Achieving Your Profitability Goals

From this article, you’ll learn how to determine the efficiency of your online advertising efforts and how to calculate the maximum cost per click for those campaigns.

The basic economics of online marketing are simple: Determine the advertising efficiency needed to make your profitability goals, then buy all the advertising inventory you can get your hands on that meets your efficiency goal.

But how do you determine the advertising efficiency needed to achieve your profitability goals? This article offers some practical formulas and advice.

Defining Online Advertising Efficiency

Ad efficiency comes down to a cost vs. benefit ratio: “What did I spend on advertising?” vs. “What did I get in return?”

The “What did I spend on advertising?” side of the equation is straightforward. The ad venue sends invoices; your accounting department tallies the funds you paid during a period.

But make sure the cost figures you use to compute efficiency match actual invoices paid. Some agencies and advertisers estimate costs via their own in-house click counts and average cost-per-click (CPC). These click and CPC estimates can differ from the authoritative costs provided by the search venue. Conduct occasional audits to ensure that the cost side of your online reports ties to your online ad bills.

The “What did I get in return?” side of the equation is more complex. Most marketers seek profit as their ultimate online marketing goal. Yet accurate profit numbers may be too hard to obtain or may be generated too far in the future. When this is the case, employ meaningful proxies such as cost per lead, cost per registration, cost per order, cost per new buyer, advertising-to-sales ratio, and margin-to-cost ratio as surrogates for actual profit.

Establish an Advertising To Sales Efficiency Target

Let’s continue using the advertising-to-sales ratio (A/S) as our profit surrogate. What’s the right A/S efficiency for your online business? Rule of thumb: Catalogers typically spend 80 percent of sales on cost of goods and marketing combined. The profit and loss statement logic here is that if marketing and cost of goods sold consumes 80 percent of sales, and variable selling expense and overhead each consume another 7 percent or 8 percent, then about 5 percent is left as pretax profit.

So a gift cataloger with a 50 percent average margin may select a 30 percent A/S for its ad efficiency target. A retailer with slimmer margins — say computers or electronics — would choose a lower A/S target.

If you sell products with widely differing margin structures, you may be wasting an opportunity if you apply a single A/S target to all your advertising efforts. Setting a single target may lead you to over-advertise your lower-margin products, while under-advertising your higher-margin products. To combat this, either fold actual margin data into your online marketing management process or apply different A/S targets at the product-category level.

Measuring Online Sales

The denominator of the A/S ratio is sales dollars. But there’s more than one way to measure sales:

• Web demand sales measures sales as seen by your Web site.
• Booked sales measures sales that actually make it into your bank account (demand sales less frauds and cancels).
• Net sales reflects sales that actually stick (i.e., booked sales less returns).

These sales measures come bundled with various time lags, as frauds and returns take days and weeks to unfold.

For simplicity, most online retailers use Web demand sales — that is, sales as reported by the Web site — as a reasonable proxy for true sales. If the ratio of Web demand sales to booked sales is relatively constant during a period of time, track demand sales and use this ratio to accordingly adjust your financial models.

Sometimes this ratio isn’t consistent over time. For example, some online jewelry merchants experience large fluctuations in fraud attempts week to week. As frauds are detected after orders come through the Web site, they’re included in Web demand sales, making this measure less than accurate. In such cases, back out frauds from your Web demand sales.

Online Marketing Channel Spillover

Just as many retailers deflate Web demand sales by a percentage to account for frauds and cancels, many then inflate the result to account for spillover into their call centers and retail stores.

While a few firms (e.g., Ingenio, Jambo, Voicestar, eStara) and search engines offer technology to track online clicks at the ad level into the call center via distinct toll-free numbers, these solutions still are new and costly and have yet to gain significant traction. But they will. Web-to-phone tracking and pay-per-call advertising will become essential for all online direct marketers during the next few years.

In the meantime, many online marketers use monthly call center surveys to estimate call center spillover (”May I ask how you heard about us today, sir?”). A handful of sites display different toll-free numbers to track clicks into the contact center at the gross engine level (i.e., visitors to the site from Google see one toll-free number, while those from Yahoo! see another).

Even fewer online direct marketers track store spillover. There are two reasons for this. First, many online direct marketers have no (or quite small) retail store networks — think Amazon, Overstock and most catalogers. Second, many large, national store retailers with strong online direct divisions still suffer from internal political struggles regarding channel importance and sales allocation.

This will change. In 2006, expect to see more national retailers providing Web visitors with offers and discounts via PDF barcoded coupons redeemable at retail checkouts. During the next few years, if you have stores, expect to be tracking the impact of specific Web promotions on your retail store sales.

How to pay for a click?

Suppose you’ve determined an appropriate A/S efficiency target for your business, either overall or by product category. And suppose you understand how to adjust reported Web demand sales to mirror actual booked revenue — some downward tweaks to account for frauds and cancels, some upward tweaks to account for contact center spillover. Given accurate click counts, you can divide adjusted sales by clicks to compute average sales per click.

The relationship between what you can pay (max cost per click, or CPC), how your Web site performs (sales per click, or SPC), and your target efficiency (A/S) is simple math:

max CPC = A/S x SPC

For example, consider a retailer seeking a 35 percent ad-to-sales efficiency with a Web site that sells $120,000 online to every 100,000 qualified visitors. After tweaking Web demand sales to account for frauds, cancels and contact center spillover, she estimates her adjusted SPC to be $1.30. To hit her efficiency target of 35 percent, she must pay on average no more than 45 cents per click for those 100,000 visits.

Warning: Don’t Manage PPC Ad Campaigns by Averages

While averages are useful to describe portfolio economics overall, they’re not helpful for the actual management of your online pay-per-click (PPC) search efforts. Averages apply to populations, not individuals.

Your actual SPC estimates will vary widely across your keyword portfolio. Different terms will have different SPC measures. And when you delve into it, you’ll find these term-level SPCs will vary with the ad copy, destination URL, season, day of week and time of day.

Returning to the previous example, that retailer may be able to afford an average 45 cent CPC across her portfolio. But many of her ads wouldn’t meet her efficiency target even at a penny per click, and a handful of her best ads easily would beat her A/S target, even at several dollars per click.

With online PPC advertising, the devil (and the profits) are in the details. Make sure your programs are managed at the keyword level, with smart bidding adjustments for copy, landing page, seasonality, time of day and day of the week.

