not from the session presenter…

• Badges and widgets CAN be amazing link bait. If it’s a useful widget and if it’s launched properly AND if the widget incorporates links back to your site in the right way, it can go viral and start building up your link popularity and PageRank.
• Inherently the widgets placed on other people’s websites aren’t helping your PageRank. That’s because most widgets are based on AJAX, JavaScript, or Flash — and links that are inside a Flash movie or within a JavaScript program are not treated like normal links by the search engine spiders. You need to think out-of-the-box if you want to capture PageRank from the page on which the widget is placed. That’s not to say you won’t get PageRank from bloggers talking about your widgets on their blogs, because you will. It’s just that the widget itself won’t (typically) pass PageRank.
• I think plugins (e.g. WordPress plugins) should have been included in the presenter’s list of viral web components. Plugins can be amazing link bait. We at Netconcepts are starting to witness this for ourselves with our SEO Title Tag plugin for WordPress. I’m hoping to get to the point that the plugin page on our site beats out our home page in terms of PageRank score. Which is a tall order when you consider that our home page fluctuates between a 7 and an 8!

Anyways, here are my notes from the session…

• An emergent phenomenon is web sites with “portable” content and functionality.
• Forrester Research says portable content is a key trend.
• There is limited business value in being on a single site…. YouTube and Google are showing the industry what’s possible.
• There’s a trend towards the “atomization” of content. Small pieces are easier to reuse and more general purpose. Microformats are the smallest pieces.
• Exploit Jakob nielsen’s Law of Web User Experience, that “users spend most of their time on other websites”. Design your products and services to leverage this fact.
• Spread your product beyond the boundaries of your site: badges, widgets, gadgets, apis, syndication.
• It should be end-user friendly.
• Build on the shoulders of giants: leveraging widgets and APIs from Yahoo, Amazon and thousands of others.
• It’s automated mass servicing of markets of low demand content and functionality (The Long Tail).
• Widgets are small applications or bits of functionality that can be embedded on the web — can be AJAX or Flash.
• Badges are displays of content pulled under the covers from other sites.
• Gadgets are more formal widget models from Google and Microsoft.
• There’s a widget standard under consideration by the W3C.
• Netvibes offers a universal widget architecture.
• Google and Microsoft have their own gadget initiatives. Both have a developer community.
• Ease of consumption and distribution is critically important. Copy and paste is best — e.g. a single line of Javascript or object/embed tags for Flash.
• Connect to their underlying sites to provide value.
• Have a business model baked deeply into it — driving site traffic, content consumption, advertising, etc.
• Widgets are often virally self-distributable, triggering network effects.
• Build a simple “dashboard” and applications (aka mashups)
• Google Gadgets directory is broken down into desktop gadgets and web-based gadgets.
• Google Docs & Spreadsheets gadget is an excellent example of a widget for your website.
• Microsoft’s gadgets directory is in the Windows Live Gallery.
• Huge directory of widgets at widgetbox.com.
• Google’s AdWords widget is probably the most successful widget in history. It turns the entire web into Google’s ad platform (The Long Tail of content/advertising). Purported 80% of Google’s revenue comes from advertising, and 80% of that 80% comes from displays within the widget (i.e. from Google’s “content network”). Key aspects of the widget: good user incentive, extreme ease of use, strong viral feedback loop.
• Widgets and badges are your front end to your APIs.
• Key design considerations include: Scalability (cost effectiveness, reliability, exploitation by others, global reach, security), Clearly thinking through the cross domain issues (sharing of personal data, will it work on mobile?, selectively allow users’ personal data such as pictures or video, no security holes), IP issues (do you have a license to redistribute the content you have, can others violate the IP protections of others? and if so, what will you do about it when you put other businesses at risk with your widget?, do you widget make it hard for others to take content out of the widget?), Ease of consumption (really must be a simple copy and paste to deploy), Leverage network effects (encourage every viewer to share it with others, letting users copy the widget), end user motivation (must do something useful for them, hsaring interesting content, providing shared access to their personal data such as photos or audio or even paying them, e.g. AdSense)
• According to Programmableweb.com, there are currently over 1700 mashups.
• Small pieces, loosely joined.
• Reuse the web palette.
• Service level agreements (SLAs) will get much more interesting in a highly composite world.
• Get experience now: begin trials to offer your capabilities via services.

