Archive: September 2009

JCPenney recently began testing POS mobile coupon redemption in the Houston metro area, claiming to be one of the first U.S. retailers to instate such a program. Accessible to consumers through the mobile coupon fulfillment service Cellfire, these coupons can be scanned directly from the phone screen at the register.

In an article published on mobilemarketer.com, Dave Owens, the development director for emerging digital media at JCPenney, talks about the company’s mobile initiatives. From the source:

“We recognize that this is where the customer is ultimately going, and we want to make sure we’re there and ready when the adoption curve ramps up. Mobile is a convenient way for consumers to shop JCPenney and redeem coupons and we want to support that need. The biggest thing is immediacy, because it puts the consumer in control. Its a competitive advantage to have a coupon right on her phone, which provides true immediacy and a discount offer she didn’t have to plan for.”

Comparative shopping. Coupon hunting. Customer service and product reviews. Location-based offers. As the real-time realities of a mobilized world settle in, this sort of instant gratification from retailers will become the norm. Read the full article here.

Online shoppers have taken a cue from  Count Laszlo Almasy from the movie The English Patient. They promise they’ll be back. And the outcome for shoppers seems to be more successful than it was for poor Almasy. From an article in yesterday’s Marketing Charts email:

According to a study by McAffee, nearly two-thirds (65%) of online shoppers who “abandon” their shopping carts actually come back after a day or more and full convert on a purchase.

So many of the shoppers who “abandon” are really just delaying. Which means your promotional strategy might require another look. For example, you might not remarket to “abandoned” prospects with discounts if really all they’ve done is delay their (full-price) purchase. Testing this premise is probably worth the time and effort.

As far as why prospects delay purchase, the study posits a few reasons: basically they all add up to prospects needing more time to consider their purchase (price, reliability, need) before committing to buy.

When taking that into account, your remarketing test might include a couple of cells that look like this:

1. Do nothing

2. Remarket with value/reliability messaging

3. Remarket with discount

Oh, and it is worth mentioning that the study also showed that the McAfee “trustmark” at checkout converted more buyers. This is not surprising, we find that lowering the barriers to purchase - one of them being security of transaction - is a “must do”.

There’s a new-ish technology called Cufón that is a serious competitor to sIFR. Before we explain what’s so great about Cufón, let’s talk about how cool sIFR is and what both technologies do.

All over the web you can see examples of sites that use non-web-safe fonts and with good reason: web-safe fonts are hideous. To get around the fact that users don’t have a fancy font installed, most sites rely on images for display text. The most common examples look like this:

<h2><img src=”/images/test/h2-how-to-apply-for-a-grant.gif” alt=”How to Apply for a Grant” /></h2>

Problems with Using Images for Text

1. Each image must be downloaded from the server. This makes pages load slower and increases the load on the server.
2. Each image must be individually produced with a tool like Photoshop. This means it is time consuming to implement.
3. Each image must be re-done if the text needs to change. It’s impossible to manage images easily using a CMS.
4. SEO is problematic. Even with an alt tag (which is often omitted), it’s still not as good as real text for SEO.

Traditionally, SolutionSet has used sIFR for websites that need dynamic text in a non-websafe-font. The big up-side of using sIFR instead of images is that it dramatically reduces the production time and even the page weight (the sIFR files are only downloaded once and then cached on the browser). Creating thousands of images and forcing user’s to download those images is a bitter pill to swallow just for some fancy text, but people do this all the time (don’t forget the SEO pitfalls of using graphics for text). There’s a technology very similar to sIFR, called FLIR, which uses PHP to generate the images on the fly which removes the issue of production time but still requires users to download a image for each header and also requires that PHP be available (this isn’t a scalable solution, it puts the load on both the browser and the server).

How does sIFR Work?
sIFR uses a combination of JavaScript and Flash to render text. There are only three basic components: a Flash file that you prepare that contains your font, the sIFR JavaScript file, and some declarations in JavaScript that describe which text to replace. Once the page loads, sIFR searches the page for — let’s pretend we want to replace all of our <H2> tags with our fancy font — all <H2> tags, measures their height and width, positions the real text off-screen and places flash file in its place. There’s already some good documentation on how to use sIFR on their site. Search engines see the plain text that exists before it is replaced with sIFR and normal visitors get to see fancy text. The best part of sIFR is the potential file-size savings over using graphic images and the increased server performance. All of the work is done by the browser when the page is loaded.