Pay-Per-Click New Customer Acquisition

Traditional catalogers know that buyer response rates drive profits. Accordingly, many catalogers spend more aggressively acquiring new customers, often running acquisition efforts at breakeven (plowing all gross margin obtained from new customers into marketing to new customers).

Track what fraction of your online customers are new-to-file. If you’re not doing so already, flag these names in your merge/purge to see how they dupe out against your housefile. If your online marketing efforts bring in a satisfactory rate of new customers, and these new customers have solid, repeat purchase rates, you can grow your business more rapidly by granting your online programs a lower efficiency (that is, a higher A/S), which in turn raises your maximum allowable CPC, letting you buy more traffic.

Online advertisers come into the online ad marketplace with their own goals. Some seek to build brand; some want sales growth, others look for profit. Whatever the goal for your company, determine appropriate efficiency targets and track your results carefully.

We here at RKG like to read. We’ve found these 9 books particularly useful and wanted to share this recent list. Some of these titles are new, some are classics, all are recommended. Happy reading! (Nov 2005)

Business & Marketing Books

The Search: How Google and Its Rivals Rewrote the Rules of Business and Transformed Our Culture by John Battelle
Looking back on Google’s short history, and forward to the future of media, advertising, and privacy.

The Cluetrain Manifesto: The End of Business as Usual by Christopher Locke, et al.
Somewhat giddy with pre-crash breathlessness, this book shows its age. But Locke et al were right: markets are conversations. Interesting tidbit: the book doesn’t even contain the word “blog.” Worth revisiting, if last read in ‘01.

The Wisdom of Crowds: Why the Many Are Smarter Than the Few by James Surowiecki
Under the right conditions, groups are smarter than all their members. Fascinating discussion of prediction markets. Relevant to decision-making processes in companies.

My Life in Advertising and Scientific Advertising by Claude Hopkins
Direct marketing advertising 101 from one of the field’s founders; as relevant today as in 1923.

Web Measurement & Usability Books

Web Site Measurement Hacks by Eric Peterson
The rare must-read for both marketing and IT types. Extremely useful code snippets and essays to help you better understand your visitors and how they use your site.

The Unusually Useful Web Book by June Cohen
More broad than deep, this is the perfect book to toss on the desk of a new employee to give a lay of the land of web retailing and site creation.

Paper Prototyping: The Fast and Easy Way to Design and Refine User Interfaces by Carolyn Snyder
Idea: test new site layouts using Post-It stickies before expensive coding. (Staples must love her.) Sure, this could be covered in 1/3 the space, but “office-supplies-usability” is a powerful tool.

Information Technology Books

Speed Up Your Site: Web Site Optimization by Andrew B. King
Web marketers pay lip service to the importance of site speed, but few commit to the changes required to gain the improvement. King shows how to trim bytes from pages, leverage CSS, and use HTTP compression. The best book on this underappreciated topic.

Extreme Programming Explained: Embrace Change (2nd Edition) by Kent Beck, Cynthia Andres
Can web IT be nimble, fast, flexible, and reliable? The XP folks argue “yes.” Radical, threatening to the status-quo, XP offers web marketers a development method which matches the urgency of online retail. One of the best software ideas of the decade. Now, could you make it fly politically?

Is your search advertising program healthy? How do you know? If you outsource, is your agency doing the job well?

It would be nice to assume that hiring an expensive specialist guaranteed a clean bill of heath, but malpractice suits are often well-justified.

Just because your search results look good in aggregate doesn’t mean the program is healthy and running at maximum efficiency. Once a year, or more often, a detailed evaluation of your program is in order. It will pay big dividends in the long run.

Gathering data on your search campaigns

Here’s how to do it. First, gather data on:

Keyword-level costs and sales. Your in-house search manager or search agency should be able to easily provide you with clicks, costs, orders and sales dollars by keyword phrase and engine for the last month or two. Don’t be content with the “top 50″ or “top 100″ terms, you need to see everything. If it isn’t easy for them to pull this together quickly, then one of two things is probably true: 1) they aren’t watching the important details of your program; or 2) they’re hiding the details because they’re ugly. Wave the red flag — your program is broken.

Google and Yahoo term lists. Have your in-house manager or agency provide the current active terms with the creative (ad text), the destination urls, the current bids and the current position (average position over the last week, or day is fine). Again, this should be quite easy to access.

Analyzing data on your search campaigns

Brand vs. non-brand. Sort the keyword-level cost and sales data alphabetically by keyword. What fraction of your sales is coming from your brand name and its permutations (store name, storename.com, www.storename)? When you look at the costs and sales of these brand terms, and more importantly, the cost and sales of everything else, do the numbers make economic sense? Don’t let sales on your brand name subsidize poor performance on the non-brand phrases.

Bid management. Sort the same spreadsheet by cost descending, so that the most expensive terms come to the top of the list. Search is a “spiky” business, and it’s important to look at a month or two worth of data, to eliminate some of the spikiness.

Looking at your top terms

For the twenty to fifty most expensive terms, pull back the current bid and position from the appropriate term list sheet. Do the bids and positions make sense for these critical terms given the economic results you’re seeing? Are there instances where you’re bidding a lot on poor performers? Are there instances where you’re bidding too little on great performers and you’re not high on the page?

Bidding on mid- to low-traffic terms is a complex statistical game; bidding on the high traffic terms isn’t. Your in-house manager or agency should be able to provide clear, jargon-free explanation about this bidding logic: If the bids don’t make sense on the most important terms, they probably don’t make sense on the rest.

Bidding logic

Now examine the rest of the bids. Delete the “brand phrases” from the performance data, as these phrases distort the real picture. Re-sort the list by click-counts descending. Look at the sum of costs and sales for terms with more than 1000 clicks. Repeat this process for terms with between 900 and 1000 clicks. Repeat all the way to the bottom, ending with terms with fewer than 100 clicks. Bid management for the lower traffic terms is tough, but if each “bin” of terms doesn’t have reasonable efficiency, that may indicate problems with the system. It’s easier to break a fine watch than it is to build one, but if you’re paying a lot for the watch, it ought to withstand this scrutiny fairly well.

