Do you really have to segment that campaign? The limits of targeting.

When theorizing about marketing campaigns there is often a tendency to want to create very detailed segments and perfect messaging that speaks directly to the hopes and desires of various personality types and demographics.  It's great in theory. The only problem is that hyper-targeting is very expensive to create and execute with any type of quality and creativity. Say you have 10 targets. This means that you have to create 10 separate creatives and manage all these additional creatives within your campaign. Then, if you are doing real marketing, you have to analyze the performance of the various creatives and campaigns designed for your targets and determine what, if anything, is actually working. I've found that it's smart to think carefully about whether targeting will have a real impact before pulling the trigger on a highly segementated campaign. Sometimes its best to simply do one thing really well.

Why They Called It the Manhattan Project

I stumbled on this New York Times article that describes just how Manhattan centric the origins of the Manhattan Project actually were.

Manhattan was central... because it had everything: lots of military units, piers for the import of precious ores, top physicists who had fled Europe and ranks of workers eager to aid the war effort. It even had spies who managed to steal some of the project’s top secrets... The borough had at least 10 sites, all but one still standing. They include warehouses that held uranium, laboratories that split the atom, and the project’s first headquarters — a skyscraper hidden in plain sight right across from City Hall.

Imagine warehouses full of uranium in Manhattan today! It shows you just how different 1940's Manhattan was. Did you know that J. Robert Oppenheimer grew up on Riverside Drive and went to the Ethical Culture School? The Times also includes an interactive map showing some of the major sites of the Manhattan project.


Don Draper at the Lenox Lounge

"When we meet Don Draper, he is sitting alone in an upholstered booth at a smoke-filled, Art Deco-appointed nightclub, sipping an old-fashioned from a rocks glass and scribbling ideas for the Lucky Strike cigarettes account on a cocktail napkin." -Mad Men's Manhattan: The Insider's Guide

The actual location is the Lenox Lounge, just up the street, which always seems to stay the same despite the fact that everything around it is changing at a furious pace -- just like most of Harlem these days.

Thanks for the intel Josh. I enjoyed our drink there last week.

UPDATE: The Lenox Lounge recently closed, but is moving up to 127th Street. A new Jazz club called the Notar Jazz Club is scheduled to open in the old location. Read the interesting back story.

App Spring Cleaning

Do you have screen upon screen of iPhone/Android apps? Maybe you've got so many apps that you've started using the questionable categorization feature on your phone to bucket large groups of apps together. If so, you probably are long overdue for an app Spring cleaning. I just did clean sweep of my phone and I feel 100% better about my relationship to technology.

Below are some of the apps that I kept:

Work: OmniFocusEvernoteSalesforceStarbucksPayPal

Reading: ReederKindleiBooks

Learning: iTunes U

Running: Nike App

Music: iTunesPandoraSonos

Of course I kept the standard apps, but that's pretty much it. Keep your tech simple, so you can live life.

Choosing a Manageable CRM and Email Marketing System (Salesforce + ExactTarget + Demand Tools)

About two months ago I launched a customer relationship management (CRM) and email marketing system for an organization. The implementation utilizes Salesforce CRM,ExactTarget’s email marketing platform, and Demand Tools for data cleansing and maintenance. For the most part, I’m very happy with the way everything turned out, but this type of project is not for the faint of heart.

Implementing and maintaining a fully functional CRM is a big data management challenge, and it is a human challenge as well. A new CRM generally touches many departments and requires changes to business processes. Even in a small organization this is type of change is bound to encounter serious resistance. Luckily, in this case, we were only focused on improving our customer support and marketing programs, so the involvement of departments outside of marketing was generally limited.

I am sharing my experience now, because I noticed a lack of objective information, while researching the various technologies available. There is a great deal of money in the business of managing customer data these days, and all the marketing campaigns for this solution or that one create so much unhelpful noise. My aim here is to clear up the confusion, and begin a conversation about a topic that I find fascinating – CRM and data driven businesses really are cool! So, let’s begin.