Original <H2>:
<h2>How to Apply for a Grant</h2>

After sIFR runs:
<h2 class=”franklin sIFR-replaced” style=””>
<object id=”sIFR_replacement_0″ class=”sIFR-flash” height=”31″ width=”335″ type=”application/x-shockwave-flash” name=”sIFR_replacement_0″ data=”/js/franklinGM.swf” style=””></object>
<span id=”sIFR_replacement_0_alternate” class=”sIFR-alternate”>How to Apply for a Grant</span>
</h2>

It’s important to remember that even though Firebug shows the HTML after sIFR has altered it, what is actually sent to the browser (the source HTML) is the plain text shown in the original. You can see this in action on the homepage of one of our clients: http://www.stuartfoundation.org/

What’s wrong with sIFR?
The biggest problem with sIFR is how difficult it is to set up. Between Flash, CSS, and JavaScript, there are a lot of moving parts. It also requires that you have Adobe Flash installed in order to implement it (you need Flash to create the embedded font file).

The second issue is its overall quirkiness. Common problems with sIFR sites are the sIFR blink and the sIFR jump. You’ll see the sIFR blink if you load a page with sIFR and hit refresh. Either the text will show using a web-safe font momentarily or the text won’t show at all until the sIFR text suddenly appears. The sIFR jump is when the replaced elements change dimension slightly after sIFR loads. Often it looks like the page suddenly gets a few pixels shorter. You can see both of these effects on most website that implement sIFR (hit refresh a bunch on the sIFR homepage: http://novemberborn.net/sifr3). Flash is a plug-in and renders in its own virtual machine outside the browser. This means that there are hard limitations as to how quickly it can execute.

Cufón is way better!
Cufón operates very similar to sIFR in principle but instead of using Flash to render fancy fonts, it uses the <canvas> tag in modern browsers and VML in Internet Explorer. Canvas and VML share a lot of common functionality and the basic idea is that they allow JavaScript to render vector graphics natively in the browser without the need for an external plug-in like Flash (there’s some really cool stuff the <canvas> tag can do). The key concept of Cufón is a recognition that fonts are essentially vector files and all browsers have a built-in ability to render vectors. Relying on built-in browser technology allows Cufón to operate extremely quickly and completely eliminates the sIFR blink and the sIFR jump.

Configuring Cufón is also a dream compared with configuring sIFR because Cufón relies on CSS for all of its styling cues. You can tell Cufón to replace all <H2> tags and even though, hypothetically, some are blue and 14px and some are green and 22px (depending on their context) they all get replace correctly with a single declaration of Cufon.replace(’h2′). sIFR doesn’t make things so easy.

Cufón also avoids some the line measuring issues that sIFR suffers from by splitting each individual word into its own canvas tag which allows the text to flow naturally. It also works well within links, even natively supporting the :hover psuedo-class. There is a great how-to on the Cufón GitHub.

Original <H2>:
<h2>Lasting Impressions</h2>

After Cufón runs:
<h2>
<span style=”width: 182px; height: 38px;” class=”cufon cufon-canvas”><canvas style=”width: 202px; height: 38px; top: 0px; left: 1px;” height=”38″ width=”202″></canvas><span class=”cufon-alt”>Lasting </span></span>
<span style=”width: 270px; height: 38px;” class=”cufon cufon-canvas”><canvas style=”width: 275px; height: 38px; top: 0px; left: 1px;” height=”38″ width=”275″></canvas><span class=”cufon-alt”>Impressions</span></span>
</h2>

What’s wrong with Cufón?
Cufón works by converting a [presumably Web Embedding licensed] font into a vector file in plain JavaScript code. Although normal usage of Cufón means that only an extremely limited sub-set of glyphs are actually converted into JavaScript — each glyph added significantly increases file size, so it’s usually a good idea to limit the characters — it’s still a “copy” of the font and it’s freely downloadable. In order to legally use Cufón you need to obtain a special font license that allows for Web Embedding. According to the Cufón home page all Adobe fonts are safe for web embedding and many of the other font foundries offer Web Embedding licensing although none of them mention the actual price. Although every blog in the universe that ever reviews Cufón mentions the licensing as a big issue, the real problem is that companies like SolutionSet are going to have to start watching their font licensing more closely and we’re going to have to start forming policies for securing Web Embedding rights for the fonts we need to use. This isn’t really the roadblock it’s been made out to be; it’s just a change in process.

