Best practices, optimization, or proven steps – which one would you choose?

This week we had a good reminder on the importance of testing your message. Working on a client’s campaign, we tested three different messages promoting a new whitepaper. We tested the messages as a lead nurturing campaign to the client’s house list, with the intention of using the best performing version for a lead acquisition campaign to a rental list later on.

The three messages were identical in format and most of the text. The only differences were in how we phrased the topic of the whitepaper, which appeared in the message subject lines, the title, and in one line in the e-mail body. The three variations were:

- Ten ways to optimize
- Ten best practices
- Ten proven steps

The results were rather striking. Open rates for all three messages were rather similar despite the differences in the subject line (24.0-24.8%). However, click through rates were significantly telling:

- Ten ways to optimize – 21.9%
- Ten best practices – 16.0%
- Ten proven steps – 32.1%

Given the significant differences in click throughs, I was a bit surprised to see the similar open rates for the three different subject lines. At the same time, it has been a trend I have been seeing developing for awhile. My take is that sender name has become much more important than the subject line (we used the name of a salesperson + the company name as the sender name, with the salesperson’s e-mail address). If the sender is someone I know and trust, I would probably take a look at the e-mail.

So what did we learn from this experiment?

1. Obviously, we know which message we will use for the rental list.
   
2. The parity of the open rates does not mean that we should stop paying attention to the selection of subject lines. I am pretty sure we could have come up with some bad subject lines that would have performed significantly worse even with the same sender name.
   
3. With sender name having a major effect on open rates, it is more crucial than ever that each e-mail we send delivers value to our audience. Once they stop reading our e-mails, even a great subject line is unlikely to reverse the trend.
   
4. Above all: we are going to continue testing as much as we can. 100% more clicks is not something we can pass on!

Please send me a note or post a comment if you have any interesting test results you can share.

Marketing by the Numbers

"Conviction is the luxury of those sitting on the sidelines." This is what John Nash's imaginary boss said in the movie "A Beautiful Mind." For those of us who choose to play rather than watch, Measurement is the compass that guides us through our mistakes.

What is your most important marketing tool? I would argue that your spreadsheet application should be high at the top of the list. In this day and age, running your marketing department without constant attention to the numbers is simply irresponsible.

The starting point is clearly defining the goals. What is the market impact goal of the specific activity? Is it the total number of leads? Number of qualified leads? Closed deals? Is it brand awareness? All these goals have to be expressed in quantified, measurable terms, or Measurable Market Impact.

Ignoring the numbers is not just irresponsible, it is also inexcusable. The great thing about electronic marketing using e-mail and the web is that almost everything can be tracked. Making smart use of these vehicles means that you design your marketing campaign results to be measurable, using trackable links, source-specific landing pages, and tracking codes. These are relatively simple to implement, so we'll assume that you've got this part right (let us know if you need any help).

Still, there are plenty of examples where marketing departments don't manage their activities by the numbers, for a variety of reasons. Let's examine some of these more common reasons, and see how you can work around them.

Example #1: "our goal is to create brand awareness, but measuring it is too expensive."

There is hard truth in this statement. Conducting market surveys to measure the impact of specific marketing activities on brand awareness can be cost- prohibitive for most small companies. What you can do is estimate the number of people exposed to your marketing message by each marketing activity. It is not the best or most accurate measurement, but it's still better than not measuring at all. Defining the results in such measurable terms allows you to compare different branding vehicles, such as print advertising and online newsletter sponsorship. Both can generate brand awareness, but which one is more effective?

Example #2: "we don't expect too many leads from this tradeshow, but we have to be there; otherwise, people will think we are in trouble."

Maintaining your company's image is a legitimate and important market impact goal, yet there are many ways to achieve it. This goal can and should be defined in measurable terms, such as the number of people exposed to your presence. In this case, the cost of the tradeshow should be allocated against two market impact goals - lead generation and company image.

Example #3: "we don't know how many people will respond to our e-mail campaign."

It is easy to estimate the results when you have a history of similar activities to rely on. Still, not having such history is not an excuse not to estimate the results or clearly define the goals. Even a wild guess is better than not having one at all; it may help you see that in some cases even your wildest dreams cannot justify the expense. If you have a large list, test a smaller sample of it before you get started (make sure the sample is randomly selected and statistically significant). Once you start generating results, it will be easy for you to go back and adjust your estimates and goals moving forward.

Example #4: "we cannot measure the results of our analyst relationships program."

You bet you can. If you're only looking for advice, the budget should not come from your market impact programs. Market impact goals of an AR program are certainly measurable, such as the number of positive mentions in the analyst reports, media quotes, and customer referrals.

Example #4a: "but we don't necessarily know when a customer was referred by an analyst."

First of all, you should know. A sales person should know how the lead was generated and who influenced the decision. But even if you don't know it in all cases, do you know it in most of them? Half of them? Put a factor on the results based on your estimate of how many you miss. Even if you're wrong on the factor, you should see a trend of improvement when you launch or increase your spending on analyst relationships. If you don't see improvement - it's a good sign to go and check why.

To summarize:

1. Define clear Measurable Market Impact goals.
2. Estimate the expected and desired market impact of each activity (some activities may have more than one type of market impact).
3. Prioritize your budget based on estimated results and Cost per Market Impact.
4. Diligently measure whatever you can. Direct all traffic to your website and use separate landing pages and tracking codes for each activity and source of market impact.
5. Estimate whatever you cannot measure directly; imperfect measurements are still better than no measurement at all.

As former NYC mayor Rudy Giuliani said when asked how he reduced crime in the city, in reference to the extensive use of metrics and benchmarks exercised by his administration:"If you can't measure it, you can't manage it."