The Goal
The goal of this project was to take multiple flat files from email newsletter subscriptions, house mailing lists, event registrations, and other customer lists, and combine these into a single relational database. By bringing together many widely dispersed data sources the new CRM would now be able to track some of the following key vital information:

  1. Recency, frequency, and monetary value (RFM) data from past purchases
  2. Customer lifetime value
  3. Engagement with communications e.g. (recency and frequency of email opens and clicks).
  4. Interactions with customer service.

We planned to and we are now using this information to greatly enhance our interactions with customers.
 

The Choices
There are an innumerable number of choices for CRM solutions. Some of the major players are SAP, Oracle, Salesforce, and Microsoft. I also considered the small business focused Highrise and the “open source” SugarCRM.

The Decision
In making my decision about the appropriate CRM for a small to mid-size business, I eliminated SAP and Oracle as these vendors are mainly focused on enterprise level multi-million dollar implementations, which did not fit with my lean budget requirements and would not have been cost effective. I also eliminated SugarCRM, because the time and labor cost of implementing and maintaining their free ”open-source” solution would be prohibitive and the cost of their professional services and hosted solution was inline with Salesforce and Microsoft.
I eliminated Highrise, because their pricing plans stopped at 30,000 contacts and there was no obvious number to call to explain that I had more contacts. This indicated to me that the company is focused on very small businesses and would generally have a hands-off approach to customer service. I was left with Salesforce and Microsoft Dynamics CRM as the real contenders for my needs, so I installed trials of both applications and I did some fairly extensive testing of both systems to understand the pluses and minuses of each.
The conclusion that I came away with from my Microsoft testing – about a year ago - is that the product still has a way to go. The interface was rather confusing and importing contacts through the web interface was not as easy as I expected. It seemed that the system was designed so that data manipulation was supposed to happen within a Microsoft Office product like Outlook or Access. This left me to wonder how effective the product would be for Mac users in the organization. Finally, when I called to ask about pricing and my problems with data importing and management, the sales and customer support representatives that I talked to were not very helpful and seemingly rather disorganized. This did not speak well for a company that makes a customer relationship management system.

In the end I decided to go with Salesforce for a number of reasons. I was able to import and start working with my contacts quickly. The interface is very intuitive, so there is a relatively shorter learning curve for new users. Salesforce customer support is very good in contrast to my initial experience with Microsoft. All common end user tasks can be easily accomplished through the web interface, so Mac user will likely have a good experience now and in the future. Further, Salesforce now had some great iPhone apps for both the general CRM and their new Chatter service (think private Twitter for business). Finally, there are lots of third-party apps – available on the AppExchange - that round out the system and help it do exactly what you want. The two key pieces of third-party software for my purposes were Demand Tools and the ExactTarget app.

On the negative side Salesforce, like other cloud based CRMs, charges on a per user basis and you will likely need at least 10 users to begin even a limited deployment of Salesforce within your organization. Each third party application that interfaces with Salesforce will generally need a user account as well as anyone working on the system. The enterprise system, which I recommend, is $125/user/month, so you should expect to spend at least $15,000 dollars per year on your Salesforce CRM and costs will escalate quickly as you deploy it to more employees. There are also data storage and API limits and additional data storage can be quite expensive. Still, the Salesforce cost structure is relatively inexpensive in comparison to other CRM solutions.

Third-Party Software for Salesforce
The Demand Tools suite of applications allows you to easily perform very complex data manipulation and cleansing routines via a user interface rather than SQL coding. The applications are only available for Windows, which is rather annoying, but Demand Tools is so useful for Salesforce administrators that it certainly justifies installing Windows on your Mac, which I did.

The following are a few examples of what you can do with Demand Tools: merge all contact records with the same email address; assign country codes based on the ending string of the contact’s email address, and perform very complex filtering to clean and merge postal mailing addresses.

The ExactTarget app allows you to send mass emails to newsletter lists and any other custom segments that you define directly from Salesforce. This app works well for sending emails from Salesforce through ExactTarget’s email engagement platform. It is also relatively easy to do basic tasks like personalizing emails.

