Jason Putorti –lead designer for Mint.com, the first personal finance application I’ve actually been able to maintain an interest in–did this great presentation on “10 Things CEOs need to know about Design”.

Great if you build, design or market web applications.

Love this quote:

“The best marketing tool you can have is a well designed application”

That is the absolute truth. I feel like I need to watch this presentation every monday just to get my mind focused for the week. So good.

View it on slideshare here

Great content, good mix of web app, open source topics, and just interesting discussion if you are that certain kind of web geek into that sort of thing — like me.

You can follow them on twitter here: http://twitter.com/devshow
And, here’s an iTunes link for The Dev Show.

Here’s my iphone/podcast listening tip (if you can call it that) : Listen to your podcasts at doublespeed. Twice the podcast goodness in half the time! It only takes a minute to get so used to the speed that it sounds normal.

And, here’s tip #2. Don’t do this while listening to a podcast:

Here’s the official Google AdWords blog post on the subject.

We’ve used a number of lead aggregators — companies that essentially sell quote requests to a pool of potential service providers. BuyerZone, and Resource Nation are two examples. These companies typically derive much of their traffic and leads from natural and sponsored search.

Often, the math works in their favor, allowing them to out-compete individual service providers in sponsored search. For example:

Say the Service Provider has a CPA (cost-per-acquisition) target of $75 for their search leads. That’s a level where they know they will be profitable. If their site converts 5% of search traffic to a lead overall, they could pay ~ $3.75/click in sponsored search.

The Aggregator however will take the leads they generate and re-sell them to anywhere between 3-7 individual Service Providers for typically $15 to $35. If they can sell these leads to six providers at $30ea, they’ll be able out-bid those same providers.

6 x $30 = $180 revenue/lead
assume 100% gross revenue return target, and 5% conversion rate

The aggregator can spend $4.50/click, while expecting to earn $9/click. To the Service Provider they are still getting leads significantly below their $75 target.

The Aggregators also often out-convert Service Providers because they have such a clear and simple goal and benefit proposition — complete this form and you’ll get six quotes from qualified providers.

You can see the impact conversion rates (and closing rates) have on the economics of this situation. It may be in your best interests as a service provider to pay a little more for that exclusive lead, knowing that you’ll close a larger portion of them.

Now, it looks like Google is going to get into the lead gen process themselves–which has the potential to totally disrupt the existing market, with the greatest potential losers being the Lead Aggregators.

Should be interesting to watch.

I was reading this post yesterday, and it made me want to estimate how much the GeoCities sites were really costing Yahoo.

I tried to make some order-of-magnitude estimates of how much the hosting of these domains was costing Yahoo. Was it really a cash drain?

Hosting:
7.5 million hosted sites.
5 Megabytes of data/site (this is probably way high)

That’s 37.5 TeraBytes of data to store.
Using the highest tier Amazon S3 pricing (we know it would be cheaper than this) of $0.15/Gigabyte, that’s $5,625 in storage/month.

Bandwidth:
10 MM Visits/month.
Assume an average of 3.5 pages/visit (probably high).
Assume 100k transferred/page (also probably high).

That’s 3,500GB transferred/month.

Again using the highest S3 data transfer price of $0.17/GB, we’re looking at $595/month.

Add the per/request fees of $0.012/1000 requests. We’ll assume there were 15 objects/page, 35 million page views, that’s another $6,300 in request fees.

So, looking at storage and bandwith costs of serving the GeoCities requests we’re talking about somewhere on the order of $12,520/month. That doesn’t seem like a lot of money just to keep the geocities pages alive. To cover those costs you’d need to generate just 3.6 cents out of every 1,000 pageviews! It seems like some contextual links could have covered that?

I’m making a lot of assumptions, and to be conservative I’ve tried to estimate on the high side of things. They’ll also need to maintain some of their own infrastructure, or outsource the actual serving of the sites–but I’d think they probably have that down to a science. Yahoo would also likely need at least some staff to manage and support the service, and some staff to plan and manage marketing campaigns to your GeoCities audience. These salaries would far outweigh actual hosting costs—but still, seems like they could have made it work.

