The Ishtar Problem: Iterative Marketing Part 1

May 14th, 2010

Topics: Measurement, Optimization

The Ishtar Problem
Road map to Success

Update: Part 2: A Culture for Change
Part 3: Iterating in the Real World.

In the February issue of Vanity Fair, Peter Biskind examines the ill-fated production of the mid-80’s comedy/model-of-epic-failure Ishtar. In it, the stars and the crew recount watching the film come undone even as they continued to push forward on it.

For many, Ishtar – like Battlefield Earth or The Postman – has come to define the concept of high-budget flop. Like nearly any great failure, hindsight clearly spells out why the film failed: Inexperienced director, bad cast chemistry, vague project goals, and enough time and money to do real damage.

At the same time though, it’s easy to see why the movie got made in the first place: Proven stars, talented writing, expert crew, and enough time and money to do something really great – more than most movies ever get.

Creating marketing programs can look a lot like making a movie and the Ishtar story isn’t unlike a lot of projects I’ve seen over the years. Call it the “Big Idea” process. You get great people together, you come up with the best idea you can using the budget you have, and hope it all works out. Sometimes it does, sometimes it doesn’t. The problem is, like Ishtar, when you’re in the middle of things, good projects can look an awful lot like bad ones. It’s not until the very end, when you’ve run out of time and money, that you have any indication which one you have on your hands.

What would be great is a way to test out ideas, identify the truly great ones early, and put your time, effort and money behind those, while side-stepping the Ishtars. With film that could difficult, but with digital marketing, it’s surprisingly straightforward.

If you’ve spent time either working in, or with, advertising agencies you’re familiar with the concept of the campaign. This is the core product of an agency and it defines the way that ideas are generated, how they are executed, and ultimately what constitutes success and failure. While each agency has their own take on the campaign, for the most part, they look something like this:
Campaign Process

In the “Brief” phase, the agency gets some sort of assignment from their client outlining the needs or goals of the project. The agency then goes into some sort of “Ideation” phase, which includes the brainstorming sessions, the identification of the key idea for the project, and the proposal back to the client. With client approval, the project moves into the “Execution” phase. This where the bulk of the time, effort and money in a project gets consumed. The Execution phase includes all of the design rounds, all of the development, all of the QA; everything that goes into making a project launch. At the end of the project, when all the time and budget is gone, the project is evaluated. It’s this evaluation phase that presents the first problem with this campaign approach.

Because agencies wait until the “end” of a project to evaluate it, when the project is launched to the public and pretty much complete, it’s only real role is to ask “did this project work?” Analysts then descend on the site to see how people are using it, maybe look at conversion funnel metrics, and then make some recommendations about how to tweak things to get better performance. That’s good, but by the time it happens, the major effort is over. The information gathered can’t be used to make actual improvements to the project because the project is over.

If there was a real chance to make a deep impact on the project, it would be earlier.

Compounding this issue is the nature of the campaign culture itself which views each stage of the project, and each project itself, as an isolated event, unaware of the projects that have come before it, not meant to influence what comes next. Once execution has begun, there is no chance to revisit the merits of the original concept, and the launch of a project is considered the end, not the beginning. Working this way, any learnings from one project are unlikely to help improve the next, and there is no opportunity to adjust a project based on its success or changes in the technical or competitive landscape.

When these types of projects succeed, they succeed big (in part because if it’s not big, it’s not counted as a success). But when they fail, they fail equally big, and they fail much more than they succeed. Moreover, their failure was just that: a failure, because it came at the end of the project, when there was nothing to do but move on.

What if, rather than waiting until the end of the project to gather and analyze data, we focused on the middle of the project? What if the launch of a project was just the first step?

At Fight, we’re big fans of big ideas. We believe that great marketing can be transformative, both culturally, and for our clients’ bottom line. So, when we went about creating our company we put a lot of thought into how we could use the kinds of thinking that emerged from the digital space to help make sure that a big idea becomes a bona fide success. Key to this is the concept of iteration.