So, start using Cufón already!
Cufón is the best technology available on the market right now for using fancy fonts on the web. It’s easy to implement, it’s well documented, it’s SEO compatible. It’s also totally safe to use with Adobe fonts and it’s possible to purchase licenses for non-Adobe fonts. SolutionSet is going to being using Cufón on its next series of projects that would’ve normally used sIFR.

Free operating systems. Free content. Free apps. Free email. Free hosting. In an article published on Fast Company, author Farhad Manjoo questions the merits of the business model that dominates the web:

Why? Because free costs too much, weighed down with hassles that you’ll happily pay a little to do without. That’s why people buy bottled water and cable TV. That’s also the model that The Wall Street Journal uses to goad people into paying for news online. Anyone can read its stories for free through Google or a news-aggregation site like Digg, but people who want the full newspaper experience pay $103 a year for the privilege. More than a million subscribers consider that a good deal.

Manjoo notes customers understand that responsibility for product quality is only truly passed when money changes hands. If a free iPhone app is busted, you’re not too terribly surprised and will just delete it. If a $4.99 app is busted, you’re ready to write an angry letter. It’s all about expectations. If you can justify a cost with a promise of quality and support, then why mingle with the businesses in the free pool that can’t? Read the full article.

This month saw the sad passing of Patrick Swayze after a remarkable fight with cancer.  I was struck by the overwhelming response and postings to this news on various blogs, Twitter, and Facebook.  It was not just the *size* of the response, but the varied nature.  Women wrote nostalgically about Ghost and Dirty Dancing.  Men shared favorite/absurd moments from Red Dawn, Roadhouse, and Point Break.  When you think about it, what other actor can claim lead roles in first-ballot Hall of Fame “chick flicks,” great rewatchable-on-cable action movies, and a top-10 SNL skit?  I can’t think of any.

What does this have to do with advice for startups?  Absolutely nothing.

I was just looking for an excuse to pour back through Swayze’s greatest hits, maybe get a new generation to rent some of his DVDs, and shamelessly look for a theme for a blog post that might get reposted.

We do work with a ton of startups here at SolutionSet and I personally have founded, worked at, raised funding for, and helped build a number of ventures.  When I get asked for advice, here are the 5 most common themes I talk about… as told through some favorite Swayze movie quotes.

Tip #1 – Disrupt from the Bottom
“It’s kind of strange, isn’t it? How the mountains pay us no attention at all.” (Red Dawn, 1984)

My advice to startups usually starts with “Have you read Clayton Christensen?”  I will pass on trying to detail the concept of Disruptive Innovation, but there is one easy lesson to take.  Some of the most successful startups are ones that “disrupt from the bottom” by introducing a new, cheaper/faster/better product into the lower end of a market. These ventures succeed because the incumbent providers dismiss the threat of this low-end offering as being “insufficient.”  Think Flip video cameras vs. Sony, Salesforce.com vs. Seibel, Open Source vs. Proprietary Software, etc.

These ventures not only benefit from “flying under the radar”, but by building a base of real, paying customers allowing the venture to move up market steadily over time as they add new, more premium features.

Tip #2 – Build Something that Does One Thing You Love
“It’s not tragic to die doing what you love.” (Point Break, 1991)

Today’s product development environment moves at unprecedented speed.  To gain traction, a company should focus on delivering a single thing that a core of customers will rave about.  Don’t worry about feature complete, worry about being memorable.  The best way to achieve this is not just find that “one thing,” but to make sure you build something YOU would love to use.  Great companies like HP, Apple, and others were built on this simple, but demanding premise.

Tip #3 - Launch Early and Ugly
“You can learn to punch in the barn, but you gotta learn to survive on the ice.” (Youngblood, 1987)

This piece of advice flows directly from the previous.  It is hugely tempting as a new venture to want to wait to launch until your product is perfect.  You’ve poured your life into this new venture and everyone you know is waiting to see it.  You want it to be impressive.  Get over it.

No matter what you think your product will be, it will be different when the market gets a hold of it.  Take a deep breath and launch with the minimal acceptable product.  People will engage with it in ways you never imagined and that data will do more to inform your product strategy than anything else.