Black Hat SEO meets Web 2.0

I’m here at the Web 2.0 Expo in San Francisco. Today (actually yesterday now that it’s after midnight) I sat in on the SEO Workshop being presented by Todd Friesen (“The Oilman”) and Greg Boser (“WebGuerrilla”).

This session turned out to be a lot of fun. It was reminiscent of their “SEO Rock Stars” radio show on Webmaster Radio. I cracked up when Todd plugged their show and explained the name of the show by saying: “Yes, we really are that arrogant.”

Much more of the session than I was expecting turned to “black hat” or “gray hat” SEO tactics — things that are outside the search engine guidelines. Both Todd and Greg believe in being pragmatic about SEO. Greg analogized SEO to speeding: nobody goes the speed limit; just don’t drive recklessly swerving in and out of traffic so you get in a wreck. He “hates the guidelines” and longs for the good ol’ days when the engines didn’t have such idealistic guidelines and if you went too far, you were simply “torched forever and you’re gone.” Greg’s guiding principle: “I don’t want to upset my mother” — i.e. he gauges whether he’s gone too far based on whether she’d be unhappy with the search results because they’re useless. They say the search engine guidelines are simply that: guidelines.

Hmm. I’m not sure I’d take that tack with an audience of webheads from Web 2.0 startups. An audience of SEOs at SES (Search Engine Strategies) is one thing, because they can be discerning about how far to take various bits of advice. But Web 2.0 geeks? That aren’t savvy enough about SEO to know when they are playing with fire. Greg and Todd made compelling arguments for playing with conditional redirects (serving different destination URLs to Googlebot than to humans). But the 4 major engines had specifically warned against doing that in the session “Search Engine Q&A on Links” last Friday at Search Engine Strategies in NYC. So you’d better REALLY know what you’re doing if you’re going to play with that stuff! Todd and Greg also testified to the benefits of cloaking. One of their client’s sites was cloaked for 3 years — a new ecommerce platform was purchased and launched by the client, but the old HTML-based site was served up selectively to the bots because it ranked so well. I’m sure that was worth a pile of money to his client. But boy if you get that wrong (like not keeping up to date with all the ever-changing list of IP addresses associated with Googlebot), things could go very badly.

Todd made the point that if you are a big brand or company, you’re not going to get kicked out of Google. Or at worst, maybe it’ll be for a day. He cited eBay and the NY Times as examples of sites that won’t get banned despite operating outside the Google Guidelines by serving millions of “related search” results and/or cloaking. Todd also cited Yahoo’s new spam-reporting feature within Yahoo Site Explorer as evidence that the engines (and Yahoo in particular) aren’t very good at webspam detection: “they’re asking you to report the spam because they can’t find it themselves.” I was ROFLMAO when I heard that one. I doubt Yahoo would find that as funny though.

Speaking of Yahoo… Greg told the audience that he has consulted for Yahoo. Wow! That surprised me; impressed me too. Nice one, Greg!

Speaking of Yahoo again… Todd loves Yahoo’s paid inclusion program (Search Submit Pro). That’s because with it, Yahoo “gives you a complete pass on the off-page factors”. Yahoo tells everyone publicly that Search Submit Pro doesn’t improve your rankings, but it’s not true according to Todd. Although you shouldn’t expect a rankings lift with big money terms like “credit report” or “data recovery,” it will lift your rankings on most non-ultracompetitive search terms. Nonetheless, I’d exercise caution with this one. That’s because, as Todd admits, when you stop paying all your listings in the SERPs “will mysteriously vanish and it’s really hard to get them back in for free.” A bit of evil advice from Todd (I think he was only half-serious! … “Submit your competitor and then turn it off”. I was again rolling on the floor laughing at that one! I’ve heard of “Google bowling” your competitors into “Supplemental hell”, but now I guess you might call this one “Yahoo bowling!”

Attendees got a little dose of conspiracy theory. Todd and Greg don’t trust Google and their “don’t be evil” mantra — and they weren’t afraid to spread a little hysteria into the audience. Google Analytics is one service they are very suspect of. They go far as to advise you don’t use it: “giving your ROI data to the company you’re buying your ads from — that’s just assinine.” Personally, I trust Google. So much so that I’m on their waiting list for the Google Implant (beta).

Greg weighed in about the Open Directory, referring to it as “a dilapidated piece of crap.” And the ODP editors? — “they take bribes all the time.” I’d love to know who the editors are that I can bribe and how much it’ll cost!

