Over the past decade, I've bought and sold many millions worth of online ads. When I ran ISPcheck, I had no real answer for prospective advertisers who wanted to know what results my customers were able to achieve. And when I became responsible for RackShack/EV1's ad buys, I found that there was no easy way to measure ROI.

All I wanted to know at the time was how many visitors from TopHosts versus TheWHIR signed up. But as I've subsequently learned from Ted Smith at Peer 1, I should have been tracking customers throughout their lifecycle. If my cost per sale from Site A is 20% less than Site B, but the average account gets canceled 50% sooner, B would be a better long term investment.

A couple of weeks ago I convinced Ben Gabler at HostNine to install StatCounter, the better to look up new customers and find out where they came from, and which parts of HostNine's website they visited before deciding to sign up. (I've also used Clicktracks and Google Analytics, which provide aggregated data on visitor behavior, but don't allow you to drill down to each visitor's click path.) It just occur to me that it'd be very cool if this functionality were built into ModernBill.

Imagine being able to generate sales reports that tabulate order amounts against referring sources? Or pinpoint content on your site that's most-viewed by your most profitable new customers? Better yet, what if you could instantly compute the lifetime ROI from those $20 Google Adwords bids? Wouldn't you like to know if customers who clicked on your "cPanel hosting" ad stick around 3x longer than those who came through "cheap hosting"?

HostNine already gives all of its resellers free ModernBill licenses, and being able to automate signup/provisioning is awesome. But what if every $19.95 hosting plan came with a business intelligence system that delivers up-to-the-minute knowledge on what website copy and ad venues work? Wouldn't that be something?

AND, what if ModernBill could collect and publish aggregate, industry-wide data on how profitable TopHost-referred customers are, relative to those who came through TheWHIR? Having been on both sides of the table, I think that would really help both ad salespeople and media buyers.

I've been reading a lot about OpenID, a free, decentralized framework for managing digital identities. You start with an URI (think of it as a master username) and store your password and other creditials with an OpenID provider. You can then log into any OpenID enabled service with your URI. The service will fetch whatever credentials it needs from your identity provider.

Simon Willison wrote a great post on cool things you can do with OpenID. My favorite is restricted single sign on. Simon suggested that if everyone in an organization had "username.internal.example.org" OpenIDs, all internal apps behind the firewall could be configured to grant automatic access to such users. This eliminates the hassle of creating/deleting accounts on each service for incoming/departing employees. Couldn't GoDaddy use a similar method to apply the same logins across its many services?

I'm also intrigued by Kai Hendry's comment that maybe OpenID can be used for brokering payments. This would allow access pass holders to view content, download songs, etc across multiple sites. Not that this technology isn't already available, but OpenID would save subscribers from having to keep track of different logins for different networks.

Anyway, Microsoft announced last month that it would make Windows Cardspace interoperable with OpenID. A couple of weeks later, AOL announced its support as well. SixApart is in (in fact, OpenID creator Brad Fitzpatrick also developed LiveJournal, and is SixApart's chief architect). And Digg. And ImageShack, WordPress, Technorati, SmugMug, 37Signals...

Recently I signed up for isabel.wang.name as my OpenID through FreeYourID (which I read about on TechCrunch). If nothing else, they managed to sell me a .name domain. My question is, why isn't 1&1 in this market as well?

I keep reading about Twitter everywhere. It's a service that lets users post 140-character-or-shorter updates on what they're up to. If you scroll through the public timeline, you might not immediately conclude that Twitter is a must have business app. (A few minutes ago, xboxbetty was eating chocolate and BrianWarren was making coffee. And as you can see, I am writing a blog post about Twitter.) But as LifeHacker points out, it can be a nifty communication tool.

1. You can use Twitter as a shorthand newsletter. The example that LifeHacker gave was a video store, whose employees can post new movies now available for rental, holiday business hours, limited-time sales promo, etc. Likewise, web hosting providers can keep customers posted on network status ("some DC2 circuits scheduled for maintenance @ 2-3am") or announce new products ("this just in: quad core servers for $199!"). As Amazon's Jeff Barr puts it, the 140 character limit forces you to be brief without wasting time on formatting. Customers can stay up to date via your RSS feed, which is both more reliable and less intrusive than email.

