Rocket Ship Marketing

Space ShuttleI think the correct way to do marketing, or the way the current landscape demands, can be explained using a metaphor having to do with comparing advertising to a rocket ship:

Whats Wrong With Big Rocket Ships (AKA Traditional Advertising):
1. Traditional rocket ships are big, complex and are explicitly designed to be controlled. These design requirements result in a huge increase in their cost and complexity, and decrease dramatically the probability of the success for the mission. Likewise, a traditional marketing plan is made to be as big and encompassing as possible. Since there is so much riding on it, the marketing must be controlled. With rocket ships and traditional marketing plans, it’s make it or break it.

2. A Rocket ship uses 80% of its fuel in lift off. Once it gets past the lift off stage its pretty much on its way. Likewise, it takes 80% of the marketing budget to launch a new campaign before it has a chance at catching on. What if the idea is a dud and you can’t tell until  after spending 80% of the budget?

3.With the rocket ship costing so much money and taking so much time, it needs to be successful. Traditionally we are continually looking for a “hit” in advertising, something that will spread and become part of pop culture. While trying to predict the likely success of a chosen brand message being a hit, inordinate amounts of money and energy are spent, often, all in vain. Who knows what will be a “hit”? Nobody.

The Solution To The Traditional Rocket Ship Is Replacing It With Hundreds Or More Mini-Rockets:
1. Mini-rockets are cheap to make and launch. And since you don’t have to rely on only one to succeed dramatically well, they don’t need to be built in with fail-proof security and reliability (the reason why traditional rockets cost so much and take so long to build.)

2. Continually measuring the mini-rockets as they go, you can build on the ones that are working and cut back on the ones that aren’t in real-time. Having a budget split up this way allows you to not waste more money than is needed before cutting back on ideas that aren’t working. By relinquishing control, the mini-rockets would be on their own, only bothering to send back whatever they discover.

3. No one knows what is going to be a hit. It’s also hard to tell which demographic will best respond to any particular media. But you stand a better chance at finding a hit with a hundred estimated guesses at a dozen different demographics than one big idea. Then, after launching the multiple mini rockets, you  can let the data tell you which one is the best and continue to fund that one.

This strategy requires the brand to forgo the single-minded brand proposition and embrace long tail thinking. Every brand has more than one potentially ideal consumer. But the big rocket ship’s only option is to target one demographic broadly to capture as many people in that demo as possible. This creates a bland campaign that doesn’t resonate perfectly with anyone since it’s trying to attract as many people as possible by being broad. Targeting the “edges” of the tail instead of the masses, or “head” of the tail, requires narrowing your list down to those most likely, most interested people and ignore everyone else.With mini-rockets you can target very specific demographics with a very specific message. Sure, the amount of people you are reaching is less but your chance for success is higher. This is also much cheaper. Target a dozen different demos and try sending a rocket to them all. These smaller niches, in aggregate, can be composed of as many people as the “head” of the tail and stand a better chance of accepting and spreading your brand.

The Big Idea Is Dead

The “Big Idea” in marketing made sense years ago when the internet wasn’t here. It was wise advice when it came to writing a newspaper ad for example. If an ad said, “we have the best location, best service, best prices and best products,” chances are the ad isn’t going to be effective because the ad is throwing too big of a net and no one is going to believe it – no one is that good. So brands would create a “big idea” to help customers understand and remember as fast as possible for example, “our brand is the cheapest.” Thats big, it resonates with the those who want the cheapest and it has the ability to chisel a spot into the audience’s mind. Simple. Big. Believable (maybe).

Since the newspaper has such big reach, hundreds  of thousands of people would see the ad but only 5%, or less of them would really care deeply about finding the cheapest brand. So a brand effectively wastes money advertising to 95% of the people who don’t use cheapest as their main filter for making buying decisions but instead care about best quality, reliability, customer service, no-hassle buying, convenience, brand image, and a whole host of who knows what. But that’s OK because usually that 5% or less was enough to make it worth it.

So back then you advertised to as many people as possible, pitching one idea that doesn’t apply to everyone but still bland enough to cast the biggest net possible, hoping that your message does reach the people that you are intending it for. Today people are much better at avoiding ads they don’t want to see and are less of a mass and more a mass of niches.