Evaluating your term list

Look at your term lists. How thoroughly do the term lists cover your offerings? A decent rule of thumb holds that you should have approximately 5 to 10 keywords for each product you sell. But numbers aren’t the ultimate measure, as it’s easy to inflate a poor list by simply appending or pre-pending “shop for”, “buy”, “online”, “on sale”, etc. to each term. Creating a truly comprehensive list with permutations that actually help capture more traffic is hard work, but well worth the effort. Remember: low traffic, detailed terms convert at a much higher rate than more general terms and are often less expensive.

Evaluating your ad copy

Look at your ad copy. Is the ad copy compelling, accurate and specific enough to the product? It usually isn’t cost effective to hand write copy for every single ad, but the templates used should be at least category specific. And, it may well be worthwhile to put extra attention into your top 50 to 100 most viewed ads.

Evaluating your destination URLs

Look at your destination URLs. Grab a random selection of phrases and copy and paste the destination urls into your browser. Is that the page you would pick to land your customer on? If not, find out why the landing page is what it is. Getting the landing pages wrong costs you sales, as many shoppers aren’t patient enough to navigate more than is necessary.

Diagnosing your search program

If you get to this point and everything looks good, congratulations, your search program is in good shape. If not, depending on the severity of the problems uncovered, ask some more hard questions or look for a new provider. If you can’t do the examination because you can’t get keyword level performance data, fire your agency or help your in-house manager get the tools they need to do the job well.

Search is complicated, and getting every detail perfect is impossible, but if you uncover grievous problems take action. There’s too much at stake to tolerate poor search health.

George Michie is vice president of client services and Larry Becker is vice president of marketing and business development with The Rimm-Kaufman Group, a search marketing and online retail consulting firm based in Charlottesville VA.

Almost any question can be answered, cheaply, quickly and finally, by a test campaign. And that’s the way to answer them—not by arguments around a table. Go to the court of last resort—the buyers of your product.

—Claude Hopkins, Scientific Advertising, 1923

Savvy catalogers have long used testing to improve their mail businesses. As the web matures, catalogers are bringing the same discipline to their online marketing efforts.

This article offers 10 tips for running direct marketing tests in the online world. The first 7 are common to online and offline. The last 3 are unique to online marketing, for they exploit the web’s high speed and low cost.

Tip #1: Move Bigger Levers First: List, Offer, Creative

In descending importance, the three essential elements of a direct marketing campaign are “list” (who, and how many, people receive the offer), “offer” (what merchandise you offer those people, at what price, and with what service), and “creative” (how is the merchandise presented, described, and displayed.)

When testing, always move the bigger levers first.

If you goal is to double online sales, your best bet is doubling qualified traffic to your site. This is generally easier than doubling average order value or doubling site conversion. (Worthy goals, too, but harder to achieve.)

If you’ve not yet tested paid search, paid inclusion, local search, affiliates, Ebay, Amazon, and so on—get out and do so. Such “list” tests offer you the greatest chance of really bumping sales.

After “list”, focus on “offer.” Is your site presenting the right merchandise at the right prices? What about shipping fees: would the conversion lift from a free free shipping offset the cost? Suggestion: when setting a minimum order size for an offer, place it above your average order size.

Finally, focus on your creative—how your site looks and works. Does your homepage highlight the breadth of your merchandise? Are your product detail pages clear, with relevant information above the fold (visible without scrolling)? Is your checkout processes smooth, fast, and intuitive?

Tip #2: Test Shouts, Not Whispers

Testing takes effort, attention, and sometimes money.

Don’t test tiny tweaks.

Favor bold tests that have the potential to really change your business. Suggestion: a sure sign of a bold test is that it may make some insiders slightly uncomfortable.

Subtle tests will, at best, yield subtle results, often too small to detect.

Tip #3: Keep Test Notebooks

For each test, document what you tested, why, and what happened. Short pre-test and post-test summaries keep you from repeating mistakes or wasting time on questions already answered.

Before the test, write down a clear hypothesis of what you’re trying to prove or disprove. Here’s an example:

Test #6, October 2005: Our hypothesis is that bringing visitors into our site from paid search to the new simplified product page template will increase conversions relative to the current grid-style product page template.

Before the test, also record your decision metrics and the roll-out plans.

If the new pages increase closing by a significant amount, that is 50 more orders than the control for the week, we’ll discard the grid template in favor of the simpler template.

After the test, record your numeric results, your interpretation, and suggestions for next steps.

Results: Despite one large order, the simple treatment stunk, actually reducing conversion a bit. Next steps: keep the grid, test another challenger later this month.

Given the value of test notebooks—indeed, they become as your marketing department’s shared institutional memory—its worth maintaining two copies.

Tip #4: Test One Thing At A Time

To isolate the effect of a variable, traditional testing mandates changing a single factor at a time.

Testing online is usually cheaper and faster than a traditional in-the-mail test. Because of this, there’s less need to cram everything into one massive test. Start with a series of fast, simple one-factor tests—you’ll quickly learn what matters.

More advanced marketing teams should look into multivariate testing (also known as MVT, scientific testing design of experiments, or Taguchi testing). MVT offers marketers the chance to vary many factors at once in a statistically valid way.

MVT test design and analysis require more skill—a little training for your team from a seminar or statistical consultant goes a long way here.

Tip #5: Separate Signal From Noise

All tests have some element of random statistical noise. Suppose you took a mailing list of 10,000 people, randomly split it into two cells of 5000 people each, and on the same day mailed each cell the same catalog. Even with exactly the same treatment to both groups, one cell by chance alone will have a few more orders, and thus a higher response rate.

Be sure you can distinguish marketing signal from marketplace noise. Familiarize yourself with basic statistical significance calculations. As a very rough rule of thumb, if you plot conversion rates over time, a test needs to increase conversion by more than 1.5 times the normal range of variability to be significant.

If your team isn’t using these statistical significance formulas yet, you can get training from the DMA (http://www.the-dma.org/seminars/statistics/), or we’ve put a Excel spreadsheet with these basic formulas on our website (http://www.rimmkaufman.com/statistics)

Tip #6: Assign Unique Tracking Codes

Online testing without proper tracking is like flying an airplane blindfolded.

To be able to read your tests, each cell needs a unique tracking code. Make sure your tracking application can report key metrics for each code—visits, sales, and ad expense.

Make sure your tracking application allows you as many keycodes as you need.

Some marketers embed meaning in the various digits of the code, where a code like “0510E02″ might mean the second (”02″) email test (”E”) sent in October, 2005 (”0510″).