However, the ExactTarget integration is not great at sending information back to Salesforce. ExactTarget’s subscription management tool for Salesforce leaves a lot to be desired, and we had to build our own subscription management form because of this. Also, the system sends every piece of email engagement data (opens, clicks, etc.) back to Salesforce for every subscriber, so you will quickly run out of space in your Salesforce database with the out of the box solution.

Although it will require a custom solution, my recommendation is to manage email engagement data in ExactTarget and pull in the needed contact data from Salesforce to do reporting on email performance by demographic. You will also find it useful to summarize and push back engagement data for individual subscribers to Salesforce, so you can understand, who your engaged/non-engaged subscribers are. Unfortunately, this will also require custom coding, which means additional time and expense. 

The bottom line is that the ExactTarget integration with Salesforce leaves a lot to be desired, but I have not found a better solution for handling email marketing at the moment. I am working with the ExactTarget support team on the various issues and custom coding, and I will continue to experiment with other email vendors’ integrations going forward.

Summary
In sum, the Salesforce, ExactTarget, Demand tools solution for CRM management is quite good and I would certainly recommended it with the few warnings outlined above. I hope this note helped you understand some of the dynamics involved in choosing a CRM solution as well as some of my reasons for going with the particular solution that I chose. Look out for my next post, which will dive into how to extract and take action based on the recency, frequency and monetary value data in your CRM.

Are the Ad Networks Missing the Boat on Privacy? Thoughts on Promoting Universal Ad Preference Management in Addition to Universal Opt-out

Long before the privacy debate erupted in its current form, Google has had a fairly simple straight forward tool called Ad Preferences Manager, which allows users to adjust the types of ads that Google serves them by changing the preferences stored in Google's tracking cookie. Unfortunately, most people have no clue what the Ad Preferences Manager is or what it means to them, and Google has done little to change this fact until recently. The fact that Google has not done more to gradually educate its millions of users on the positive aspects of controlling the ads that they see online represents a serious missed opportunity to shift the privacy debate away from "tracking" users online and towards empowering web users to take control of the ads that they see.

Despite the lack of marketing and consumer education around the Ad Preferences Manager, I actually think that it represents the best model for the kind of tool that the advertising industry should be providing and promoting to web users. It is simple and easy to use, and it promotes the positive side of controlling the ads that one sees as opposed to just promoting tracking opt-outs, which would kill the online ad industry, if adopted en masse. I find it strange and troubling that the industry has rallied around promoting opt-outs without simultaneously promoting the advantages of setting one's preferences and receiving more targeted advertising. I completely respect every web user's right to opt-out of being cookied and anonymously tracked, however, if the industry promotes only the benefits of opt-out without simultaneously providing the tools for users to opt-in to more personalized and meaningful advertising, then many users will simply opt-out considering the benefits of a more personalized relationship to advertising. Further, the ad networks are missing a golden opportunity to ask individuals directly what types of ads they want to see and deliver on these preferences.

There are a number of issues related to providing a universal ad preferences manager, but it seems that the benefit far out way the difficulties. The first problem is that Google, Yahoo, Microsoft, AOL, and other ad networks have not yet decided on the standard classification system that could be used to define the ad preferences that a user could choose from. Google, which recently bested Yahoo, as the number one provider of display advertising could push the industry to adopt something like its classification system, but the reality is that in display advertising, Google commands nothing like its 65% of the search market, so it has less influence and less ability to get other parties to follow its lead. The solution here is that the various large ad networks need to come to an agreement on unifying the categories on which online advertising is targeted and agree to abide by these user preferences when serving ads.

The second problem is that cookies only persist in web browsers for a limited amount of time; either until the you clear your cookies, upgrade, or changes the browser you are using. The ability to keep your ad preferences stored across all browsers and devices that you use would require logging into a service that keeps track of your preferences. Google's Advertising Cookie Opt-out Plugin does exactly this by permanently saving an opt-out cookie in your browser that persists even when you clear all other cookies. However, Google offers no plug-in that allows you to permanently store your ad preferences despite the fact that this would be a relatively straightforward modification of the current opt-out plugin. In my view, this is a missed opportunity for Google to both offer web users more personalized advertising and an alternative to tracking opt-outs, while at the same time leading the industry to a more sustainable model that gives web users direct control of the ads they see.

