Folgende Maßnahmen werden erwartet:

1. Google Analytics tracking code um die Funktion AnonymizeIP erweitern
2. Schriftlichen Vertrag mit Google zur Auftragsdatenbearbeitung abschließen
3. Datenschutzerklärung der Website ergänzen
4. Alte Google Analytics Profile löschen

Warum das alles?

Ich bin sehr besorgt um meine Privatsphäre im Internet. In all meinen Web-Browsern sind 3rd party cookies deaktiviert, ich lösche ohnehin alle Cookies regelmäßig, verwende AdBlocker und NoScript, und ich habe viel Zeit mit den Facebook-Einstellungen verbracht, nachdem mein Profilbild ohne meine Zustimmung in einer Werbung im Online-Telefonbuch erschien.

Ich bin trotzdem der Meinung, dass die o.g. Maßnahmen wenig Sinn machen. Fast schon grotesk.

Es geht vornehmlich um die Anonymisierung der Benutzer IP-Adresse, weil IP-Adressen einen Benutzer persönlich identifizierbar machen und auch der Standort des Benutzers ersichtlich ist. Ich werde deshalb lediglich auf den ersten der vier genannten Punkte eingehen. Wer allgemeine Kritik auch zu den anderen Punkten sucht sei auf den Internet-Law Blog verwiesen.

Mit der AnonymizeIP Funktion im tracking code soll eine „Anonymisierung“ der IP-Adresse gewährleistet werden.

Dazu muss man folgendes wissen:

1. IP-Adressen können nur vom Internet Service Provider (ISP) einer Person zugeordnet werden. Firmen wie 1&1 werden Google auf Anfrage sicher nicht den Namen ihrer Kunden mitteilen. Da müsste Google schon selbst der ISP sein, das käme aber für mich genau aus diesem Grund sicher nicht in Frage.

2. IP-Adressen identifizieren das client-Ende einer TCP/IP Verbindung. Das ist heute i.d.R. ein Computer oder ein (DSL-)Router. Hinter einem Router können sich viele verschiedene Computer verbergen, und jeder davon kann von mehreren Person verwendet werden. Wer meine IP auswertet, wertet demnach das durchschnittliche online-Verhalten aller Personen in meinem Haushalt aus, nicht mich persönlich. Auf meiner Arbeitsstelle teile ich die öffentliche IP Adresse unseres Proxy-Servers ohnehin mit mehr als 1000 Kollegen, also auch hier kein „personal tracking“.

3. Es ist technisch nicht möglich, die IP Adresse vor der Übertragung an Google zu verändern. AnonymizeIP sorgt lediglich dafür, dass der tracking code Google auf die erforderliche Anonymisierung hinweist. Google hat sich verpflichtet, in diesem Fall das letzte Octet der IP Adresse vor der weiteren Verarbeitung der Daten (bereits in Europa) zu entfernen. Damit kommen 255 verschiedene IPs in Frage. Data mining Spezialisten können mit einer verkürzten IP zwar weniger anfangen, nutzlos ist sie dennoch nicht. Zusammen mit anderen tracking Parametern kann ein anonymes Benutzerprofil relativ einfach erstellt werden. Für den Versuch, diese Daten einem bestimmten Computer oder Router zuzuordnen, sollte die Zahl 255 keine allzu große Hürde sein (Stichwort EverCookie).

4. Google verwendet die IP Adresse für die (mehr oder weniger ungenaue) Bestimmung des Ortes, von dem aus der Benutzer eine Website verwendet hat. Diese Orte werden in den Statistiken angegeben. Das letzte Octet zu entfernen, wird die Genauigkeit der Ortsangabe in vielen Fällen gar nicht verschlechtern. Es wird ohnehin nicht der Standort des Benutzers ermittelt, sondern der seines ISP, und der ist für alle 255 anonymisierten Benutzer gleich.

5. Viele Benutzer haben wechselnde (dynamische) IP-Adressen. Meine private öffentliche IP Adresse ist die eines 1&1 Servers. Sie wechselt täglich, und das nicht nur innerhalt des letzten Octets.