Had the ability to target and measure the effectiveness of your media in real time existed back then, I doubt the “big idea” would  have ever been created. Back then you couldn’t accurately or affordabley target males 18-21 with HHI of $80,000 +, who like japanimation and Nine Inch Nails. That audience would have been too hard and expensive to find and wouldn’t have been big enough give a return on investment to justify the effort.

Not anymore.

Want to target stay at home moms, age 30 – 45, HHI $100,000, who are green conscience and like scrapbooking? OK. Why not choose the 10 most ideal micro-demographics and tailor your media just for them? They will find it super relevant and have a much better chance of connecting with the brand and spreading it to others.

Earning Attention In Marketing

I think there is a lot to be said about having a marketing strategy to create content/earn attention as opposed to buying media/buying attention. I’ve gotten used to most advertisements being  avoidable if I don’t want to pay attention to them, so when there are ads that I can’t avoid, like video pre-rolls or in shows on Hulu, I find that I have half the amount of patience with them and dislike the companies more as a result. Don’t try to force me to watch something I don’t want, which to me, does more harm than good. This is a good video below explaining the idea. From the video: “how can I help you accomplish the task you are looking  to accomplish versus interrupting you and distracting you from that task?”

PPC Strategy: Match Type vs. Negative Keywords

When making a pay-per-click campaign, the goal is for your ads to show up only for the people that they are most relevant for. Hence, match type and negative keywords as a way of doing just that. But what’s the right combination of match types and negative keywords? Here’s my theory:

If you sell ballerina shoes, for example, bidding on the broad term ballerina shoes will result in your ad showing for the most random search querys such as:  “name of ballerina wearing red shoes in that one movie.”

So you may decide to you phrase match, as in, “ballerina shoes.” Impressions will go down, cost-per-click will go down, but your click throughs should go up as you’re reaching a more targeted audience. Run another Search Query Report in AdWords and you will see traffic coming from querys like: ballerian shoes pictures.

Obviously you don’t want your ad appearing for people looking for pictures. So you can choose Exact Type, as in, [ballerina shoes]. This way you’re only getting people searching for those two words in that order. Impressions and CPC will go further down but click thoughs should still go up (relatively) and conversions should also go up since you’re getting an even more targeted audience.

Now take a look at this (PDF):

number-of-keywords-report

Searches with 5+ words have increased an average of 10% year over year. People are using more words in their querys which  means your exact match bid on [ballerina shoes] will be missing a lot of potential customers since using only two words in a search query is down 5% year over year.

New strategy: Keep your exact match bid going on [balleria shoes] since the your average cpc on those keywords will be lower. Then keep the broad match keywords but load up on the negative keywords. Run Search Query Reports and look for all the words that don’t have to do with your product and add them to your campaign as negative words: -images, -free, -download, etc.

This way you can keep out the unrelated random searchers but keep attracting the people getting very specific with their search querys.

Improving Grocery Stores With Data

groceryThe app Shopsavvy for the Android and Redlaser for the iPhone are pretty sweet. With the iPhone 3GS you can use its video camera to capture a barcode, then the app shows you price comparisons for that product online as well as comparing the price to other stores near you.

This makes me think of all the cool things grocery stores could do if they harnessed the internet’s data collection and social networking abilities. A few ideas:

1. Grocery stores already keep track of purchases with their loyalty cards when you check out, why not share that information with the consumer? Give the customer a personal page on their website that shows their shopping habits and make recommendations like Amazon does – 90% of people who purchase Cool Ranch Doritos also buy Cherry Coke.

2. Being able to see everything that you have purchased and the quantity of what you have purchased would help you plan your shopping better. Like what Mint does for personal finance, you would know more accurately how often you need to buy milk. You could see pie charts that show you how much of your food purchases are made up of candy.

3. With more data, grocery stores could give highly relevant and targeted coupons designed individually for the consumer. With enough time the grocery store will know which kind of offers – buy one get one or % off – and on which products incentivize customers to buy. They could figure out that my cookie of choice is Oreos and any discount below 20% off won’t make me buy, but as soon as an offer comes for 30% off Oreos, I’m there. The store could effectively maximize every purchasers buying ability.

9. Brands could set up loyalty products for each of their items. Your 10th Kraft purchase gives you 10% off your next purchase of cheese.