Other marketers use sequential numbers for their cells, and log the meaning of each code in a spreadsheet or database.

In either case, make sure your team maintains scrupulous electronic records matching keycodes to their corresponding cells and campaigns, with accompanying information regarding the creative and offer. Many a test has been lost to keycode mismanagement.

Tip #7: Manage Expectations

Many ideas when tested will prove to be duds. Be patient and try again. Make sure your team understands that most tests produce null results. Because your current marketing approach represents years or decades of thoughtful improvement, many alternatives won’t test out better. And the more successes your testing program obtains, the harder it becomes to move the needle.

Should the difficulty of hitting home runs stop you from stepping up to the plate?

Not at all. While testing wins may be scarce, the big wins than can follow more efficient marketing are well worth the effort.

Tip # 8: Online vs. Offline Testing: A Difference in Kind, Not Degree

In the offline world, versioning is expensive and often limited. You face physical limits on what you can vary due to printing and bindery constraints.

In the online world, versioning is usually inexpensive, often free.

Suppose you’re testing an holiday sales email. You might have six decent ideas for the subject line, two possibilities for the large seasonal image, and three reasonable candidates for the lead “hero” product. Considering all combinations of subject, image, and product, you could send 36 different email versions (6 x 2 x 3).

Which version will prove most effective?

If your testing framework can support it, and if the creative can be built in a “cookie-cutter” format, then send them all.

This brute-force “shotgun” approach can be surprisingly effective at finding winners.

(Note this isn’t multivariate testing described in #4 above—MVT would attempt to test all 36 version with only 12 or so cells.)

Tip #9: Testing Within, Not Between

In the offline world, tests are typically slow events, requiring time to prepare, launch, and analyze.

With online testing, results come immediately, often within hours. Smart marketers can exploit this early information to test within a campaign, rather than just between campaigns.

Here’s how.

Suppose you mail a monthly email to your house file. Getting the subject line right can have a large impact on open and conversion rates.

Many catalogers test multiple subject lines on each mailing. Then, after the campaign, the marketing team gathers over coffee to review results by subject line, learning what worked best, so as to improve future emails.

Instead, pull off a random 10% of your email file to mail first. Randomly nth these emails across the various subject line ideas, and mail them. Wait six hours. Evaluate the subject line ideas using their 6-hour open rates for initial blast. (You’d prefer to use conversion rates, but for many catalogers six hours is likely too short to see meaningful sales.) Then—and here’s the trick—mail the remaining 90% of your file using the winning subject line from the initial blast.

This allows you to reap the benefits of sending the better email immediately, not down the road in a future email.

Tip #10: Automated Testing

In the offline world, a test is a discrete marketing effort. Each test takes careful planning and execution by your marketing team. Testing is anything but automatic.

The online world is beginning to offer an exciting new opportunity: automated test robots.

Automated testing may not yet be supported by your e-commerce platform or any of your advertising partners. Within 12 months, they will.

Here’s how automated testing works. You load up a set of different marketing messages (be these ads on an advertising network, or featured products on your homepage) and instruct the platform as to which metric you’re seeking to optimize (such as click-through rate, clicks, or sales). The platform then rotates through the set of messages, tracking which performs best. As the platform learns which message performs best, it preferentially serves the winner, improving your results.

Early examples of automated test robots include Google’s “automatically optimize ad serving for my ads” adwords option (http://www.adwords.google.com), and Offermatica’s “mbox” platform (http://www.offermatica.com).

Online Testing Gives Insights

Claude Hopkins’ praise for testing rings as true today as when written 80 years ago.

May your online tests surprise you, bringing you fresh insights to grow your business!

Alan Rimm-Kaufman, PhD, leads the Rimm-Kaufman Group, a direct marketing service and consulting firm helping catalogers with online marketing and paid search. Alan may be reached via his website at http://www.rimmkaufman.com.

Direct Marketing Significance Calculator

Recent trends in online marketing

Big changes are afoot online.

It’s still early, but I see two interesting trends which will have large impacts on online marketing. The first trend involves the sharing of content. The second trend involves the sharing of applications.

These two trends haven’t fully arrived. They don’t have well-established names yet. But their early glimmers are visible today in the growth of RSS (Really Simple Syndication) and the growing popularity of web service APIs (Application Programming Interfaces).

These trends — let’s call them “open content” and “open apps” — are coming fast. Over the next few years they will revolutionize the web, and in doing so, revolutionize online marketing. This article describes why open content and open apps will matter to your business in the future, and how savvy catalogers can start using them today.

An expanded list of links can be found at the end of this article.

What is open content?

I’ll use the phrase “open content” to describe sharing your online content — articles, product information, specs, reviews, musings, customer feedback, etc — in machine-friendly format, usually XML.

The best example of open content today is RSS (Really Simple Syndication).

Using RSS, you can share information with customers, vendors, the media — in short, with the world. Because RSS is a computer friendly format, RSS makes it easy for others to search your data, fetch it, republish it, and use it in all sorts of interesting ways.

RSS and Blogging

Many people think of RSS as related to blogging. And indeed, blogging platforms use RSS to syndicate content across the web.

Digression: If you haven’t checked out the blog phenomenon yet, do so today. The blogosphere is growing at a phenomenal rate. Technorati, a blog search engine, reports the number of blogs has doubled in the last five months, reaching 16.4 million blogs in September 2005. For starters, check out www.bloglines.com, www.newsgator.com, www.feedreader.com, and www.blogger.com.

RSS beyond the Blog

But RSS extends far beyond blogging.

Here’s a grab-bag of some creative non-blog uses of RSS.

Newspapers use RSS to syndicate their articles (www.nytimes.com). Real estate agencies use RSS to list properties (www.citycrybs.com/). Coupon distributors use RSS to push coupons to shoppers (www.dealoftheday.com). News wires use RSS for releases (www.businesswire.com). Companies do, too (www.ibm.com/press). Schools use RSS to send information home to parents (http://www.udsd.k12.pa.us/rss/). Tech firms use RSS to distribute documentation (msdn.microsoft.com/aboutmsdn/rss). The State of Montana uses RSS for hunting advisories (fwp.state.mt.us/news/rss/hunting). The Red Cross uses RSS for disaster information (www.redcross.org/websites/rss/). There’s even RSS for tracking Britney Spears websites (http://www.iq451.com/music/rssfeeds/britney-spears-web.rss)

The list goes on and on. I’d wager more than half of all organizations with significant web content offer some RSS feeds today; and I suspect almost every such organization will do so within a year.