Mit anderen Worten… viel Wirbel um Nichts. Ich sehe jedenfalls meine Privatsphäre auch dann nicht in Gefahr, wenn Google meine vollständige IP-Adresse erhält. Zum Glück hat dieser Auswirkungen keine unmittelbaren Auswirkungen auf die Funktionalität und Benutzerfreundlichkeit einer Website. Lediglich die Datenschutz-Informationen werden nochmals komplexer und umfangreicher, was lediglich dazu führen wird, dass sie noch seltener gelesen werden.

A new model

Most presentations were oriented around the same general theme. The traditional pharma sales model is changing quickly and dramatically. Some US presenters even stated that it is about to collapse because many physicians don’t (want to) see reps anymore.

Less blockbuster drugs, shrinking product differentiation, and ageing product portfolios force companies to adjust their marketing approach, in particular:

• closing the loop,
• moving from push marketing to dialogue marketing,
• recognizing the value of the “long tail”.

Many of the marketing terms used during the conference reflected these trends. Here are some of them:

• Closed Loop Marketing (CLM)
• Relationship Marketing
• 360° view
• Micromarketing
• Multichannel marketing

The underlying concept for all these terms is the same. Doctors spend less time on listening to elaborate product detailing presented by sales reps. Instead, they prefer on-demand information tailored to their specific needs and presented to them at the time and place of their choosing. Online media in particular have enabled them to be in control of all important aspects of communication, the what, the how, the when, and the where. A recent Manhattan Research study shows that in the EU 85% of physicians want online product information. They prefer a mix of channels, including email, web, and traditional offline communication.

New marketing skills

All big pharma companies are in the process of adapting their marketing communication to their customer’s preferences. The online medium offers great tools to customize marketing communication for both doctors and marketers. eMarketing concepts and tools therefore are an integral part of pharma’s new marketing communication strategy. This also affects the brand marketer’s daily work. Offline-online marketing integration is essential. Understanding the Internet and how eMarketing can be added to the mix was identified as being a key skill for advanced pharma marketers. More and more job titles on the delegate lists of marketing conferences express this trend. Taken from this event’s list:

• Multi Channel Marketing Manager
• Director Integrated Business
• Marketing Excellence Manager
• CRM and eBusiness Manager

The delegate list also contained a large number of marketing roles without an “e”, so brand managers want to keep up.

Role of the sales force

Several presentations addressed the role of the sales force in this process. Reps are still considered a very important part of the relationship building process. They are needed for closing the loop, so they must support CRM. A successful sales force uses all available customer information to deliver the right messages, and they feed new insights back into the system. eMarketing tactics like email marketing or online meetings can be made more effective if sales reps participate.

Conference take-home message

The pharma marketing model continues to transform from product-oriented “spray and pray” push marketing to integrated relationship marketing involving time and place shifting. Online communication tools have some unique features to support the new model, and therefore play an ever increasing role in this process.

Internal policies and procedures need to be established to make sure that relevant website standards are defined, implemented and maintained. For new websites some level of standards compliance is usually applied today. Finding a budget for making existing sites compliant can be challenging though. The site may not look much different afterwards.

“So what did you need all that money for?”

Don’t answer this question by just listing insider terms such as brand consistency, usability, browser compatibility, accessibility, W3C compliance, or search engine optimization. If you have read PAS 124, your answer around these terms will be much more elaborate and meaningful. The document explains what web standards are, why they are important, how they save money in the long term, and why they improve the quality of your sites to provide a better user experience.

PAS 124 is not a comprehensive technical reference for the standards themselves. Please refer to other resources if you want to learn more about website accessibility or semantic HTML.

A free summary of PAS 124 can be downloaded from the sponsor’s website. You can purchase the full document at BSi’s web shop.

Web standards are crucial, make sure that your web agency and internal staff are familiar with them.

Looking at the actual numbers, mobile web access is only about 0.3%, so still plays a minor role. However, the Net Applications data support common belief that mobile devices will be widely used to surf the web when they provide a better browsing experience.