7. Or how about instead of going through a checkout line you put your cart through a conveyor track, like an x-ray machine at the airport, that scans all your items immediately and gives you the price.  No more paying price checkers and no more lines.

8. Or, what if your fridge had bar scanners on the side of it so that it knew everything that you had in the fridge. It could tell how often you take things out so it would alert you if some food was about to expire soon. If you needed to go grocery shopping, just push a button on the fridge and have it print out everything you have run out of, or better yet, send it to your handset.

5. What if every time you put an item in your cart, a digital read-out of the total price of everything in your cart was displayed on the cart handle; take something out of the cart and the price goes down. This would be awesome for customers to be sure they weren’t spending too much while picking out the groceries. Of course, grocery stores probably like that we don’t know how much everything costs until we get to the check out line. But think off all the cool stuff you could do:

6. Supercook is an online tool asks what’s in your kitchen and then uses that information to provide dozens, if not hundreds, of unique dishes that you probably would never have thought of on your own. What if the grocery store kept track of what you put in your cart and gave you ideas of what dishes you could make while in the grocery store  – add Country Crock butter and Pace Picante Salsa, and you have all the ingredients you need to make zesty enchiladas.

4. Looking at the data you could tell in what order most purchases happen in. So they find out that statistically, after people buy meat, the next thing they buy is beer. Up-sell by placing selected items next to, or in-between the meat and the beer for that purchaser to see like barbecue sauces or beer coasters.

10. Let customers connect with other customers who buy similar things. Looks like you buy a lot of spices and ingredients, would you like to join a recipe community in your area? You BBQ a lot, compete in your local community cook-off.

Maybe some of this stuff seems a little too much big brother, but I would let companies know my purchasing habits in exchange for relevant coupons, food suggestions and insights to my food buying habits.

Links From Twitter = Better Traffic

800px-moscow_traffic_congestion2Saw this on TechCrunch this week:

Twitter [and Facebook] “will surpass Google [as a source of traffic] for many websites in the next year.” And just as nearly every site on the Web has become addicted to Google juice, they will increasingly try to find ways to get more links from Twitter. Because Twitter equals traffic…these Twitter links “convert better” than search links because they are often pre-filtered and come in the form of a recommendation from someone you are following.

This is a good point. Now that there are no more gate keepers to filter what gets published, everything gets published and we rely on filters (bloggers, friends online) afterwords to help us find the best stuff. Twitter is a great tool for networking with like-minded people to help you find the stuff you’re interested in. I rely heavily on the opinions of bloggers that I trust to point me to where the good stuff is at. This is the best reason, so far, for getting a business involved in Twitter.

P.S. Google has added a Creative Commons filtering capability to its Image Search results to allow you to find images without having to pay a stock photo company or steal an unlicensed picture from Image Search. The option isn’t currently available on the user interface, but you can enable the filter by adding a certain parameter to your search URL (in this case, just replace “mountains” with whatever you are actually searching for)

  • For public domain images — http://images.google.com/images?q=mountains&as_rights=cc_publicdomain
  • For images licensed with Creative Commons Attribution (that is, images you can use as long as you attribute the image back to the creator) — http://images.google.com/images?q=mountains&as_rights=cc_attribute
  • For images licensed using Creative Commons Attribution Share-Alike — http://images.google.com/images?q=mountains&as_rights=cc_sharealike

I got the image above from Wikimedia Commons. Nice!

The Former Audience Are Now Particiapants

Great talk at TED by Clay Shirky:

In a world where media is global, social, ubiquitous and cheap – in a world where the former audience are increasingly full participants – in that world, media is less and less about crafting a single message to be consumed by individuals, and is more often a way of creating an environment for convening and supporting groups.

What an exciting world we live in. The ability to create movements, be heard, and make things happen has never been easier. Now it’s about asking the question, what do you want to create?

Business Websites Will Change

I think pre-internet, most people trusted big companies more than little ones because they were bigger. If you have never heard of a company then how can you trust it? Now that any company is findable online the tables have turned. Big companies that have static brochure websites, where they have their corporate jargon and stock photos on every page leads one to think, what are they hiding? Meanwhile little companies whose websites consist of blog posts and insights into who is behind the company and what they think leads one to believe, these people are just cool normal people who do what they do well. Cool.