RSS as Open Content

As organizations publish mountains of great information via RSS, people search and read this information using RSS feed readers. This is all well and good.

But because RSS is a computer-friendly format, machines can read RSS too. That’s where things get really interesting.

With relatively little programming, you can build a software agent to watch for great real estate deals in your neighborhood. Or to see if today’s school lunch is something your child likes to eat. Or to monitor your key competitors.

This trend towards open content will happen with or without you. It can influence your brand even if you don’t participate directly.

Example: Open Content Can Influence Your Brand

Consider the case of Kryptonite Lock.

On September 12, 2004 a blogger described how the formidable black bike lock could be easily picked with a Bic pen. Within two days, other bloggers had picked up on this weakness, posting video demonstrations of the lock’s vulnerability.

Kryptonite issued some press releases, but the blog storm quickly spread. The story reached the New York Times on September 17^th. By September 19, Technorati estimated that almost two million online blog readers had seen some sort of post about the vulnerability.

To quell the storm, on September 22 Kryptonite announced it would replace all affected locks. The incident cost the company $10 million in direct costs and far more in loss of reputation — and this blog storm blow up in just ten days.

The Effect of Blogs on Your Brand

There are many other cases of bloggers harming brands (Mazda,Captain Morgan). There are also many cases of blogging helping brands (Scoble at Microsoft, Zawodny at Yahoo). If it hasn’t already, the blogosphere will exert influence on your brand, too, both positive and negative. Realize that in the next two years there will be thousands more external voices with the power to shape your brand than there was two years ago. Be ready. Be aware. Be a good corporate citizen.

Using RSS Today

As a catalog marketer, how can you use RSS to help your business today?

Before anything else, make sure you and your team have some familiarity with RSS and blogs. Start reading before you start writing.

Consumption and Creation

On the content consumption side, monitor RSS feeds for intelligence on your brand and your competition. Use RSS feeds to syndicate fresh news relevant to your industry onto your website.

On the content creation side, provide customers the option to read your marketing emails via RSS. (Big advantages: no spam concerns, highly trackable, and free!)

Use RSS feeds to announce the arrival of new products, special offers, and price reductions. And set up a blog to let your experts speak to and with your customers.

What are Open Apps?

The second important trend online trend today involves “open apps.”

Most important web sites are front-ends to powerful computer applications. Consider search engines, e-commerce, online banking, online games, and online travel reservations. The exposed public websites are simply human-friendly input layers for the applications beneath.

As with open content, things get interesting quickly when computers start speaking to other computers. APIs let this happen.

Using APIs

In short, an API is just a technical specification, a written document which describes how one computer can ask another computer to do something. For example, using Ebay’s API, you could have your computer instruct Ebay to post a new auction, without having to click your way through all the web screens.

Sure, programmers can write scripts to interact with web sites without APIs. These scripts impersonate a person using a browser. The approach is called “screen scraping.”

Using APIs to Avoid Screen Scraping

Screen scraping isn’t fun for two reasons. First, scraping is often disallowed by a site’s terms of use, and breaking rules never feels good. Second, scraping is difficult because it is brittle. As it relies on the exact details of the HTML source, scraping code breaks each time the site owner modifies their screens.

APIs (Application Programming Interfaces) solve this problem. If a site wants to grant visitors access to their applications programmatically, they provide an API to do so.

Who Uses APIs?

Many of the most interesting sites online today offer comprehensive APIs: Google natural search (www.google.com/apis/) and Google paid search (www.google.com/apis/adwords); Yahoo! paid search (http://searchmarketing.yahoo.com/af/yws_api.php), EBay (http://developer.ebay.com/common/api), PayPal (https://developer.paypal.com/), Amazon (www.amazon.com/gp/aws/landing.html), Salesforce.com (http://www.sforce.com), and the United States Postal Service (http://www.usps.com/webtools/), just to name a few.

Some of these APIs are public, some are not. Some of these APIs are free, some are not.

APIs to Control Web Services

Why are these APIs so interesting? Just as RSS allows your computer to read content, APIs lets your computers control web services.

Instead of a human struggling to manage bids manually on your paid search campaigns, a well-programmed computer can do it faster, cheaper, and better. Instead of using a human struggling to track all your shipments manually, a well-programmed computer can do it faster, cheaper, and better.

Computers excel at repetitive algorithmic tasks. Smart people excel at exercising judgment and handling special cases. The best results come from pairing a smart person with a well-programmed computer, each tackling the tasks they handle best.

Building Your Own Integration Systems

Today, your organization probably doesn’t have sufficient business need (or available IT resources) to commit to the costly effort of building your own integration systems to tie into your vendors’ APIs. And your organization certainly doesn’t have sufficient business justification to even consider creating your own API to let partners or customers interact with your internal IT systems.

But your some of your agencies might have built this integration layer. Your search marketing agency should be using the Google and Yahoo! APIs (and, soon MSN) to directly manage your advertising on your behalf, with smart people and well-programmed software driving those APIs. (Disclaimer: my firm provides such services.)

And some of your software vendors might provide this integration layer. Expect that as you upgrade your operations software over the next few years — your apps for web, call center, warehouse and accounting — the software that you buy may start offering web APIs.

These APIs would allow you to permit trusted partners to access your back-end systems. It will take a few years for such APIs to arrive, and a few more for businesses to invent creative uses of them, but both will occur. APIs will bring you closer to your vendors, and your vendors closer to you.

Merchandising and Media Buying

Just as open content will change marketing communications, open apps will also change merchandising and media buying.

In September, Google began testing ad brokerage for print publications, as well as moving into VOIP (Voice Over IP) telephony. I predict the large search engines will morph into general business brokerages, changing the cost structure of entire industries: television media buying, real estate brokerage, travel, and telecom.

Conclusion: The Future of Open Content and Open Apps

Open content and open apps will influence your business over the next decade. Approach these new technologies strategically. Start getting ready. Start reading. Start testing. Use and offer RSS.). Watch for APIs.

Like the inventions of the 800 number the credit card and the web browser, these technical innovations will reshape the catalog industry. Best of luck!