It is estimated that 3 billion mobile phones are in use today, and only about 1 billion PCs. Yet almost all web browsing is done on PCs. Why?

The conventional 1.5 x 2 inch screen on most mobile devices is just too small. According to Mobref.com, more than 60% have a horizontal screen resolution of 176px or 128px, or even less. Most web sites do not offer a mobile interface. Viewing such sites on a 1.5” wide screen requires lots of scrolling. Unless a phone is 3G, data transfer is slow. Even with an iPhone which is not a 3G phone yet, surfing the web is reported to be painfully slow using At&T’s mobile connection (EDGE). Only with its WiFi option, the iPhone provides fast enough web browsing.

Phones will become more powerful, with faster access, larger screens, and better web browsers. Mobile Internet access will become cheaper, too. I strongly believe that despite current low mobile access stats, web site owners should no longer hesitate to build a site version that supports mobile devices. If your site is not web standards compliant yet, make XHTML plus CSS your first priority. Unless your site targets a mobile audience, there is no need to have a tailored version for all of the 50+ different mobile browser models out there today. One style sheet that optimizes the site for mobile access can easily be implemented and for most sites would make a huge difference already.

The W3C has published mobile web best practices. Testing your site with Ready.mobi will reveal key areas for improvement.

Not many mobile phones in use today offer decent internet browsing like Apple’s iPhone or recent Nokia models. Surfing the web on an iPhone is easy because it uses a full featured Safari browser and can flip to portrait view when the phone is turned 90 degrees (accelerometer).

A significant part of all web surfing will be done on mobile devices only if and when:

1. most important web sites have (CSS driven) mobile interfaces
2. mobile devices have larger displays and regular web browsers (good bye WAP)
3. prices for 3G phones come down to $200-250
4. high speed mobile internet access at affordable flat rates

Most web site owners today are not willing to invest lots of recources into building a number of tailored mobile web interfaces that suit different mobile devices. When mobile devices have standards based web browsers  there will be a need for only one such interface, if the site’s main interface is not sufficiently compatible with mobile phones already.

Thanks to mofuse.com this blog has a mobile interface, too: mobi.soldierer.com

t.b.c.

<span class=”fn”>Walter Soldierer</span>

How do you tell if a web page uses microformats?

You can’t tell by just looking at the page. You either need to view the html source code or run a tool to inspect the page. Check this blog’s About page, for example. There is structured address info, but even though it is heavily microformatted as a hCard, it does not look special.

So what are microformats good for anyway?

Microformats allow programs to categorize the information on a web page, find related content, and offer useful extensions. Here are some typical examples for the 3 fundamental data types that microformats are most used for:

WHO?

By exporting a vCard, address information on a web page can be added to program’s list of contacts.

WHEN?

Event information can be added to calendar applications.

WHERE?

Location information (city, street) can be used to display a map or directions.

When microformats are widely used, you no longer register an event at upcoming.org. Instead you only add it to your own web site’s events page, properly tagged as an event using microformats. There it will be found by upcoming.org, Google, and many other sites, and listed together with tens of thousands of other tagged events that were published anywhere on the web.

What applications use data tagged with microformats?

Search engines are already taking advantage of microformats. An experimental Yahoo site called Microsearch nicely shows what can be done. The engine extracts microformatted information from pages and gives it a prominent place in the search results. Properly tagged location data shown on a map, providing a user experience similar to mashups.

Another really nice example of how software can take advantage of microformats is Operator, a Firefox 2 add-on to detect microformats in web pages. Now that I have used Operator, I really look forward to Firefox 3’s native microformat support. Operator lets you view and export the tagged data. It nicely integrates with applications like Google maps, Yahoo maps, Delicious, Youtube, and Upcoming. Operator can even search Flickr photos that match the tags used in your blog post.

Operator add-on at work

Are microformats an eMarketing tool?

Very much so. Internet users consume information with the tools that they find most convenient. Microformats make information accessible to such tools.

If you don’t use email newsletters and RSS feeds to promote your brand, a significant part of your target audience will not receive your message. Microformats add more choice to how users receive information, just like email and RSS.