Another issue is that the whole privacy debate has moved from the space of technological innovation to one of legal wrangling where very powerful interests are out to gain competitive advantage through the legal outcomes in Washington D.C.. This does not create a particularly great climate for innovation. In fact, some of the fastest growing ad networks are the social networks like Facebook and LinkedIn, which have quite a lot to gain from more stringent tracking and opt-out legislation, as they operate walled gardens of the web with vast troves of personal data against which they serve ads. They are relatively immune to the effects of privacy legislation, as users willingly provide their personal information and agree to the privacy policies of their social networks of choice. Therefore, I doubt that we will see innovation on privacy or any kind of universal ad preference manager coming from the social networks. The social networks are undoubtedly very comfortable with were they stand vis-à-vis the privacy debate and will likely provide a challenge to any standardization that is not based and dependent on their standards.

In sum, despite the various challenges, Google has a simple and easy to use platform for managing individual relationships with advertising and a slightly dominant market position that it could use to shift the debate on privacy from simply promoting and enforcing opt-outs to enabling more personal management of advertising preferences where one option is universal opt-out. Up to this point the ad networks' singular focus on opt-outs has not served the industry well and the technology for enforcing it has been clunky and ineffective. It is time for a company or group of companies, with the required market share and technical acumen, to build a universal ad preference manager that can be easily installed in a web browser and used to either control the ads served or opt-out of tracking completely. A universal ad preferences manager would be a boon to the industry and is fundamental to its future health and success in the face of privacy concerns and more stringent legislation.

Travels and Redemption of a Stolen iPhone

My wife lost her iPhone last Saturday night. When we realized this Sunday morning, I first tried using Apple’s MobileMe service, but the locator service was disabled. I then signed up for AT&T’s FamilyMap service, which is designed to help track ayour kids, but is also a last best hope should you loose your phone. FamilyMap pinpointed the location of the phone to a restaurant in Harlem where we ate dinner the night before. I called the restaurant and was told that the phone was nowhere to be found. I would have to wait for the manager to find out if it was put away the night before. When I tried to locate the phone again, FamilyMap indicated that the phone was not available.

The connection between my phone call and the phone dropping off the radar made me very suspicious, so I walked over to the restaurant, which is only a few blocks away, to see what was up. I found two people cleaning the restaurant and their supervisor, who I had talked to on the phone. He and I looked everywhere and then I just decide to give up and wait for the manager. The most likely scenario seemed to be that someone found the phone while cleaning up and then turned it off when I called the restaurant. When the manager arrived she assured me that nothing was found and they would continue to question the staff and look throughout the day. There was really nothing I could do so I left and continued to try and track the phone throughout the day.

Later in the day I called and reached the restaurant owner, who said they had been looking throughout the day but could not find anything. They had gone over the whole restaurant and questioned all the employees. They even went through the linens from the night before to see if it got caught in those. I thought this was really going above and beyond, but as it turned out this was just the start. Soon after hanging up with the owner I got another signal from an address in Queens, so my wife called back and asked if anyone on the staff lived at the address. It turned out that one of the employees who cleaned up that morning lived at the exact address. Busted!

The owner called the employee and told him to come back with the phone or she would call the police. He showed up at the restaurant an hour later, but he completely denied taking the phone. The restaurant owner asked us to come over to show him the evidence we had, so we bundled our young daughter up and headed over to the restaurant hoping to quickly resolve the issue. No such luck. I showed everyone including the thief the detailed maps with the location of the phone at 10:30AM and the location of the phone at 6:30PM. Still the thief denied taking the phone or any knowledge of how the phone managed to get from the restaurant to his apartment. We told him we would call the police if he didn’t give us the phone but he persisted in denying it.