Tip #4 – Pick Your VC Wisely
“You ain’t seen bad yet… but it’s coming” (Next of Kin, 1989)

Okay, so now you have launched and have customers/traffic.  You will inevitably face the choice of accepting venture money.  Before you read too much into my quote, I’m not reflexively anti-VC.  They can be hugely helpful in the right circumstances… but there are a few key things to know.  First, choose the specific Partner, not just the firm.  Do your own due diligence on both the person and the firm and make sure you are comfortable with the values, reputation, and focus of each.  Talk to not just portfolio CEOS, but also VPs down the chain (who often will be more willing to share honest feedback.)  But do your research; the wrong decision can be catastrophic and the right one can help you build a much better business.

Tip #5 – Beware of the Big Customer
“He’s stuck, that’s what it is. He’s in between worlds.” (Ghost, 1990)

You’ve built a Version 1.0 product and raised money.  You are now getting introductions to customer prospects you only dreamed of meeting.  Seems great, right?  Beware… many companies have died at the hands of a won deal.

Big customers mean big money.  That’s a good thing.  But they also mean big demands.  Startups can get trapped in an endless cycle of dedicating all of their resources to a single “hail mary” customer.  Instead of building a product for everyone, they get taken off track building a ton of features specific to “Big Customer” that do not help them build a product for the larger market.

The result is a company stuck between two worlds.  You are a trying to build a product company, but instead you are functioning as a professional services firm working on a “project” for a client.  If you want to build a product company, resist building for a world size of one.

* If you liked this blog post, or even if you hated it, please consider donating to the American Cancer Society.

For a services firm like SolutionSet, hours billed equals revenue in the bank and improved profit margins. Hours worked but not improperly allocated or worse not tracked affect our bottom line negatively.  To help insure our financial success SolutionSet implemented processes, an organizational culture and most importantly a web based system, OpenAir (internally referred to as “OA”) to streamline the management and visibility of our business pipeline, active project billings, resource allocation and timesheet management  to name a few.

With almost two years of OpenAir under our belt we have a well oiled machine, not perfect but well oiled. The benefits of the web based system are endless, but it is the highly trained and technical customer support team at OpenAir that enables us to continually refine various modules and business processes that  impact our bottom line. As project managers, we found that the time tracked by staff, integrated with the project financials, made running reports to determine if we were on budget that much faster and easier.

There are several benefits to the system. In the first part of this posting, I’ll discuss the benefits of OpenAir to Project Management. In the second, I’ll discuss the benefits of OpenAir to other aspects of the company.

Project Management
We manage multiple aspects of a project from OpenAir. When we get a client lead, we put the client information into OpenAir as an Opportunity. When we get a project lead, we add it in as a Deal. Once the account managers solidify all the particulars of a contract with the client around the project, the Deal becomes a Project. The project contains several pieces of information that help project and account managers keep on track:

• Project information

This information includes the project name, what client the project is for, the estimated or fixed fee for the project, and who the account and project managers are.

• Phases/Tasks

Tasks are the services employees track their time to. This information is usually up to the project and account managers to determine. For the most part, it is usually the project role (project management, development, creative, etc.), but it can also be by milestone (requirements document, design signoff, etc.), or by phase (define, design, development, deploy). There is also the option to upload a Microsoft Project file into OpenAir. This allows for employees to track their time to specific line items within the schedule. However, this option is a bit cumbersome and gets confusing at times.

• Financials

As with any company, tracking financials on a project is very important. OpenAir allows us to indicate billing rules that automatically set up invoices to be generated by the accounting team. For fixed fee projects, it usually means a fixed amount will be billed when a milestone has been reached. For time and materials projects, OpenAir will read how many hours have been tracked to the project (as well as whatever expenses have been billed to the project, if any) and create invoices based on rates put into the system for that project.

It also allows us to estimate what our revenue will be each month. For fixed fee projects, we can estimate how much of the total will be used per month. It may not be an even amount for each month of the project, so the estimation comes from both experience with past projects and an estimation over time (example: if you start a 3 month project in the middle of July, you probably won’t be doing 30% of your project in July). For time and materials projects, we can estimate the revenue in a given month based on how many hours the client has booked for the said month. These numbers then get updated at the end of each month to the actual revenue amount.

• Reports

OpenAir has many reports and most of them are customizable to however you wish to use it. We can also save reports for future use, and share them with others. Each department (project managers, account managers, resourcing, accounting, etc.) has a set of reports all their own. The reports are essential to managing any project within OpenAir. One of the most important reports is to determine how much time employees have spent on a given project. Others we use often include how employees are booked on projects, what availability employees have over time, and how projects are looking from a financial point of view (which is tied to how employees are booking their time).

