You can't increase ROI without some experimentation. CrownPeak Technology's VP of marketing and product strategy describes the process for turning failures into success.

Consider the following question recently posed by a vice president of marketing at a web technology conference: "I have tried to figure out which online marketing tactics will give me the best results. I've looked into email, Search Engine Marketing, Search Engine Optimization, traditional banners and contextual advertising. Now, I don't know where to put my money, because I might fail and management expects me to keep improving ROI."

The answer for this particular person may seem obvious. But, at the heart of the question was something many of us who cut our teeth on the reach-and-frequency school of marketing face on a daily basis: With the continuing challenge of delivering improvement to our marketing investment, how do we know where to place our bets so we don't screw it up?

So, whether you're a grizzled veteran of a failed set of email marketing campaigns, or struggling to find one more tactic that works as well as your online banner ads, there is a set of best practices I've experienced that seem to work well.

Let's first look at where we started
I'll be the first to admit that I came out of the reach-and-frequency school of thought. (And, yes, at some point I did have "Flock of Seagulls hair.") Using that model as a marketer, you built an expensive soapbox, you climbed up on top of it and you yelled your message as loud as you could. Then, you built another expensive soapbox down the street, and ran down to it and screamed it again, hoping the same person would be walking by.

In his classic book, "Permission Marketing," Seth Godin draws the metaphor that this model is like the marketer walking into a bar with a wedding ring. He asks each female patron if she'd like to get married. Of course he's destined to fail. And, when he fails, he blames the suit maker, the ring maker, the bar; everybody except himself. Godin suggests instead, why not try dating?

So, the key here is that it is expensive and risky to build soapboxes. And we've come to expect a "low conversion percentage" as just a marketing rule of law.

What's this thing called "paper click"?
Then along comes online marketing, and the rules change. Unfortunately, for many, while the tactics have changed, our outlook in how to manage them hasn't. Many online marketers still bring the reach-and-frequency attitude to the online marketing toolbox. 

While we use email marketing, we're still not tracking what the recipients do on our websites, and using that to refine personalized messaging. While we support Search Engine Marketing, we're not continually testing the messaging, landing pages and offers to make it efficient. And, most importantly, while we embrace online marketing, we're not experimenting with different types of tactics, funding the ones that work and killing the ones that don't. 

In short, we're still bringing a Marketing 1.0 philosophy to a Marketing 2.0 world.

So, what now?
If you're wrestling with this whole notion of how to build an online marketing engine, consider a few things from a philosophical shift:

Get permission to fail: Work with your management team to allocate your marketing budget and success based on continuing improvement over a longer term, and with the understanding there will be a learning curve and experimentation which may or may not bear fruit.

Hedge your bets, but make some: Definitely keep your successful core tactics, but allocate a certain percentage of your budget to experiments. With most online tactics (especially performance-based PPC or PPA tactics), you can run experiments based on budgets.

Use a stair-stepping methodology: First, focus on establishing your base, from which you can measure anything you do. Make sure your website and landing page infrastructure is solid, and that your analytics and lead measurement is in place. Then, focus on the experiments: the outreach. Whether that's SEM, SEO, email or even offline events, make sure everything comes back through your website and can be measured. Finally, tie all that back into your top five metrics (whatever they may be). Measure like crazy, and then fund the experiments that are working and cut the ones that aren't. Be patient. Give yourself at least a quarter or two for any tactic.

Stop thinking in "campaigns": If the word campaign appears in your online marketing strategy, rethink it. Online marketing isn't like old warfare anymore; it's like special operations. You will certainly have broad-based creative that you apply to your strategy (just like special operations is driving toward an overall goal). But using smaller, strategic tactics is less risky and develops success that grows into permanent investment.

Next: Helpful safety tips