How to Use Simple Math to Get Better Insight PPC

I highlighted how you can assess the value of your campaigns average KPIs and how to identify outliers for better campaign insights. I will continue showing you how to improve the value of your campaign with some simple math tips. We will continue to use the example of setting up a new ad group to promote fireplace components.

If you remember from my last post, after we identified that we were going to focus on quality and service as the selling points for this new ad campaign, we developed two different forms of ad copy to reflect each campaigns’ different value proposition. I’ve listed out the results below. Now, what do you think? Should we remove Ad Copy #2 and just keep Ad Copy #1 because its average CTR is higher?

EXAMPLE One: CTR Performance Comparison Between Two Ad Copies



Step One: Use Standard Deviation to Remove Outliner

As illustrated in the Part I of this series, before comparing the averages, we need to separate outlier (i.e. values that fall outside 2 standard deviations from the average) from the rest of the values. As you can see, the CTR on Day 6 of Ad Copy #1 is outside a reasonable range, so we don’t include Day 6 while comparing the performance between Ad Copy #1 and Ad Copy #2.

EXAMPLE Two: Use Standard Deviation to Remove Outlier



Step Two: Use T-Test to Confirm the Difference

After removing the outlier, the updated average for Ad Copy #1 is 4% vs. 5% for Ad Copy #2. We can use a T-Test to confidently declare those two ad copies perform significantly different. (I am aware of any terms with “test” within could be intimidating so allow me to comfort you a bit by revealing that the T-Test was originally developed to test quality of beer. Call it Beer-Test if it soothes you. J) The following are things you need to know about this simple test:

• T-Test is an one-way test, if the outcome doesn’t support our assumption (i.e. two ad copies perform differently) it does NOT suggest the opposite is true (i.e. two ad copies perform the same)
• You can use it to validate the outcomes from Before/After Test on one set of objects
• You can use it to validate the outcomes from A/B Test on two sets of objects
• In general, T-Test value > 0.05 implies your assumption is NOT correct (i.e. the chance of your assumption is not correct > 5% )
• The easy way to perform a T-Test is using Excel T-Test function

EXAMPLE Two: Excel T-Test Function

1. Array 1: the first data set
2. Array 2: the second data set

o Tails

1. To validate the conclusion of one data set performs better than the other
2. To validate the conclusion of one data set performs differently from the other (i.e. could be better or worse)

o Type

1.  For Before/After Test
2. For A/B Test with a same standard deviation
3. For A/B Test with a different standard deviation other (i.e. could be better or worse)

EXAMPLE Three: Use T-Test to Confirm the CTR Performance Difference Between Two Ad Copies



To compare two ad copies with different standard deviations, we should apply 2 for tails and 3 for type while using Excel’s T-Test function and the formula looks like “=ttest(array of Ad Copy #1 CTRs, array of Ad Copy #2 CTRs, 2,3)” with an outcome of 0.4 which means: we cannot confidently declare those two ad copies perform differently because the chance of our assumption is wrong is 40% . Is it because there is no difference or we need more samples? Let’s add more power to our analysis.

Step Three: Use Power Analysis to Identify Sample Size

In a statistical test, power is the probability that the test result could help us confidently confirm our  assessment and Power Analysis is a simple math we can use to calculate the minimum sample size which enables us to confidently decide whether those two ad copies perform differently. Before conducting Power Analysis, you need to know:

• Required parameters

• Tails
•             One-tail: To validate the conclusion of one data set performs better than the other
• • Two-tail: To validate the conclusion of one data set performs differently from the other
• Test value: the value you would like to compare to sample.
• Sample average
• Sample size
• Sample’s standard deviation
• Confidence level

• Analysis Outcomes

• Result of Power Analysis is between 0 and 1.
•             0:0% possibility to confirm the assessment based upon the given parameters
•             1: 100% possibility to confirm the assessment based upon the given
• For conducting T-Test, we need an outcome of 0.8 or higher from Power Analysis to confirm our assessment parameters

Many universities and research institutes offer Power Analysis tools on their websites to help us conduct this analysis with ease – for example, for example, here is a good one from DSS Research. For our case (tail: 2, test value: 4%, sample average: 5%, sample’s standard deviation: 2%, confidence level: 95%), after entering those parameters into its Power Analysis tool, I learned 32 good samples are required to confidently confirm our assessment. Conclusion: Let’s keep collecting data for now.

In Part III of this series, we will re-visit those two ad copies’ performance. But until then, today’s takeaways are:

1. Having the right samples is a critical step for any analyses. Use Empirical Rule (aka 2-Standard-Deviations Away You Are Out Rule) to control the quality of the sample and use Power Analysis to check its quantity.