2. Twitter could also be useful for managers who are responsible for scheduling sales or tech support coverage, especially if your reps are scattered across multiple locations. You can subscribe to a feed that aggregates your entire team's real time updates ("running 15 minutes late", "leaving for meeting, back @ 3pm", etc), which they can post via SMS (or a web browser, or IM or an internal app that you integrate with Twitter's API).

3. Lastly, LifeHacker recommends Twitter as a to-do list, but Anne Zelenka says it might be better as a "already done" list. For instance, if you and your two partners are traveling to three different cities, you can use your aggregated Twitter stream to keep each other informed on important events that don't require follow-up action ("contract signed with Company X", "new batch of switches will be delivered next Tuesday", etc). It reduces email clutter and is more flexible than IM; each member of the group can catch up at his convenience.

Technorati, CNN, BBC Video and Google News all have Twitter channels; shouldn't you? It's free, it's super convenient - and according to Kathy Sierra's "Asymptotic Twitter Curve" (which I came across via David Terrar's blog), it might help you capture a little more of your audience's attention.

Between Jan 2 and Feb 7, TheWHIR published 23 announcements of shared hosting price reductions. For instance...

* HostMySite, Hostway, Sibername, MyDomain, 4Domains and Web World Ireland are among those who cut domain prices.

* AIT, HostICan , LunarPages , NetPivotal, CWCS, and Netazen increased bandwidth and/or storage allocations.

* Webair, Hosted and 1&1 ran half-price promotions.

Many of these announcement mention a desire to make it more affordable for small businesses to maintain a web presence. I don't know about you, but I've never come across a company that held off on building a website because it can't come up with $5 or $10 a month.

And of course, all of these promotions are intended to enhance competitiveness. But as CrystalTech's Robert Cichon puts it, "survival will be brutal, marketing impossible and customer acquisition badly flawed" if one's only sales propositions are price and resource allocations.

I think the average small business owner is probably shorter on time than web hosting budget, which is why I feel like BT might be onto something with BT Tradespace. (As David Terrar reports, this is BT's third small business initiative, following their SaaS marketplace and online collaboration service. BT also offers traditional website/ecommerce hosting.)

Here is an example of a BT Tradespace. The site owner can provide a company description, upload photos and post news updates via a web-based interface; no FTP client or HTML knowledge required. Visitors can rate the company, map its location, request an appointment, or contact it via email or VoIP. As David points out, it delivers a lot of usefulness for very little effort.

No, BT Tradespace's functionality won't be enough for everyone. Microsoft's Windows Live Spaces has a relatively limited feature set as well, but it signed up 650,000 new users in January alone! The trick to bringing new users into the market, I think, is to lower the entry barrier - not in terms of monthly fee, but time investment required.

PS - BT Tradespace and BT's "Workspace" collaboration service are both powered by SMBLive. Telus is another SMBLive partner.

Ok, check out this graph. Blue represents Google search volume for "wiki", red is for "web hosting".

I decided to look this up on Google Trends after seeing that "wiki" is one of Google's top 10 search terms for 2006. Wiki is also #2 on the top how-to's

Also interestingly, a friend who works at the Department of Labor mentioned that he's just attended an information session on the Bureau of Labor Statistics' upcoming wiki. He's so excited about the idea of web-based information sharing that he wants to sign up for SocialText and start collaborating within his group immediately.

Update: Just found out via Techcrunch that Zoho has launched their hosted wiki. It looks awesome: along with all the usual wiki features like edit history tracking and reversion to previous versions, you can insert graphs/spreadsheets/slideshows/forms from Zoho Office, as well as videos/photos from third party sites.

Zoho's other apps include word processor, spreadsheet, presentation, file sharing, project management, CRM, calendar, database, chat and polling. A couple of months ago, the company implemented single sign-on across the entire suite. Zoho is hosted at Savvis, on internally developed grid infrastructure. They even offer a application monitoring service that their sysadmins themselves use.

Read/Write Web calls web office products like Zoho one of 2006's major technology trends. (And indeed, Zoho has more than 150,0000 users - which is 50,000 more than when I first heard about the company in October.) Might there be an opportunity in partnering with Zoho to offer your hosting customers a web office solution?