List of Links from this Article

• What is RSS : http://www.xml.com/pub/a/2002/12/18/dive-into-xml.html
• Bloglines.com : http://www.bloglines.com
• Newsgator.com : http://www.newsgator.com
• Feedreader.com : http://www.feedreader.com
• Blogger.com : http://www.blogger.com
• Nytimes.com : http://www.nytimes.com
• Citycrybs.com : http://www.citycrybs.com
• Dealoftheday.com : http://www.dealoftheday.com
• Businesswire.com : http://www.businesswire.com
• Udsd.k12.pa.us : http://www.udsd.k12.pa.us/rss/
• Microsoft.com : http://msdn.microsoft.com/aboutmsdn/rss
• State.mt.us : http://fwp.state.mt.us/news/rss/hunting
• Redcross.org : http://www.redcross.org/websites/rss/
• Iq451.com : http://www.iq451.com/music/rssfeeds/britney-spears-web.rss
• Fortune.com : http://www.fortune.com/fortune/technology/articles/0,15114,1011763-3,00.html
• Wired.com : http://www.wired.com/news/culture/0,1284,64987,00.html
• Wikipedia.org : http://en.wikipedia.org/wiki/API
• Google.com : http://www.google.com/apis
• Google.com : http://www.google.com/apis/adwords
• Yahoo.com : http://searchmarketing.yahoo.com/af/yws_api.php
• Ebay.com : http://developer.ebay.com/common/api
• Paypal.com : http://developer.paypal.com/
• Amazon.com : http://www.amazon.com/gp/aws/landing.html
• Sforce.com : http://www.sforce.com
• Flickr.com : http://www.flickr.com/services/
• Evdb.com : http://api.evdb.com/
• Usps.com : http://www.usps.com/webtools

Your Affiliates Should Not Compete Against You

Catalogers must continually generate new buyers to keep their house-file healthy. They’re always keen to discover hot new channels for acquisition, and to uncover tricks for wringing more new buyers from established channels.

Over the past five years, online affiliate programs have become increasingly popular among catalogers. Many of these catalogers also run search marketing campaigns. Managed wisely, search and affiliates can co-exist peacefully.

When these programs are left unchecked, however, many catalogers find themselves competing against their own affiliates, driving up their overall marketing costs, increasing their use of discount offers, decreasing their average margin, and possibly harming their brand. Here are some tips on taming your affiliate programs.

Use Selective Addition to Manage Your Affiliate List; Not Selective Deletion

Because an affiliate program requires strict guidelines that are consistently enforced, most marketers will find it more efficient to let the right affiliates in one at a time, rather than ousting affiliates after they have done harm to your marketing results or your brand. The top 5% of your affiliates will generate 95% of your sales. Evaluate the economics of choosing to ignore all the smaller affiliates – you may find they take too much work for their paltry return.

Don’t Let Your Affiliates Bid on Your Brand Name

Google currently only permits ones advertiser per keyword per domain: When the keyword is your brand name and the domain is your site, you, not your affiliates need to win that placement.

Some retailers hypothesize that it’s sufficient to be number one in organic results for their brand name. In fact, however, there’s proven incremental value in being visible in both paid and unpaid search results.

Allowing your affiliates to advertise against you on your brand can increase your cost without increasing your sales. Don’t let them. Mid-sized retailers report cost savings of $50k and more from this simple change.

Examine the Core Economics of Your Affiliate Program

When an order results from a paid search ad placed by one of your affiliates, you may pay three intermediaries: the search engine, the affiliate network, and the individual affiliate.

For the marketer doing paid search without affiliates, at least one middleman is removed.

In the case of a cataloger running their paid search in-house, there is only one entity to be paid: the search engine serving the advertising. In the case of a cataloger running their paid via a search marketing agency, there are two entities collecting fees: the search engine for the advertising, and the agency, for managing the campaigns.

Even without studying the numbers, fewer middlemen usually means greater profit.

Study your numbers. Understand what value is being added by each link in the affiliate chain, and the expense of generating that value.

Make Sure your Affiliates are Providing Incremental Business

Your affiliates should be helping you tap new veins of customer acquisition. Ideal sources include churches, synagogues, PTOs and PTAs, clubs, schools, as well as professional and community organizations.

It’s counterproductive for affiliates to simply be competing against you for the same pool of online customers.

Insist on Transparency

Avoid the “black box.” The methods your affiliates use to attract new customers should be completely transparent to you. Where are the affiliates advertising your site? What do those pages look like? Are the affiliates using your offers correctly? Is their presentation of your company brand-appropriate? Allowing affiliates to advertise on inappropriate sites can lead to embarrassment or worse.

Reduce Reliance on Discount Programs

Many affiliates will drive traffic to your site with ads offering discounts and coupons. Retailers who allow their affiliates to market this way should revisit this decision for at least two reasons.

• Cannibalization: When an affiliate offers coupons in response to searches on your competitive terms, or worse, your brand name, you often sacrifice margin on orders you would have received without the discounting. It’s unlikely that you would pay someone to hand out discount coupons in the lobby of your retail store. Likewise, allowing your affiliates to distribute coupons on paid search rarely makes economic sense.


• D.I.Y: If discounting is part of your overall marketing strategy, then you can probably execute more effectively by doing it yourself. Affiliates realize that many people begin their online shopping with a search on the phrase “[your brand name] coupon.”

  Working with an in-house team or paid search provider, you can craft your own discount ad copy and destination pages. You can then test the effect of these campaigns against a range of other approaches within your advertising program. And you can do so without paying a commission on a discounted sale.

  Giving up margin hurts. When you must do so, decide where this tactic makes the most sense. For example, rather than offering online discounts to new customers as a result of a loosely managed affiliate program, catalogers may choose to invest in a reactivation program.

The bottom line? Know what your affiliates are actually doing for you, how they’re doing it, and what you’re paying them for. With this knowledge, you can do more of what is really working, and less of what isn’t.

Break out Brand vs Non-Brand Paid Search Economics

Digging into the details can help you win sales from each advertising dollar. To maximize return on their paid search efforts, savvy direct marketers slice and dice their reports, analyzing their results by phrase concept, copy, landing page, engine and time of day. The resulting insights help retailers spend more on the advertising that performs well and less on the advertising that doesn’t.

One of the most important analyses a retailer can perform is the breakout of brand vs non-brand paid search economics. As you drive your program to its target metric, evaluate its success excluding the effect of the sales and cost associated with sales on your brand.