Power Analysis to check its quantity.

2. Don’t jump into conclusions based on face value of the numbers. Instead, apply a T-Test to confirm your assessments.


3. T-Test results only confirm one side of story (i.e. “those beers taste no different” does NOT confirm “those beers taste the same”), so clarify your assessment before put it to the test.

Email : naturalseosolution@gmail.com | Mo : 8763723377 | Skype : sunmoon.mohanty

Continue

Market Research for International PPC Success

International advertising is more than just translating campaign components into new languages and changing targeting settings. International paid search advertising in particular is even more complicated given that we are directly replying to search queries in different languages, cultures, and economies. In-fact, simply translating campaigns that were initially built for US geo-targets may be the biggest mistake I see advertisers make when moving into international markets. Well…this and simply applying international geo-targets to existing campaigns written in English. It’s just not that simple guys.

Before setting up your campaigns, or even beginning your keyword research, there are some key pieces of market research you much carry out first. These can really make or break your international campaigns and keep you from wasting resources on marketing efforts that don’t pan out. Here are a few specific areas of investigation that will help you localize your campaigns for success.

Search Behavior

Search behavior differs by country, region, and culture with differences in device choice, search engine choice, and keywords length used to search. For example, did you realize that according to mobiThinking, 30 percent of the world’s mobile users live in India and China? This is a huge percentage of mobile users, so a mobile strategy may be vital to your expansion strategy into these markets.

What about alternate search engines? Do you know about Eniro in Sweden? What about Navar in South Korea? There are a multitude of search engines out there. Google isn’t the heavy hitter in every market. Many of these engines have their own pay per click ad platforms, some of which back fill with Google AdWords, while others stand on their own two feet completely. Know your options; know the people’s choice in each market.

Online Buying Behavior

Online buying behavior is easy in the US, we love eCommerce. Sure, there are still some people who don’t trust their financial info on the web. I’m altogether too trusting when it comes to this I’m sure, but for the most part we’re all about buying online in this country.

What about other countries? Not every culture is as comfortable with this as American’s are and in some countries there are additional barriers to consider. One good example of this I had to learn the hard way while running a campaign in Brazil. The Brazilian economy doesn’t support large credit backed transactions. Brazilians are open to purchasing online, but in order to make most purchases they have to use more than one credit card. Use of multiple credit cards with small balances is a regular and accepted part of the Brazilian economy and payment terms are often referred to as x2 or x3.

Had I known this when I started my campaigns, I would have included this language in my ad copy and advised my client to include it on their landing pages to ensure I was giving the target audience payment terms they were comfortable with. Instead, I spun my wheels for months trying to figure out why my campaigns were under performing in terms of online conversions but my call volumes were high.

Cultural Differences

Cultural differences are probably the most obvious area of research when you sit down and think about it. Do people in Japan or the UK know what Black Friday is? What about Cyber Monday? People in these markets may know what these major commerce days are, but they do something different. European culture has January Sales that start around the same time the major US holiday shopping season, mid-November. It would be a real bummer to run your eCommerce holiday campaigns advertising Black Friday sales in the UK when users are actually looking for January Sales events.

Another good example of a cultural difference is in how people in different countries speaking the same language use different terms to describe the same thing. The classic example is mac and cheese in the US vs. Kraft dinner in Canada. If you advertise coupons for mac and cheese in Canada, you probably won’t do as well as if you were to use the correct terminology for the target market: Kraft dinner.

The same phenomenon is seen with tennis shoes in the US vs. trainers in the UK. We’re speaking English in both markets and we’re talking about the same item, but cultural differences cause us to refer to them in different ways.

Socioeconomic Market Segmentation

Pay per click advertising uses market segmentation and personas just like any other advertising medium. These have to be created specific to each culture or market due to differences in income, occupation, and education, as well as other indicators such as lifestyle, price sensitivity, and brand preference.

Different brands have holds in different markets; a 28 yr. old woman in Mexico will have different interests, responsibilities, and societal expectations than a 28 yr. old woman in Japan. Thus these two women will respond differently to different messaging and will search differently.

These four areas are very important to research and account for in your international search campaigns. Of course, there are additional items to consider like rules and regulations specific to the country you are targeting, language and location settings within your tools, international competitors, barriers to market, and others. But if you start with a good foundation of research in these areas you are better equipped for a successful international PPC campaign.

What research do you conduct before starting your international campaigns? Share with us in the comments!

Email : naturalseosolution@gmail.com | Mo : 8763723377 | Skype : sunmoon.mohanty

Continue