Josh Kerbel

The prerequisite of making money online is getting visitors to your site.  This means having promotional system in place that taps into the problems that people are trying to solve.  Take a look at some of these banners and PPC ads to see if and how they work.  Here is a sampling from around the industry.  It’s far from complete, but gives you an idea.  The point is to think of who this ad is targeting.  What you are seeing are ads that have been tested with lots of money and lots of page view, so even though these ads may seem explicit, irrelevant or offensive to you, they work.   These ads could be tested with upwards of $1,ooo,ooo worth of promotion.   Every word that is on the page is there for a reason.  Stop blushing and think about how this applies to how you can promote your auto supply business,  your  flower shop or your IT consulting operation that you are trying to get off the ground.

Have you found out the emotional hot buttons that move you clients?  Where there pain (or pleasure) is?   The first ad taps into someone looking for an escape from the grind of reality.   The second ad goes after people’s lack of knowledge of sexual activities and confronts the

awkwardness of discussing the subject of a lack of sexual gratification with your partner.  Talk about going after a PAIN POINT

The third screen shot shows some PPC ads , that, well, could honestly be better written, but some do attempt to address whole issue of discretion and online dating.

Click on the image to read the PPC ads.

Like I said, these ads, and the landing pages they send you are the product of a huge amount of time and testing.  Unlike many of the ads you see for doctors, lawyers, flower shops, concrete cutting,  a surprising amount of brain  and computer power went into them.

Next up, we are going to pick a site for Robert to sign up at and see the method it uses to extract money from his wallet.

If you find this topic distasteful, think about it this way.  You are getting millions of dollars of education for free.  Once we take a look at the conversion process, we will then shift our focus to the back end and marketing formula

The bane of anyone who has ever thought about advertising a product to promote its awareness is that you commonly run in to silly ideas like “Oh, we can’t measure advertising, but know know it works” or ” This is valuable for your image” or “Give me your money and don’t worry about a thing”.  Brand/Image advertising has traditionally been held unaccountable for its performance,  you simply do it because you are supposed to do.  It’s what the cool kids do.

Well guess what, it’s all a load of crap.

The problem isn’t with the advertising itself, its the fact that no one bothered to come up with an operational definition for the business objective of “promoting product awareness” in a way that can be measured.  Let’s get a little creative and see how you could operationalize increased product awareness.

1. New Visitors to Your Website

2. Repeated Visits to your website – do people like what you are saying enough to come back

3. Increase in sales/converstion rate/average revenue

4. Increase in conversions in non-sales activities – mailing lists, coupon downloads, white paper downloads

5. Search Activity – was there an increase in the volume of search activities related to your firm

This is proof that you can add the accountability of direct marketing to the traditional “you just can’t measure” brand advertising

Marketing is way more systematic than creative.  So rather than a primer on Photoshop, I though a primer on testing and system development would be more appropriate

Start with a hypothesis

Beyond saying you need to try different things, you need to test specific things. You need to have a hypothesis that you can test and measure (which means you need a clear success measurement for the goal you are trying to achieve) All of this means that you need to ………

Create goals

You just can’t measure for the sake of measuring, you need figure out what you are measuring and what you are measuring it against………you need a scale………….which means you need……

Baseline Data

You have to accept that you need to get something up and running in order to test against it. You need to make sure that you have enough baseline data to make statistically valid decisions when you start testing.

Now repeat………

You need to go through the above process for every step in your marketing system.

If you been around sales for any period of time, you know it is all about numbers.

Unfortunately, it’s usually about the wrong numbers.

Plenty of sales efforts are still based on cold calls. Or in some cases, unsolicited email.

Doubt me, take a look at Craigslist or any job site and see how many sales jobs emphasize the need to make cold calls

The basic idea being that if you interrupt enough people in the middle of the day (or even worse, at home in the evening) a certain percentage will agree to speak with you and certain percentage of those might actually become qualified prospects and a some small percentage of those might actually buy something from you. (An off color joke comes to mind here, put I will refrain…….for now)

Often sales managers say that sales people need to contact 100 people to make a sale. A 1% close rate. There you go. For every 100 people you contact you will get a sale. Seem pretty straight forward right?

Well, what about all those people that you didn’t manage to contact?

Either they hung up on you, didn’t answer, the gate keeper shut you down………………..how many of those might have been qualified prospects.

But you’ll never know because you don’t have a system for screening them.

At the end of the day, the number that really matters is how many “qualified prospects” did you speak to.

That’s it………….its still a numbers game. You just need to be counting the right numbers

How do you know what you don’t know or know what you should know.   The only way to do that is to conquer this challenge is to divide up numbers in a way that produces some actionable intelligence.  Or in marketing speak, you need to segment your customers/audience.  Segmentation is what gives context to your stats and alerts you to problems or opportunities, or  better yet, the ability to turn problems into opportunities.

Let’s run through some examples

Visitor – If we just look at visitors, we get everyone who comes to our site.  What is important is who sticks around, so you segment your visitors by duration, say those who say more than 10 seconds and those who stay more than a minute.  Alternatively, you can segment by which search engine referred the visitor.

And then, if you really want to get fancy, you can combine duration and source………..say Google users who say more than 7.5 minutes on your site and buy 4 items

When it comes to direct marketing you are trying to get a grasp on 3 main aspects of your prospects behavior.