Now this was just getting stupid, but also interesting. We caught him red handed, he might be arrested if he does not give us the phone, but he continues to deny it. The logic perplexed me, however, a smart thief would not have turned the phone back on after getting home from work, so I was thankful for his stupidity. At this point we had no choice. We called the police and I went home to put our daughter to bed while my wife waited for the police with the owner and restaurant managers. It seemed that everyone had become fascinated with the turn of events.

While waiting for the police, my wife tried to locate the phone and found that it was now a few blocks away from the restaurant and it kept moving up and down the block. After putting my daughter to bed, I logged in to see this interesting activity. We guessed that the thief had left the phone with a friend, who was getting impatient and pacing the block.

At this point the police arrived and he continued to deny any knowledge about what was going on despite the evidence and the indication that the phone was just around the corner. However, after more questioning and the arrival of additional officers he finally said that he might have "dropped" the phone we were seeing it on the map. The police, my wife, the restaurant owner, and the thief all piled into squad cars to go find the phone. But, the adventure was not over yet!

The thief still, “did not know where it fell,” so the police began combing the entire area. They looked through everything on the street corner. It was really impressive how helpful they were! While everyone else was looking for the phone, the thief just stood there smirking. It became clear that he knew exactly where the phone was. At this point the owner of the restaurant told thief that she was giving his remaining wages to my wife unless he told us where the phone was. All of a sudden he seemed to remember that the phone was tucked under a parking lot fence in a CVS bag. Mystery solved!

We are sending flowers to the restaurant today. We are not sure how to reach the police officers, but they truly showed themselves to be New York’s finest. We will be sending them something soon. We are so happy to have all our picutures, videos, and memories back. Thank you all!

Search Marketing Terms You Need to Know

The first thing that clients often ask when we start setting up a search marketing campaign is, "What do all these terms and acronyms mean?". Here is an overview of the basic search marketing terms you need to know when running campaigns with Google AdWords, Yahoo Search Marketing (YSM), Microsoft AdCenter, and other search marketing networks.

Ad position: The the location of an ad on the search results page. Position 1 is the top of the first page. On Google, if your ad is in position 1-8, it will almost certainly appear on the first page of search results, though positions 1-11 may appear on the first page of search results depending on the configuration of advertisements.

Bid price: The maximum amount of money an advertiser is willing to pay for a click from a given keyword. Advertisers generally pay less than the maximum bid price set in the Google, Yahoo, or Microsoft ad management systems.

Call to action: Direction within an ad or a web page for the reader to take an action.

Conversion: A desirable action taken by a web site visitor. Conversions can include joining a mailing list, buying a product, calling a phone number, visiting a web page, or downloading a file.

CPC (Cost per click): The amount an advertiser is charged for a single click. Different keywords cost different amounts, depending on competition.

CTR ( Click-through rate): The number of clicks an ad receives divided by the number of impressions. The higher the CTR, the more effective ad management systems (AdWords, YSM, AdCenter) generally consider the ad. A high click-through rate doesn't necessarily mean your ad is effectively meeting your conversion goals, but it certainly means that Google, Yahoo, or Microsoft are making more money from your ad. For this reason, ad management systems generally rank ads with a high CTR above other ads.

Impression: The display of an ad on a web page.

Landing page: The first web page shown after an ad is clicked. The page is constructed to appeal to the same desire as the ad.

PPC (pay per clck): The advertising model that charges advertisers only when their specific ads are clicked.

Split test: Test that divides online traffic randomly between two or more creative approaches and measures which one generates more conversions.

Traffic: The number of visitors to your website.

Visitor value: How much money, on average, a single visitor to your web site is worth. Determining the visitor value will help you set your bid price, so that you can be sure that you are not losing money on your PPC campaign.

Launched Tadias.com

I recently launched a news style website for Tadias Magazine. The site is my first stab at deconstructing the WordPress content management system and the idea that personal publishing should always adhere to the rather boring blog format.

At root WordPress is a fully functional open source CMS that is quite easy to play with and modify. Many media companies use their own customized versions of WordPress to power large complex sites, so it shouldn't be that difficult to give the average user this power in the form of feature rich easily customizable templates.