OpenAir has made managing project financials much easier for the Project Managers at SolutionSet. In the next posting, I’ll discuss how it aids in other aspects of the project.

The best way to develop an excellent customer experience is to use input from actual customers.  Conducting usability research will allow you to explore and/or validate design options and focus on knowledge that enables you to meet the needs of users/customers.

So, let’s say that you would like SolutionSet to perform usability research on your behalf, but how does it work and how can you get the most out of observing each session?

Typical usability research methodology

Unlike focus groups, usability research is typically a one-on-one sessions, set up to gather users’ natural reactions, without outside influence. A trained moderator will spend the duration of the session in the room with the participant and observers watch the study from a separate area. There are many logistical options for observation rooms and recording technologies. It really depends on your research objectives and budget. Learn more about budget usability options from a previous blog post.

Participants will be asked to complete tasks with the website or prototype that we would like all users in the real world to be able to accomplish without difficulty. The participants will be assured that the website is the center of attention and that they themselves are not being tested in any way. They will also be encouraged to think out aloud to facilitate the recording of their thoughts, expectations, and reactions to the pages.

Getting the most out of observation

Users often express opinions that do not coincide with our expectations. Study participants sometimes are confused about your company’s offerings, have difficulty completing tasks, or they criticize the appearance of a page. These reactions can be the most important data we collect! It is better to gather an unscathed and fully honest assessment of our work now than to release products into the public where the same confusion may occur.

Observing can be a mentally taxing exercise, especially for those who attend all the day’s sessions. Please maintain focus, keep conversations to a minimum, and take notes you would like to share with the group during the debriefing.

What to observe

Every signal the participant gives counts! Hand gestures, body language, facial expressions, and pauses in speech all count in addition to what the participant explicitly says and does. These tacit signals sometimes signify areas of improvement we will be able to act on.

Make note of similarities in participants’ feedback, for example when multiple participants are experiencing the same pitfalls or expressing similar criticisms. These commonalities typically identify the most obvious areas for optimization.

Record questions – both for the participant and to the group – that arise in your mind during the session. Near the end of a session, time permitting, the moderator will enter the observation room to gather questions for the participant during the wrap-up portion.

Debriefing

The purpose of the debriefing is to come to a mutual understanding of what occurred during the session. We would like to discuss and identify what you found to be really impactful and help us prioritize findings from a business perspective. This is a rewarding but sometimes arduous process, as there often may be multiple interpretations of the participant’s reactions.

Thanks for observing, a successful usability research project depends on it!

Some nuggets from a retail analyst appeared in an otherwise grim New York Times article covering back-to-school results. Retailers who are succeeding in this tough climate are doing so by coming up with innovative and fun promotions, like the Zumiez “5 for $85” deal. From the source:

John D. Morris, a retailing analyst with BMO Capital Markets, said more mall stores, like Gap and Aéropostale, offered buy-one-get-one-free sales. Others offered breaks on bundled merchandise, Mr. Morris said. He said that the teenage apparel chain Zumiez rolled out a “five for $85” promotion — three T-shirts and two pairs of jeans for $85.

“That deal started at $80 and it went to $85,” Mr. Morris said. “That tells you it worked. That tells you it is registering with the consumer.”

At Gymboree, customers could take 30 percent off whatever merchandise they managed to stuff into a Gymboree bag.

“The retailer is learning the native tongue of the new consumer,” Mr. Morris said. “It’s not only that you need to dangle value in the face of the consumer. It also tells you that the consumer is bored with the same old promotions. The smarter companies are going in and changing the marketing message.”

This made us think about the endless me-too promotions that are crossing our desk from direct marketers—Free Shipping! 20% Off! Buy One, Get One!—blah blah blah.

Direct marketers have a great ability to tailor their promotion to specific groups or even individuals. Far too few do anything with that power. So, since we are all going to be forced to promote this Holiday, why not use that promotion power to manage the customer life cycle?

How about getting some cross-category purchasing going by offering our apparel-only buyers, “Buy anything in our homewares section, and take 40% off first item”? How about offering our new customers a no-conditions $20 coupon for making their second purchase? How about acknowledging our best buyers and offering them concierge level treatment, lifetime unconditional returns and the like?  Wouldn’t we like to come out of this recession with more multiple buyers and fewer lost best customers? Since you’ll have to promote to stay aloft, try to do it creatively and strategically.