When I was first hired by EV1Servers, I sent then-CEO Robert Marsh a loooong wish list for internal apps. One item near the top was a dynamic customer map. If I were going to a conference in Chicago or visiting friends in Portland, wouldn't it be cool to find out which major customers are nearby? If 50 companies from a certain city lease 1000 total servers from EV1, shouldn't we fly out there to meet with these folks - and see if they have contacts who might be interested in our services?

There was no Google Maps API back then, so building such a tool didn't seem feasible. But now that there's a Flickr layer and a Wikipedia layer on Google Earth (I read about those in this O'Reilly Radar post) and map mashups for everything from apartment rents to handmade jewelry, why not do the same for your customer base?

You could use an internal map that shows customer count, server count, support ticket volume, recent orders and cancellations... by location. In addition to identifying sales opportunities, the map could also help set content localization priorities. If customers in China seem to submit an unusual number of tickets for very basic issues, maybe they'd appreciate a Chinese FAQ.

More interestingly, you could offer customers the (optional, of course) ability to put themselves on a public map. For instance, Etsy, an arts and crafts marketplace, has a geolocator where buyers can find local sellers. Your customer map could have an ecommerce layer for retailers, a web services layer for developers and designers, a local vendors layer for doctors, lawyers and plumbers...

I'm guessing there are hundreds - if not thousands - of GoDaddy and 1&1 and DreamHost and BlueHost customers in just about any major city. It costs very little to encourage these companies to do business with one another, but hosting providers could earn major brownie points (and longer-lasting customers relationships) by facilitating these connections.

PS - On a totally unrelated note, I just read on Wired that there's a Santa tracking layer on Google Earth.

Niel Robertson from Newmerix wrote a very insightful blog post last week about inversion as the cornerstone of the Web 2.0 revolution. (Niel used to be VP Research at Exodus.)

A successful web app, Niel says, looks like ZingFu, which puts a silly frame around photos that users upload. ZingFu offers about 0.1% of Photoshop's functionality, but it's instantly accessible to 99.9% of the population. In contrast, the other 99.9% of Photoshop's features might only be understood by less than than 0.1% of the population.

Following Niel's logic, a successful web hosting service might look like MySpace or YouTube or Flickr. Each of these sites offers a tiny % of the average web hosting package's functionality, but their simplicity appeals to an *enormous* audience. In contrast, the other 99.9% of web hosting features might not be readily accessible by the general public. As a point of reference, none of my friends outside the industry have any idea what FTP is.

My favorite simplified web hosting case study is PhotoBucket, a company founded by Alex Welch, a former Level 3 exec. Wired reported a couple of weeks ago that PhotoBucket passed the 30 million user mark, and is signing up 80,000 new users per day. According to PhotoBucket's official blog, 30 million is also the number of (a) dog owners in the US, and (b) hot dogs consumed in baseball parks. Not coincidentally, neither activity requires HTML knowledge :)

Google co-founder Sergey Brin says that success will come from simplicity. Nicholas Carr calls simplicity the new spaghetti and tries to consolidate Google's many, many products into 5 categories (web presence, web office, content sharing, ad platform and search platform) - because 5 is the magic number; Eric Schmidt says that's as many as most people can remember.

As Brin puts it, "if we continue to develop so many new individual products that are all in their assorted silos, you will have to essentially search for our products before you can even use them... I would rather have a smaller set of products that have a shared set of features."

This reminded me of my recent visit to 1&1's website. There are *lots* of features on this chart; I counted to 50 before losing track. That ain't simplicity! iPowerweb has a similarly long list, as does LunarPages. (LunarPages has a great blog, by the way. I found it about it from Ben Welch-Bolen at ResellerGuide.)

Yes, all of these companies are very successful - but isn't Google? If Google thinks it's important to build easier to understand products with more tightly integrated features, might they be on to something?

For instance, many hosting companies have photo albums that don't integrate with blogs, shopping carts that don't have similar-looking templates to site builders, web stats software that isn't set up to track results from SEO and email marketing tools... As an end user, I'd rather have a seamless feature set than another 1000 GB of bandwidth I'll never use. Wouldn't you?

Update: Jon Udell from InfoWorld talks about Clayton Christensen-style disruption in his latest blog post. Google's office apps are collaboratively adept and functionally lame, he says. And Microsoft's apps are adept and lame in the opposite ways. He cites a reader's example of Google Docs adoption within her company and among her customers:

"We started using them around the office for convenience and they've taken off. We found the version control, collaboration and invite system outweighed the limited feature set. For the most part, they have the very basic functionality covered."