One step removed from reality, Second Life or This Life?

I started playing Second Life (SL) this weekend. For the uninitiated, Second Life is a virtual world where users can do pretty much anything you do in the normal world. Users (residents) can buy property, build houses, start businesses, buy services, host events, chat with others, modify their appearance, etc.. Generally the laws of physics and societal norms that govern SL mirror the real world, however there are some important exceptions. Residents can fly and teleport, and I don't believe residents ever get hungry, thirsty, tired, injured, or dead.

One of the interesting aspects of SL, especially for the non-gamer, is that residents can buy Second Life currency, Linden dollars (L$), with real world currency on the LindeX currency exchange. Linden dollars can then be used to purchase goods and services in the SL world. Even more interesting is the fact that people actually spend real currency to purchase Linden Dollars. The LindeX data shows that the current exchange rate hovers around L$270 / US$1.00.

As in the real world it is difficult to make the cold hard cash (Liden Dollars) in SL. It seems that the most lucrative profession is as a Second Life marketer in the real world. Residents get L$2000 for every member referral who becomes a premium subscriber. Designing items for the virtual world can also be quite lucrative, and the world's oldest profession seems to flourish as well. However, there is also a bit of a welfare system in SL and users receive a certain number of Linden Dollars per week just for logging in.

I'm still waiting for my check, but my stipend is supposed to be L$400 per week. This isn't to shabby considering that an entire virtual apartment building can be purchased for L$6000. Also, I'd actually make US $1.50 per week off the Linden dole, if I sold my Linden Dollars for U.S. dollars at the current exchange rate. With the injection of real cash, this virtual world becomes a lot less virtual, but until the exchange rate gets a little better, it's still a game to me.

 

Regardless of the financials and all the marketing hype, the looming question still remains, 'why would anyone want to spend significant amounts of their real and very short life in a virtual world?' Judging by the list of the most popular places in second life the draw is mainly sex followed by gambling. The Reuters, CNET, Wired spaces were totally dead when I stopped by.

I don't think I'm going to be spending much more of this life in Second Life. I judge all internet phenomena based on whether they make my life better, and I just don't see SL doing this. SL interests me mainly as a tool for facilitating discrete tasks like building a park, modeling human behavior, or holding a meeting (though video conferencing seems more practical). The media companies who are tripping over each other to setup shop in SL are over excited about the marketing potential of the platform, but who can blame them. Their job is to live in the moment and virtual worlds make for great headlines.

Eric Rice has a good post on all the real difficulties that SL is now facing including a customer rebellion over customer service and increased prices, as well as emerging competition in the metaverse marketplace. I'm not sure what companies likeCrayon think they are going to do in Second Life that they can't do in the real world. No doubt, they got lots of links from their recent dual launch in both the real world and in SL. However, I think that after a couple months hanging out in SL, Crayon will soon realize that they could be doing more productive things in the real world and on the real internet.

Okay, it's back to reality for me folks.

The Microsoft AdCenter Experience: Day 1

I just had enough time to sign-up for AdCenter today and upload a few keywords. The interface for adding keywords is a little clunky. The dropdown menu on the keyword page upload page showed me gibberish instead of options. Once the keywords were added I had to click on 50 individual check boxes to chose phrase match, and I then I had to click on 50 individual checkboxes to deselect broad match. Yuck!

It seems that AdCenter also relies on a Yahoo like approval process rather than letting you edit ads on the fly. I'm still waiting for my ads to be approved and go active, so I'm not sure if AdCenter does Google like A-B testing to automatically figure out which ad is best and serve it more. I love that feature in AdWords!

The day parting and demographic targeting look great, but it is going to take a few days to see if these features work as advertised. Everything is on hold for now, because as of 9pm EST I got an error message that I could not log into AdCenter due to an "unspecified data error." At 10pm EST I got a message that AdCenter had "no record of my user name." I double checked and it was definitely my user name. Hmm... not very promising, but I'll try again tomorrow.