If you’ve been reading the Webtrends blog for a while, you’ve no doubt seen many references to “Iterative Marketing“. Taking it’s philosophical cues from agile development methodology, iterative marketing seeks to break down a larger strategy into smaller, but conceptually complete elements. By releasing these smaller elements to your audience, you can take advantage of analysis of real-world data early in a project’s lifecycle when the information can still inform project direction. This notion of optimizing the middle of a project, rather than the end, is key to iterative marketing. Building projects this way, not only do you preserve the fundamental drive towards big thinking, it actually enables you to achieve far bigger ideas than traditionally possible.

But more than just increased efficiency, iteration allows you to approach projects in a more web native way. By re-contextualizing your time frame from single, all-or-nothing  campaigns to a more long-term, continuously improving approach to developing projects, you’ll be in a position to develop the kinds of projects that can more effectively deliver actual value to your audience and compete more effectively with digital native projects.

While working in an iterative way represents clear advantages in the digital space, it also presents a fundamentally different way of looking at projects organizationally. In the next installments, I’ll look at what the organization impact is of working in an iterative way and how to actually implement this within your organization.

Update: Part 2: A Culture for Change

Justin SpohnGuest Author Justin Spohn
Justin is a co-founder of Fight, an interactive research and strategy firm in Portland, Oregon focused on making the Internet more valuable for brands by making brands more useful on the Internet. He is one of the foremost thinkers on iterative marketing and digital marketing strategy. Justin recently did a presentation at the Webtrends Engage conference about iterative marketing, you can flip through it on his blog. You can follow him on Twitter @adognamedpants.
Marko Muellner

About the Author

Senior Director of Marketing

With over 14 years in digital, Marko has managed cross-channel, integrated and multi-touch marketing strategy and creative development for MillerCoors Brewing, adidas America, Webtrends,, Razorfish, Magnet Interactive, Digital Evolution and others. As the Sr. Director of Marketing he manages the Webtrends Marketing Studio and leads all marketing strategy, execution, measurement and optimization. Marko is passionate about iterative and insight-driven marketing and about the alchemy of remarkable digital marketing.

6 Responses to “The Ishtar Problem: Iterative Marketing Part 1”

  1. Monte Montoya

    Great Post. I’m so tired of long processes that deliver no value at all in the end.

    As I push forward and try to integrate Marketing / Scrum, It’s a breath of fresh air to hear someone else talking about concepts of iterative and incremental delivery. It works!

    We are long overdue.

  2. Justin Spohn

    Michael and Cory – thanks! I’m glad you guys got something out of it.

    Michael – good point. “Hindsight” I think is exactly the problem. While shorter cycles may be important in certain areas, even more important for me is the idea testing ideas, gathering data, and then using that data to move projects in better directions. The standard approach to waiting the project is done to evaluate it makes that evaluation a lot less effective.

    Cory – Agreed, many executives aren’t up on this. That said, I’d say it’s our job to be the leaders here. Right now, I think a lot of marketing is seen as an expense, but if we work harder to show just how important our work is to the business of our clients, we become a lot harder to ignore.

  3. cory huff

    I enjoy reading your stuff Justin. Personally, I’m always a little bit surprised at some of the very high level executives I meet who don’t have any idea how to track ad performance on the Web, or who just plain do it badly.

    If the majority of executives are still at this stage, how in the world can they ever get to shorter cycle ad campaigns?

  4. Michael Griffin

    Excellent writing! I’m *really* looking forward to the next couple of posts in this series as I feel shorter iteration cycles have huge social implications.
    To wit: the limited effectiveness of post-execution evaluation is a problem across industries. In fact, as the adage “hindsight is 20/20″ illustrates, it’s a challenge present in all human endeavor.


Leave a Reply

  • (will not be published)

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>