Don’t Let Your Brand Name Searches Mask Poor-Performing Terms

By brand we mean your brand itself (”Toyota”), as well a variations, misspelling and domains (”Toyota motors”, “Toyota USA”, “Tyota”, “Toyota.com”, etc.).

The brand vs non-brand breakout is significant for every marketer. How to approach it depends on the nature of your business.

For all retailers, searches on your brand are often among your most profitable campaigns. Indiscriminately rolling brand name results into your programs overall results can mask the performance of less effective terms. This leads you to continue to over-investing in poor performing terms.

Marketers working with third parties on paid search should make sure their partners also pay heed to this distinction. Your affiliates and your search provider will always be happy to report strong aggregate results. It’s up to you the retailer to determine how much of the good news should be credited to the messenger.

Catalogers Should Recognize the Influence of Print

If you’re a cataloger, searches on your brand name are a product of the catalogs that you mail and any other offline advertising you buy. Therefore, to truly evaluate the success of your paid search program, you need to report on these ads tied to your brand, separately from your other ads.

Failure to do so can lead to inefficient cross-subsidization among your marketing programs, with your catalog paying the true costs associated with brand –phrase orders credited to paid search

Some catalogers may in fact be willing to tolerate some subsidization by their brand name terms, allowing their print media to offset a portion of paid search cost. Likewise, some marketers may choose to allow brand-name phrases to bankroll bids on more competitive terms. But the decision to subsidize should be always be made deliberately not due to lack of visibility.

Brand vs. Non-brand for Non-catalogers

For non-catalogers, the brand vs. non-brand breakout is less clear cut.

If you’re both a direct retailer and a manufacturer, and you compete against your retail resellers your brand name is simply a “word in play.” As with all other phrases associated with your business, you will be bidding on this term alongside all your resellers.

The same holds true if your company’s name contain a generic term describing your good and services. For example, a retailer of men’s trousers doing business as “Best Possible Pants” could not attribute orders on the phrase pants to people searching for his brand.

The bottom line? When analyzing your search marketing campaigns, report on results with and without your brand terms. Partitioning your analysis in this way will likely provide you key insights on what is working and what isn’t.

Bidding For Position Is Dangerous

Faced with the bid spasms of some of their competitors, marketers may be tempted to measure their success by the most readily visible metric: Position on the page.

But bidding to position is dangerous. Clicks typically do rise with position but sales-per-click may not, and worse, this approach to bidding ignores data that is essential to performing successful bid calculations.

If you’re Coca-cola, the size of your Google spend relative to your investment in , say, TV, is small enough to let you go for position 1 at any cost . But for the typical direct marketer, a more nuanced approach is called for.

Avoid Dangerous Twitch Bidding

Responding to a competitor’s bid spasms can be just that: a dangerous twitch very much like the movements favored by stock market day traders. These momentum investors often flame out in the long run.

If you want your paid search program to profitably endure, it pays to adopt the measured stance of a Warren Buffet, or his Columbia University mentor Benjamin Graham. Applied to search marketing, the mode of the value investor dictates taking a scientific approach to your own business and the competitive marketplace.

Bidding Wars Benefit the Engines, not Retailers

There’s neither art nor science to bidding for position. Indeed, the engines make it easy; Within Yahoo! Search (formerly Overture), the dashboard lets you do just that.

But bidding wars benefit the engines, not retailers: Convince more than 3 advertisers than they need to be in one of three top positions and the race is on.

This kind of competition can mean economic devastation for your search program.

Consider this comparison to the twentieth century arms race: The former Soviet Union had a nuclear arsenal capable of razing the world many times over, but locked into competition, it kept upping the ante, eventually bankrupting itself. Worth keeping in mind the next time you’re tempted to race a competitor up the page.

The smartest marketers realize that just as in their offline programs, success in pay per click advertising depends on managing to target efficiency.

As always, applying the fundamental economics of your business is key:

For retail sales, subtract COGS (Cost of Goods Sold,) ad cost, and variable expense to compute ad-level profitability. (For businesses using paid search for lead generation, divide ad cost by lead to compute cost per lead.)

Audit Your Data

Given the foundational role of cost, make sure your data is accurate—or even relevant.

Tip: Make sure your internal team or search provider is reconciling ad cost with the authoritative data provided by the engines themselves. If you or your vendor rely exclusively on internal data to report on clicks and cost, you’re not capturing the reality seen by the engines, upon which your actual search advertising fees are based.

Govern Your Program with Relevant Rules

With fundamental profitability metrics in place and all data including cost, audited for accuracy, your program is ready for more precise adjustments. How will you deal with the debut of new terms that have yet to receive an order? Where is the threshold for lowering or raising a bid, and within what window of time?

The answers to these questions matter. For example, consider a retailer of high-end goods for whom conversion follows careful consideration. Suppose their target A/S ratio is 20%. Suppose that in week 1, a given term incurred $200 of cost with $0 of associated sales. Should bids for the term be decreased?

Suppose that in week 2, the same term again incurred $200 of cost and but generates a $4000 order. Depending on the retailer’s interval for term evaluation, the A/S ratio is either infinity or a very attractive 10%.

We painted our example with broad strokes and for a single term; but real-life ambiguities play out across thousands of search terms. The need for statistically relevant rules to govern your efficiency-based program is clear, and the effort will pay off.

Each Element of a Search Ad is Ripe For Testing

Savvy direct marketers understand that ongoing testing is the key to improving results.

Done right, paid search advertising lends itself well to this rigor: Each element of a paid search ad- keyword, copy, and destination page- can be the basis of a profitable test regimen.

You’ll notice that many of the same best practices from offline testing apply online as well.

Test Shouts, Not Whispers. Test One Thing at a Time.

Testing takes effort, attention, and sometimes money. Don’t test tiny tweaks. Favor bold tests that have the potential to really change your marketing. Subtle tests will, at best, yield subtle results, often too small to detect.

Multivariate testing (aka scientific tests, design of experiments, or Taguchi testing) offers marketers the chance to vary many factors at once in a statistically valid way. We strongly recommend a walk-before-you-run approach. Start by changing one element at time, determining if that elements helps or hurts in isolation. After you have a strong testing program in place, you can later bring in a consultant or software to help with more advanced testing methods.

Make Sure Every Ad Has a Unique Tracking Code. Make Sure Your Tests Are Statistically Valid.