Ways they could have made money from that traffic:

1. Cross Sell other Yahoo! properties from the GeoCities pages. Use a small top-of page banner of some kind.
2. Serve their own PPC contextual ads on the Geocities pages
3. Remarket Yahoo’s hosting services to the Geocities site owners (they made a small effort in this regard at the end, but you could continue to market to your 7.5 million site owners month after month.
5. Add AdSense to the GeoCities sites — Heh, heh, just kidding but it would work

So, what do you think? What am I missing?

In the end they probably just decided the revenue upside was just not worth the effort. But it makes me wonder what other Yahoo! properties are in exactly the same situation.

Four years ago IAC bought Ask Jeeves for stock worth 1.85 billion (at the time). But, since that IAC stock is now worth about 30% of its value back then that would place the deal’s present value at about 550 million—still way, way too much. Now it seems IAC is ready to put Ask (minus the Jeeves) on the block.

We were Ask Jeeves advertisers once upon a time. Back when they had their own ad management system instead of the horrible outsourced Looksmart clone they implemented in 2006—a decision I’ll never understand.

Back there in late 2005 there did seem to be a glimmer of hope that Ask Jeeves may grow their 3% search share to something bigger. They had a memorable brand with the Jeeves butler. But, IAC showed him the door in a rush to turn them into the next ho-hum online destination.

The butler didn’t fit with their rush to be just like Google. See, they’d upgraded their search technology so you didn’t have to do your searches in the form of a question. But maybe asking a question to the butler was the only fun thing about the site! I mean he was in the Macy’s day parade. Come on!

Interesting Utah factoid, one-time super-star Utah agency DSW created the original Jeeves brand (along with Intel Inside among others). DSW also gave COGBOX one of our very first jobs back in ‘97.

Good luck unloading Ask.com. I can’t imagine who would line up to buy it.
You had potential at one time, and it was squandered.

Here’s more on the topic.

Fast Company pulled all the videos of their other robots together here.

The company is called Boston Dynamics.

That first day also marks the zenith of banner ad click through rates. A banner with just two words, “Click Here”, achieved a 78% click-through rate. Each click was followed by the user sighing and saying “oh, it’s just an ad”.

Here’s an interesting account by Frank D’Angelo at AdAge.

Yesterday, while working on a COGBOX.com refresh my up-till-now trusty iMac suddenly froze. And, on restart I saw the dreaded flashing-question-mark-folder screen—not good.

While my iMac is still not back in service, I’m back in action with a combination of my laptop and other office machines.

There are a few choices that have made this hard drive crash much, much less painful than it could have been. I thought I’d post them here to give credit to the products/services that have helped. Maybe my experience will help you.

Hosted Subversion Repositories – Beanstalk
We use Subversion for version control on all our client sites, and we host our Subversion repositories off site with Beanstalk. Beyond all the other many benefits of version control, hosting at beanstalk makes picking up development right where I left off super-fast. Three other development tools we use that are subversion friendly : Versions, BBEdit, and Coda.

Google Apps – Gmail, Calendar and Tasks
About two years ago we moved all our cogbox.com email over to Gmail for Business. It would be impossible to go back. Besides the amazing speed with which I can search through 40,000 messages in my inbox, having our email as a web service means no hiccups in the case of disasters like yesterday’s. Just lately I’ve started using Tasks from in Gmail and really like it, easy fast to-do lists (they look good on the iphone too).

So I’ve got my client’s sites & development projects, my email/calendar and my to-do lists. What about other work files?

Time Machine
Fortunately, Apple came through for me. I’d stopped paying attention to whether my backups were working months ago. But Time Machine was there in the background storing my files every hour.

My only regret is excluding some of my system files and applications from my Time Machine backups to save space. When I do get my iMac back in working order I’ll have to spend some time re-installing a bunch of apps. What I should have done (and will be doing in the future) is let time machine do a whole system backup to a larger network storage device like a Drobo. Then my recovery would be as simple as a hard drive swap and restore from Time Machine.