Suppose that we want to send a travel newsletter and we need to target our email subscribers based on their favorite travel destinations. The newsletter signup form would offer a selection of destinations of which our subscribers choose one or more. How can we then make sure that subscribers only receive content that matches their destinations of interest?

1. Sending different newsletters to separate subscriber lists

This segmentation option is the most basic one. We could even implement it with an email program like Outlook. Each travel destination has its own newsletter and is sent to a specific list of email addresses. We’d have a France newsletter and send it to the France list, and the Italy newsletter goes to the Italy list.

Main advantage:

Each newsletter can have its custom name and design.

Main disadvantages:

A subscriber intersted in multiple destinations would have to receive multiple different newsletters. Furthermore, if one newsletter issue contains content that is of interest to multiple subscriber lists (e.g. Southern Europe), it needs to be separately sent to all the lists concerned.

Implementing separate email templates and lists usually is the preferred option if either the newsletters or the target audiences are very different. Using this model is therefore probably not the right way to target our travel newsletter.

2. Sending different newsletters to one segmented subscriber list

With this option the France and Italy newsletters are sent to one big traveller list. To implement the segmentation, areas of interest are stored in the recipient’s record as a profile field:

Traveller List

Record   Email                 First   Last   Interest           
1        tom.foo@abc.com       Tom     Foo    France,Italy,Greece
2        cliff.bell@hdef.com   Cliff   Bell   Spain,France,Germany
... etc ...

Main advantage:

If the email marketing program supports the selection of dynamic sub-lists that are based on the main list’s profile data, the same newsletter can be sent to multiple interest groups with just one mailing in a highly targeted fashion. Example: “Send Southern Europe News to the Travellers list if the recipient’s Interest field contains Spain or France or Greece”

Disadvantage:

For narrow targeting (Spain only, Italy only) separate emails need to be created, sent, and tracked.

3. Sending one segmented newsletter to one segmented subscriber list

This is the most advanced of option. It can only be implemented if the email marketing software/service supports inline segmentation. With inline segmentation, content blocks are defined as belonging to certain subscriber profiles. This is usually done by placing IF…ENDIF statements around content blocks:

This is the most flexible of all targeting solutions, but it requires careful testing of all targeting criteria against all dynamic content.

At a conference last week I met Simon Revell, a Pfizer UK employee who had successfully helped implementing enterprise 2.0 in his organization, mainly blogs and wikis. Very interesting presentation.

First of all I was surprised that he is an IT manager, so the IT department brought this project forward, not Marketing, not Communications. I guess it would not have been possible otherwise because there was no social software available in the organisation, Simon’s team simply set up a LAMP open source environment using Drupal and hooked it to the network. Nobody but IT can do this in a company where everything is running on Microsoft technology.

Simon had an interesting story to tell about the difficulties of kicking off enterprise 2 (”Who gave you permission to do this?”), marketing the new “corporatepunks” culture, and getting their blog known to UK and international colleagues to make it a lively communication medium.

You can find out more about Simon.

Definitely check out the “Meet Charlie” presentation created by an IT colleague of his.

Now that Sharepoint 2007 is running more and more corporate intranets, enterprise 2.0 is easier to implement, at least technology-wise. Sharepoint supports blogs and wikis, not feature-rich but good enough to get started. What are we waiting for?

Daniel,

We purchased domains with and without Sedo’s help.

Are you dealing with a professional domain seller? If this is the case, entering the name in a browser will usually redirect you to a page where it is offered for sale. Be prepared to pay a lot more than just the regular registration fee and parking costs. Don’t get shocked by a ridiculously high first offer though. The real price is often much lower and will mainly depend on three factors:

1. Size of your organization

Bigger organizations are assumed to have bigger budgets for domain purchases. The seller will therefore always try to find out who you are.

2. The domain name itself

Domains containing brand names are sold at higher prices, and .com versions in particular.

3. Number of potential buyers

If there is a certain amount of competition for a name, the seller will only sell it at a low price after several high price attempts were unsuccessful. So if your budget was too low, you could try again after some time.