Each code should describe a distinct combination of phrase, search engine, destination url, and ad copy. Without accurate tracking, you can’t distinguish losers from winners. Detailed tracking also lets you “slice and dice” your results after the fact along different dimensions, gaining more insight from the same testing effort.

Make sure you can separate marketing signal from marketplace noise. Familiarize yourself with basic statistical significance calculations. As a very rough rule of thumb, if you plot conversion rates over time, a test needs to increase conversion by more than 1.5 times the normal range of variability to be significant.

Keep a Testing “Notebook

For each test, document what was tested, why, and what happened. Just a few brief sentences can prevent you from repeating mistakes or wasting time on questions that have already been answered.

Before the test, write down a clear hypothesis of what you’re trying to prove or disprove (”Test 6: Our hypothesis is that bringing visitors coming into our site from paid search to the new simplified product page template will increase conversions relative to the current grid-style product page template”).

Write down your decision metrics, and the roll-out plans. (”If the new pages increase closing by a significant amount, e.g. 50+ more orders than the control for the week, we’ll discard the grid template in favor of the simpler template.”)

After the test, record what actually happened. (”Despite one very large order, the simple treatment did not work, actually reducing conversion, but by a statistically insignificant amount. Next steps: keep the grid, test another challenger later this month.”)

Accept That Many Tests will be Duds

Be patient and be prepared for the fact that most tests produce a null result. And the more successful your program becomes, the harder it is to move the needle. Because your current marketing approach represents years or decades of thoughtful improvement, many alternatives won’t test out better.

Should this realism about the difficulty of hitting home runs dissuade you from testing?

Not at all. When you do test your way into an improved marketing strategy – be that better copy, better destination URL choice, or better landing page design – you’ve struck gold, more conversions for the same cost. While success may not come easily, the potential rewards are well worth it.

Get into a Testing Groove

As much as direct marketers recognize the value of testing, many admit that they test far less frequently than they think they should. Work with your team to establish a rhythm to your testing. Instead of one-offs driven by the latest buzz, establish a plan for a series of tests and stick to it.

Introduction
Computers are great at simple repetitive tasks requiring extreme precision.

Much of paid search marketing fits this description. If you’ve managed paid search campaigns by hand for more than a few thousand terms, you know what we mean.

Updating thousands of bids daily, testing copy across thousands of ads, keeping myriad of tracking codes straight—a human quickly goes crazy trying to keep it all straight.

Bid Robots
Enter the bid robots.

These helpful pieces of software automate much of the intricate drudgery of tracking, managing, and optimizing bids for paid search campaigns. An intelligent and reliable bid robot is a key weapon in a paid search marketer’s arsenal.

But as useful as a smart bidding robot can be, even more important to success in paid search marketing is making sure a smart person supervises that ‘bot, particularly during key selling seasons.

The danger of driving with the rear-view mirror
Bid robots operate on the basic premise that tomorrow will look like today—that is, the recent past and present are reliable predictors of the near future. And this premise is usually right. Want to predict tomorrow’s weather? Take a look outside today. Crazy as it seems, predicting tomorrow’s weather by stating what happened today is on average more accurate than the National Weather Service.

The problem is that predicting the future based on the past breaks down as soon you’re not in steady state. And a direct marketer’s calendar is full of exceptional events, only some of which can be anticipated. There are the winter holidays, your peak selling season, changes in the competitive landscape, and new developments in the search engine’s algorithms, to name but a few.

A machine may know that bids on blue-striped widgets are best set by looking at the last few weeks’ worth of conversion data, but it doesn’t know that you just mailed a catalog. Or that Madonna was just spotted in Paris wearing blue stripes. Or that your top competitor just went broke. That’s why you need a smart human collaborating with your robot, watching it and twisting the dials to keep it on track.

What the robot knows and doesn’t know
Consider Q4, the peak selling season for most catalogers and retailers. As you enter the season in October, business can be pretty slow, and your robot could place bids accordingly.

In late November, as consumer spending ramps up, your robot follows suit. December week 1 is strong, week 2 stronger, week 3, stronger still.

Your robot is tracking bids to results every step the way. In fact, based on recent history, the robot might stomp the accelerator to the floor during December week 4.

What your robot doesn’t know is that, given your merchandise and shipping schedule, suddenly it’s too late for your shoppers to buy from you and get it in time to place under the tree for Christmas. Headed toward a brick wall, your ‘bot needs a smart human to step on the brakes, as recent data don’t contain any hints about the end of the season.

During holiday 2004, we heard unpleasant stories of retailers who left their robots unchecked, leading to costly and unprofitable bidding overshot the holiday. These retailers privately shared their dismay that their bidding systems kept spending in early January like it was still early December. Ouch.

Slow reaction times of bid robots
On the other side, robots can also be slow to react to the beginning of a holiday. The migration of shoppers to the web has made quick reflexes even more important. Over the past decade, we’ve seen holiday selling seasons become increasingly compressed. Online shopping and consumer willingness to wait for ever-stronger discounts make for a late start, rapid acceleration, and a sharp peak.

Online, when holidays do come, they come in like a freight train. Again, a smart human supervising a smart robot can help you get the jump on the holidays.

Robots don’t know what your customer is thinking; you do
Unlike a machine flying solo, a seasoned marketer also understands that conversion is always a function of the kind of traffic coming to your site, not just the quantity.

Consider the person shopping for Christmas ornaments on December 15. She’s likely busy and rushed. She waited to the last minute. If she’s paying for expedited shipping to get the goods in time, she’s more likely to be affluent.

Compare this person to the shopper looking for same ornaments on January 15. With more time than money, this second shopper is in the parking lot before the store opens, price-sensitive and looking for a deal.

Robots cannot see beyond their data
At its heart, a robot is just a collection of rules and algorithms. It can’t see beyond its data.

When you talk to your search provider or in-house expert, you should be able to ask questions about the basis of their robot’s approach.

While you can’t expect an agency to reveal the code behind their technology, you should expect specific concrete answers. Your agency’s explanation of the bidding approach being used to spend your precious advertising dollars should make sense to you, and should gibe with your understanding of direct marketing. A bid robot may be proprietary, but should not entirely be a black box.

Use robots and humans
The bottom line?

Understand that bidding requires both smart humans and smart machines. And dig in a little, asking your agency or technology partner about the approach their faithful bid robot uses when buying clicks for your firm.