All-in-all, relatively painless. I’ll update my post as my recovery proceeds.

Update:
I forgot one other app that has made my life much easier with this crash —Yojimbo. It’s another app from the makers of BBEdit that allows me to store (and encrypt) thousands of logins, notes, code snippets and other bits of information, then sync it between multiple machines. With all the mental space I’ve freed up with this app I can now sometimes even remember where I’ve left my keys!

Update 2:
I found the difference between an Apple HD replacement and a local Mac Reseller (MacDocs) HD replacement is about $150 and 1 TeraByte. Apple Store: Replace my 500 GB drive w/another just like it – $350. MacDocs: Replace with 1.5 TeraByte drive $200.

(Normally something as simple as a HD replacement is something I’d do myself, but the Aluminum iMac replacement process involves suction cups and major dis-assembly—well worth avoiding for the $75 mac docs fee.)

Search Submit allowed you to pay a fee to guarantee inclusion of your pages in the Yahoo Search index. We actually used that service years ago. It was a convenient way to ensure that you had new pages indexed quickly, and that large sites like ecommerce sites were fully indexed.

These days failure to get your site fully indexed has more to do with poor site design than with failure to cough up some extra cash to Yahoo. And, for commerce sites there is always Yahoo Product Submit.

The program will end at the end of this year. Buh bye.

You can nominate a location here.

From the size, it looks like the street view trike would be a thigh-burner:

Here’s one:

This chart shows how the value of on-page optimization declines as the level of competition in the search results increases, and that at a certain point a SE will begin relying more heavily on inbound links to determine position than on on-page/site factors to determine position. This jives with our own experience with the relative importance of these tactics.

Several other good charts in the full article here.

According to the article over the course of 3 hrs the story went from:

1. Entertainment Niche Site
2. Larger Entertainment Site
3. Wikipedia
4. CNN Twitter
5. first Bit.ly link (others tweeting)
6. Story hits Digg front page
7. massive RTing
8. MSNBC confirms
9. CNN confirms
10. Story on Google News

The story really took fire on social media, with the MSM and Google News following later. I don’t think a 3 hr delay is too bad on Google News’ part, but it is interesting to see how quickly stories like this can spread now.

Read the rest here

http://www.seobook.com/worthless-hype

More than a few SEOs out there have claimed that their proprietary testing processes have shown huge benefits could come from manipulating Google’s “link juice” with nofollow tags. I mean, they’ve tested it, and so they should know right? Well, turns out those SEOs may need to learn how to test.

Now, we do SEO, (and have since man walked on four legs), but we’ve always maintained that most of it is more common sense than secret sauce. But, we’ve seen the lure that SEO salesmen have on clients when they start talking about dark arts like page rank sculpting with nofollow tags.

The author, Aaron Wall–who I would love to buy a beer (or coke)–has a couple of theories why SEOs may have perpetuated this nofollow myth:

1. It makes them look cutting edge and allows them to sell more services.
2. Writing about things which are new, uncertain, and untested yields links.
3. For every person who is an SEO expert there are 1,000 ditto-heads linking to whatever sounds new or important.

You know, that sounds just about right to me.

http://www.getpersonas.com/persona/31913

To install:

Go here: http://www.getpersonas.com/
and click Get Personas for Firefox.

Then go here: http://www.getpersonas.com/persona/31913
and click “Wear this”.

That’s it.

They asked 30,000 web types their opinions on topics from their “level of excitement” to “how long since your last raise” for this year’s survey. And, if you are so inclined you can even download a copy of the raw data yourself to play with.

Go to the survey.

It’s called Coda, by Panic Software — the same folks who make the very nice Transmit FTP client for Mac — and it combines a bunch of tasks that a developer on a mac will do over and over.

From within a single app, you have:

A text editor that rivals BBedit
Integrated SVN version control
A webkit powered preview mode — basically Safari built right in.
An FTP Client
A Terminal shell window
A library of programming reference docs
A CSS editor

So instead of jumping back and forth between applications while you’re developing, it’s all right there for you!