Their recommended SEO tools were useful. For example: TouchGraph. They gave a very cool little demo of the tool in action. Todd said: “It’s like Diggswarm, but actually useful. Is Kevin Rose in the room? He’s going to kick my ass.” LOL!

Some of the other free link research tools recommended:

• Yahoo’s Site Explorer
• Netconcepts’ Link Checker
• WeBuildPages’ Neat-o Tool
• SEO for Firefox
• Tattler

Hmm. That’s funny, this list of tools (except for the last one) looked mighty familiar. So did some of the previous slides. I couldn’t resist… I got the microphone during Q&A, and I alluded to the slides, saying “Thanks for the plug for Netconcepts on that Tools slide.” Well you won’t believe what happened next! They admitted it to the audience that they lifted one of my decks from our website and then customized it! OMG!!

They praised me for making such excellent Powerpoints available online. Umm, thanks… I think! Geez!

Actually I’m not mad. Perhaps I should be, but they fessed up and poked fun at themselves for it — in front of 300 people! — so no harm done.

Just don’t do it again, k? Or if you do, at least keep my Netconcepts logo on the slides!

What other nuggets did they share? Todd shared that Endeca can be a secret weapon for SEO when you know how to make it work for you. I concur with that statement, totally. For example, with our (Netconcepts’) client Northern Tool, we were able to blow past previous search traffic records by deploying our GravityStream technology on top of Endeca Guided Navigation. Todd said that one of their (Range’s) biggest SEO successes was with Endeca. Todd gave a word of caution though: “Out of the box, Endeca is just a piece of junk for SEO”. The good news though is that, if you know what to ask for, “they will customize.”

They also recommended Google’s “Domain Park” program if you are a domainer with a bunch of domains you want to monetize.

They also advised folks to stay out of Google’s contextual advertising space so that you don’t get your AdWords campaigns onto the domain parking pages, because it is junk traffic that doesn’t convert.

What was the most interesting revelation of the day? That Todd was a Viagra webspammer before working for Range. I knew he was a reformed black hat SEO, but I never knew what industry. Greg joked that in Todd’s space, “PPC” stood for “Pills, Porn and Casinos” rather than “Pay Per Click.” Todd regaled us with tales of “churn and burn” SEO going after Viagra rankings — he could make $10k in a week, and all it cost him was a day of his time and $7 for a domain name — then the site would get banned and he’d move on to using another domain name. “Viagra” would have to be the most competitive keyword in SEO bar none (well, with the probable exception of “sex”). Todd *must* have been a “rock star” at SEO (but on the Dark Side of the Force of course! to make such a good living from it. So the name of his Webmaster Radio show really IS fitting and merited. I highly recommend downloading some audio recordings of previous SEO Rockstars episodes; even Matt Cutts has been known to listen to their show!

Rock on!

Google to own a SEM and SEO firm?!

Wow, what news eh! That Google is acquiring DoubleClick for $3.1 billion. What a windfall for the private equity firm that bought DoubleClick for a little $1.1 billion dollars in 2005.

This is no ordinary acquisition though, at least from an SEO consultant’s point-of-view. That’s because of what comes with the deal: the DoubleClick subsidiary Performics, a company that offers search engine OPTIMIZATION services (as well as paid search consulting services). Who would have predicted Google would be selling SEO consulting!

I see some potentially sticky situations ahead, like when Google/Performics employees are asked to offer their big name retailer clients assistance with ‘reputation management’ of the top 10 Google results for their brand names — particularly when there are already less-than-complementary results occupying the top spots that need to be pushed down and made less visible.

Can “Do what’s good for the user” and “Do what’s good for the client” happily coexist together? Hmmm. What do you think???

I can think of four possible scenarios ahead for Performics:

1. Google sells the Performics subsidiary
2. Google spins off Performics and somewhat “disowns” it (i.e. keeping it at an arm’s length and not allowing the SEOs and the Google engineers to fraternize)
3. Google keeps Performics and discontinues selling “natural search optimization” services, sticking only with selling paid search
4. Google keeps Performics as is and endeavors to offer the most ‘pearly white hat’ SEO consulting this industry has ever seen! (severely hampering the SEO consultants’ ability to do their job in the process)

Did I miss any other potential scenarios? (other than the deal falls through before it closes)

It should be interesting. I’d love to be a fly on the wall at the Googleplex right now.

(By way of disclosure, Performics is a partner of Netconcepts, reselling our GravityStream SEO proxy technology as Performics’ “NSO Proxy”.)