1. Acquisition

This lets you know how your marketing budget is being used to generate traffic to your website.  These aren’t even leads yet.  These are just people who have come to your site as a result of any of your on line investments in marketing.  You just want to know how much you spent trying to get them there.

2. Behavior

What are people doing on your site?  Are they doing what you are hoping they would do?  If not, why not?

Better yet, can you develop aspects of your site that serve people with different behavior patterns.  Those who come from blogs, those who view 6 pages or those that take 10 days to purchase

3. Outcomes

This is where the rubber hits the road and you can find out your total investment in attempting to hit your business objectives (the overall reason you were on the web?  remember that?).  Were you looking for purchases?  Maybe leads?

One of the great things about online marketing is that you can build a formula that leads to marketing success.  But first, you need to actually define what you consider success.

Before you put pen to paper or code to page, you need to know what you are hoping to accomplish with your website, so, step one is……….

1. WHAT DO YOU WANT TO ACHIEVE WITH YOUR SITE – Your Objective

-Sales, signups, leads, emails, love letters…….

2. Steps

Now you have to think, OK, I want to sell more of X, what do I have to do.  You need to break down the steps you need to take in order to get to your business objective.

These could be decrease bounce rate on the landing page, increase conversion rates, increase signups from twitter, increase your reach.  You need to sit down and break up your sales process.

3. Strategies

Strategies are the specific techniques you are going to use to achieve each of the steps in point 2.  What will you to decrease bounce rates, increase sign ups, etc……

4. Metrics & Performance Indicators

These are just numbers.  What you need to do is transform them in to something useful – like performance indicators or even a Key Performance Indicators  (if it says key, it must be important).   A key performance indicator (KPI) is a metric that helps you understand how your strategies are performing in achieving the steps needed to reach the business objective.  This could be average order size, cost of lead acquisition, return visits.  Like everything else in life, the simpler the better.

6. Targets

What you aim for.  This is an operational definition of your business objective.   Put it in numbers, make sure these numbers have some basis in reality.
7. Dimensions

A dimension is a characteristic of a site user.  Time on site, referring source, # of clicks on your site, number of return visits, sex, income are all dimensions.  Think of dimensions as demographic information

8. Segments

Segments are how you slice and dice your information based on user dimensions. Checkout the dimensions I am using to segment my website traffic to understand performance better.

• Analyzing people just from a geographic regions;
• People who come from a specific referring site; and
• Specific type of device – laptop, phone, desktop.

You want to look at how you acquire customers – (search, PPC, 3rd party sites) and how they behave in the process of becoming a customer (clicks, time on site, return visits, etc)

Two can be as bad as one
It’s the loneliest number since the number one

No is the saddest experience you’ll ever know
Yes, it’s the saddest experience you’ll ever know
`Cause one is the loneliest number that you’ll ever do
One is the loneliest number, worse than two

- Three Dog Night

Three Dog Night must have been talking about statistical analysis when they wrote this song.

The worst thing you can ever do in starting to analyze the performance of your website is to look at one or two numbers in isolation.  Pageview mean little in isolation.   All a pageview is when a page is loaded in a users browser

You need to see how views of an individual page relate to the overall performance of you site.  Does your most popular page have a high bounce rate?  Is it getting better?  Is it getting worse?

Or how about looking at the number of total pages views on its own.  Say you see a big spike one day.  Was it because there were more users coming to the site, or the same number of users looking at more pages

One of the neat things about the online marketing is you can watch how your customers make decisions in close to real time.  You can see how they navigate your website, well, sort of.

You see it all depends how complex your site structure is.

Say you have a microsite – one distinct target, either a sign up, download, impulse sale and you have designed the site with limited navigation tools, path analysis  comes in handy here.  You can see where visitors lose interest in your site.  But then you have to consider the LONG TAIL

What about sites with 10 and 100′s of pages.  The permutations and combinations of ways a user can navigate your site start to multiply……..FAST and this is where you need to really know how to use your stats.

For any direct marketing, performance comes down to one thing…….numbers.  This job isn’t about creativity.  It’s about finding a formula that generates business.  Think of it like a slot machine that is rigged so that for every 10 dollars you wager, you win 20 or 30.

And just so you don’t get any ideas of easy money, getting to this point requires, well lots of testing and lots of money….well, here we will focus on some of the testing using Google Analytics and other performance tracking statistical packages.

The basics…..

Bedrock

What drives online statistical analysis is, come on, I know you can do it………………VISITORS.  Seems pretty basic right.  Well of course there are a number of ways you can slice a visitor……..figuratively, not literally.  Without going for overkill, lets draw and quarter our visitors.

Total Visitors

All the people that came to your site within a specific period of time.  They may have come once or 10 times

Unique Visitors

The number of individual visitors.  Say you had 300 total visitors in a 2 week period, but 299 of those visits were you checking the site, you had 2 unique visitors

New Visitors

As it sounds, the number of people who have come to your site for the first time

Returning Visitors

People who have come back for a second visit

Where some of these stats become merky is when people start clearing their caches between visits

Going forward, every performance metric we look at will come back to these basics

Enough with the nitty gritty, why does a non-stats nerd care about this

You want to answer 5 basic questions

1. How do visitors use a site;
2. How do I account for my marketing investments;
3. Is my content effective;
4. Are people completing the tasks I want them to complete – Buy/Sign up/Download; and,
5. How can I make it better in the future.