For a lot of our clients the process of an update used to look something like this:

1. Open BBedit or Dreamweaver.
2. Open the appropriate docs.
3. Make your changes.
4. Save your changes, switch to FireFox or Safari to preview
5. Repeat steps 3-4 until satisfied.
6. Switch to terminal, navigate to the site, and commit the changes in to the repository (or switch to Versions, another SVN GUI app we’ve been using).
7. SSH in to the remote server and check out the changes to the working copy.

In Coda, this goes something like:

1. Open Coda, click on the site.
2. Click the doc you want to open in editing mode.
3. Make changes, preview in preview pane.
4. Click the Source Control Status button to show your SVN status. One more click checks in your changes.
5. Click terminal tab, you’re automatically SSH’ed to the server. Just update the working copy, and you’re done.

Same work completed, much less time and hassle! See it for yourself.

“To give advertisers more control over their ads, we plan to remove a limited number of keyword mappings on July 29, 2008. “

Match Driver is their system for matching synonyms and plurals of certain phrases together into one bidding group. For instance, if you wanted to bid on the keyword “timeclock” and “time clocks” on YSM, they are lumped together by Match Driver. The system was first introduced by Overture way back in 2002 with the claim that it would help advertisers drive more traffic–by forcing them to bid on synonyms that they may not have wanted in the first place.

The problem with Match Driver is that it has always been a little bit of a black box. Instead of having the ability to choose an easily understood match type like “exact match” (where your ad will only be triggered by a search for exactly your defined key word/phrase) the best you can do with Yahoo is to use their “Standard” match –which will still be matched to multiple phrases through match driver.

This causes all kinds of issues when managing keyword sets across multiple networks as well. For instance, say you have a set of five phrases in Google Adwords, each with their own destination URL, match type, and bid. If you take these phrases and import them into the YSM system they will only accept one of them, and reject the others as duplicates. The one version they accept will be the the one you submitted first, so now you’ll be targeting all five phrases with the bid, and destination URL intended for just one. And, since you didn’t know their match list prior to submission, it is quite difficult to know precisely which of your submissions will be the accepted version.

You can look at their list of unmapped phrases here:
http://searchmarketing.yahoo.com/developer/index.php

What strikes me about the list is that they are not so much unmapping phrases, as mapping them to a new cononicalized version. For instance, they are taking this set of phrases :

history and automobile
history and of and the and automobile
history automobile
history of an automobile
history of automobile
history of the automobile
history on automobile
history on the automobile

and matching them to the phrase “automobile history” instead of “car history”. Not a huge change in my opinion.

When you search Yahoo for “automobile history” and “history of the automobile“, you’ll get the same set of advertisers — many of whom are selling used car history reports. So, is it in these advertiser’s best interests to force them to use the same bid, ad copy, and destination URL for those two searches? Certainly “history of the automobile” converts at a much, much lower rate. And, notice that on AdWords, where advertisers have more control, those advertisers avoid “history of the automobile“, while they do appear for the search “automobile history.

Rather than tweak Match Driver, I’ve got a much better idea. Get rid of it.

“lexpionage, the sleuthing of new words and phrases. These aren’t “stunt words” or “sniglets,” but new terms that have appeared multiple times in newspapers, magazines, books, Web sites, and other recorded sources.”

Here’s one that I’ve experienced myself at times:

semisomnia n. Mild but chronic exhaustion caused by poor sleep habits.

More here: Word Spy

Pretty cool, and something I think I’ll actually use.

This is backed by Sequoia Capital, who’ve previously been involved with little web companies like Google, YouTube, Yahoo.

As a side note, one of the first speakers we saw after starting COGBOX in 1997 was Mark Kvamme, now one of the partners of Sequoia. This was at an AIGA brand design conference, and really got us excited about our new business. Our paths have remained almost eerily synchronous ever since. He went on to fund Google, we…ok, maybe not perfectly synchronous.