The price of a domain is no more and no less than what a buyer is willing to pay for it. I know, having to pay a premium for “just a name” is painful, but bear in mind that selling domain names is a legitimate business, intellectual property right issues aside.

If the seller is not a professional “domain grabber”, the domain usually either goes nowhere or loads some sort of “coming soon” page. You may be able to bag yourself a bargain, but you may also not get the domain name at all if the current owner really wants to use it themselves. We once tried to buy a domain name for one of our products which happened to have the same name as a small German IT company. They kindly told us that the name is not for sale unless we were “prepared to pay a ridiculously high amount”, which we weren’t.

Contacting the seller anonymously can help if you represent a larger organization or if you want to buy a domain for your brand. However, at some point during the buying process you need to reveal your identity and there is no guarantee that the price will then not increase again. If you want to remain anonymous during the entire process you need to ask a third party to do the purchase for you. Often this is a company employee who buys the domain name in private and then transfers it to their employer. Be sure that this person cannot be associated with your company by simply searching Google. And as with all anonymous buying, also observe your company’s policy for conducting business.

If you want to buy SomeBrand.com and your company is the most logical single interested buyer for that name, the seller will suspect that you represent it. Using a third party just to remain anonymous may therefore not work. In a situation like this we nevertheless used Sedo.com to successfully buy a domain name. Sedo did a great job to independently evaluate the domain name and negotiate with the seller on our behalf. The initial price was 20,000 USD. Sedo was able to convince the buyer that our maximum budget was MUCH lower and that it would be difficult for him to sell the name to somebody else. They finally managed to buy this name for us within the small budget we had.

Our organization owns hundreds of domain names. Buying them is a rare exception though. I often recommend to use a different top level domain (.net, .info) or a variation of the name like …..-online.com or …..-info.com.

Hope this helps,

–Walter Soldierer

User protest

It was only a matter of time when online Scrabble fans would start fighting back. Even though I do not believe that public pressure will impress Hasbro, there is already a new Facebook group called “Save Scrabulous! Give us Scrabulous or Give us Death!” with more than 2600 members on January 16. A Scrabulous petition can be signed on ipetitions.com, with about 300 signatures as of today. Customer service links on Hasbro and Mattel web sites were posted on the web so that Scrabulous supporters can utter their protest directly.

Copyrights, copyrights…

Hasbro (US, Canada) and Mattel (other countries) own the intellectual property rights to Scrabble worldwide. Since I joined Facebook, I have become a big Scrabulous fan and I had already asked myself whether this application was licensed from Hasbro, a company that took action against competitors in breach of copyright before. I am convinced that Hasbro’s legal claims are 100% vaild, and a disclaimer notice like this one will not let the makers of Scrabulous get away with breaching copyrights.

Talk, don’t sue!

Are take-down-notices and legal actions the right things to do in this situation? No information is available on the Hasbro and Scrabulous web sites, so we do not know whether the companies had entered into discussions about working toghether before. I would hope that Hasbro isn’t one of those non-flexible monopolic giants who prefer suing over negotiating.

Hasbro is missing out their biggest Scrabble promotion opportunity ever. Since I play Scrabulous online, we are again playing the Hasbro original game in our family. Fortunately I found the dusty Scrabble box in the attic, but if I hadn’t, I would have surely purchased one on Amazon.com.

This case shows again that many companies still don’t seem to get it. Either they adapt to the new princples of marketing, involving online marketing and – in this case – social networks, or they will soon be in serious trouble. It was estimated that the two Indian brothers cash in about 25.000 dollars each month in Scrabulous advertising revenue alone. Of course Hasbro as the rights owner to Scrabble needs to get their share of the pie, and the Agarwalla brothers would be stupid not to cooperate. Another big opportunity for Hasbro lies in promoting their hardcopy version of the game on the Scrabulous web site and Facebook application. With millions of users playing the game, Scrabulous built an amazing marketing machine for Hasbro. But rather than seeing their revenue opportunities, they are trying to kill it.

Helloooo! Hasbro! There is money to be made! Scrabulous is promoting your product! Just talk to them about working together. And let me continue with my current game please, I’m about to win…