I want By Data Be Driven to be my personal Internet marketing consulting brand but I choose to make my site a blog that shows who I really am. I think that goes father than a lot of pages about services offered and my unique strategies of how to “effectively market your brand online using integrated search marketing to help companies get noticed, retain customers and continue growing.” If you really want to know what I can do you’ll find out from someone else other than me, so why put up the front and just be as personal as I can?

I think before long, most sites will have a blog, feeds from their interaction with others and interactive widgets/applications on their home page instead of a stock photo of a happy customer, their tag line and links to about us, services and contact. Just wait.

Super Crunchers

I finished reading the book Super Crunchers by Ian Ayres. I thought it was good. I liked his explanation of randomized a/b or multi variant testing done online and off.

He explains that randomly dividing prospects into two groups and seeing which approach has the highest rate is one of the most powerful super crunching techniques ever devised.

When you rely on historical data, it is much harder to tease out causation. The sample size is key. If we get a large enough sample, we can be pretty sure that the group coming up heads will be statistically identical to the group coming up tails. If we then intervene to treat the heads differently, we can measure the pure effect of the intervention…after randomization makes the two groups identical on every other dimension, we can be confident that any change in the two groups outcome was caused by their different treatment.

Of course, randomization doesn’t mean that those who were treated differently are exactly the same as those who were not treated differently. If we looked at the heights of people in one group, we would see a bell curve of heights.  The point is that we would see the same bell curve of heights for those for those in the other group. Since the distribution of both groups becomes increasingly identical as the sample size increases, then we can attribute any differences in the average group response to the difference in treatment.

In lab experiments, researches create data by carefully controlling for everything to create matched pairs that are identical except for the thing being tested. Outside of the lab, it’s sometimes simply impossible to create pairs that are the same on all peripheral dimensions. Randomization is how businesses can create data without creating perfectly matched distributions.

The power behind randomized testing is undeniable. So should we just have computers make all our decisions for us? With that question in mind is were he goes throughout the majority of the book.

Randomized trials require firms to hypothesize in advance before the test starts. Historical data lets the researcher sit back and decide what to test after the fact. Randomizers need to take more initiative than people who run after the fact regressions.

The most important thing that is left to humans is to use our minds and our intuition to guess at what veriables should and should not be included in the statistical analyisis. The regressions can test whether there is a casual effect and estimate the size of the causal impact, but somebody (some body, some human) needs to specify the test itself.

So then the question becomes what do we test, and after we test the question becomes, what are the results telling us?

Social Media Marketing comes down to 2 things

I finished reading the book World Wide Rave by David Meerman Scott. I thought it was OK. I’ll explain the problem with these kinds of books at the end.

While I was reading it I came to the conclusion that success at social media marketing and having your business idea spread virally comes down to two main factors: creativity and understanding the tools that the internet provides. You may have a good understanding of how to use all the internet tools: blogging, Facebook, Twitter, YouTube, Squidoo, ebooks, monitoring your brand using Tweet Scan, Google Alerts, Boardtracker, Social Mention, trying to build competitions, interactive tools, applications, widgets; but if you don’t have something creative that’s worth spreading it won’t spread. The creative content part is harder than the tools part.

One way to start getting creative is to put your message in terms of the need that your business solves. Scott says:

By truly understanding the market problems that your products and services solve for your buyer personas, you transform your marketing from mere product-specific, egocentric gobbledygook that only you understand and care about into valuable information people are eager to consume and that they use to make the choice to do business with your organization.

Once you’ve gotten down to the needs that your product fulfills, then you can start trying to come up with creative ideas around communicating your solution to that need. After that there is not much advice one can give on how to be creative. I guess you could try brainstorming ideas. Another good idea in the book is to try lots of things in hopes that at least one of them sticks.

Many attempts will be duds that won’t spark any interest; a few will generate some notice and basically pay back your investment of the time required to make them; and a handful will spread to thousands or even millions of people and make the entire program of 10 or 20 initiatives worthwhile.

Realistically I think you’re more likely to make a hit if you try 50 to 100 initiatives.

So there’s the problem with most of these internet marketing books; they can’t explain to you how to creatively use the tools that the internet provides to spread your message. They can only show you the successful ideas other people have had. That’s the hardest part and only you can figure it out.