In other words, convenience outweighs a large feature set. If Google Docs can take on Word and Excel, might the also-simple Google Pages be a contender against your hosting plan?

I've been thinking about Dan Golding's comment on my post from yesterday. If classic shared hosting is an endangered species, what's next?

Many people I know would disagree that shared hosting's days are numbered. After all, customers are continuing to sign up. And while they aren't happy about churn, they think the solution is more consumer education and better customer service. But are informative newsletters and fast trouble ticket response times enough to save the day?

I think the answer is in this recent MarketWatch article. Bambi Francisco points out that Lycos, Excite and Alta Vista were once among the highest traffic sites on the web. Their portals generated huge amounts of revenue. Back in 2000, advertisers spent $3.8 billion on display ads, versus just $109 million on paid search. But fast forward a few years and paid search dominates. Google alone will have sold $7 billion+ in keyword ads in 2006.

Has the fortune of display ads declined because customers weren't aware of the benefits of online advertising? Or did ad reps from Lycos, Excite and Alta Vista drive them away with poor service? I think the answer is neither. Instead, Bambi's explanation is that Google has closed the audience/revenue gap.

Earlier search engines attracted enormous crowds way before Google was on anyone's radar screen - but they didn't monetize this particular type of traffic. Google, on the other hand, has managed to shape its product (Adwords) around consumer behavior (search).

Last night a friend and I were looking at 1&1's hosting plans. She's interested in building a website, and she thinks it'd be cool to have Live Chat. Unfortunately, that's only available with 1&1's larger hosting plan, which includes 100 GB storage and 1000 GB bandwidth. "Why can't I buy the features I want separately instead of paying for capacity I'll never use?" she asked. "It's like a leasing agent telling someone that if they'd like a sofa in their office, they have to rent 100,000 square feet of space."

[She doesn't know that hosting providers oversell bandwidth/disk space. She thinks if you sign up for a 100 GB plan, you get a dedicated "virtual hard drive" with that amount of space.]

I thought of her real estate analogy when I read today's MarketWatch article on Cyworld. Cyworld offers free web space, in the form of 3D "minihomes". It sells $300,000/day worth of virtual home decor: sofas, dining tables, paintings, etc. The company plans to offer branded virtual items from Gucci, Apple and Nike, soon. MarketWatch calls its approach a logical evolution in social networking, because it allows us to express ourselves through what we collect.

Might web hosting be headed in the same direction? The bandwidth + disk space formula clearly doesn't reflect many customer's views of their needs. The question is, what would be the equivalent of Google's search and Cyworld's virtual iPods in web hosting? In addition to storing and transferring data, what common and popular customer desires and activities can we create products around?

You know how Flickr and YouTube and del.icio.us let users tag the photos/videos/links they submit? Why can't web hosting providers do the same for customer websites?

What for, you ask? I once spent a whole weekend looking up every incoming EV1Servers customer's website. I wanted to experiment with customizing keyword ads and landing page text for major customer categories instead of bidding on "web hosting" like everyone else. And I thought I might learn something if I cross-tabulated different types of customers against what they bought. In the end, I realized this wasn't a 2-day project. To really see the big picture, I'd have to continuously monitor and classify new signups, and I didn't have time for that.

I'm still convinced that there's more where they came from - they being your customers, whoever they are. And guess what? You could crowdsource the kind of research I spent hours and hours on to customers themselves. Just ask them what keywords they'd associate with their websites. Not only could you use this info for marketing and product development, you'd be able to share all sorts of interesting aggregated stats: 62% of cat herders prefer Linux, perhaps, while 74% of basket weavers choose your premium versus economy hosting plan. You could also put up a tag cloud to reassure prospects that you host many companies just like theirs. Or publish an opt-in list of customer websites associated with each tag.

In a recent blog post, Harvard Business School professor Andrew McAfee writes about channels versus platforms for communication. Traditional web host-customer relationships are private channels; interested third parties (such as prospects, or even your marketing department) don't benefit at all from the two sides' interactions. Flickr/YouTube/del.icio.us, on the other hand, are platforms which support the accumulation of knowledge - knowledge that allows you to connect more efficiently with prospects. Doesn't that sound like a good thing?