When it comes to understanding your ecommerce clients, you need the following information:

#1 Geographic data of your best ecommerce clients
#2 Demographic data of your best ecommerce clients
#3 Psychographic data of your best ecommerce clients
#4 Understanding of the behaviour of best clients
#5 Understanding of, how your best clients are being acquired
#6 Understanding of the type of contents being consumed by your best clients
#7 Understanding of the type of products being purchased by your best clients

In this article, I am going to explain all of these factors in great detail.

There are two main advantages of knowing about your best ecommerce clients:

1. You can improve your business bottomline by reducing the cost per acquisition.
2. You can target your best clients more effectively by structuring your marketing campaigns around them.

But first, you need to be with me, on why you need to focus on acquiring more of the best ecommerce clients and less of low-value clients.

Your best clients are almost always more profitable than average/low-value clients

Unless you operate on a very low profit margin, where any decrease in sales volume will quickly erase profit and result in net loss, there is absolutely no excuse for you, to not to focus on acquiring more of the best clients and less of the low-value clients.

Companies like Walmart, Tesco, and other FMCG giants operate on a very low profit margin.

Since they operate on a very low profit margin, they have to focus on improving the top-line growth, like sales volume to remain profitable. If they don’t, then they will incur huge losses and will soon be out of business.

But this is not the case with the majority of businesses out there.

The majority of businesses out there are small to medium size, and they do not operate on a very low profit margin, but yet they continue to function in a way like, that they operate on a very low profit margin.

So they need a lot of clients, any client will do.

According to Pareto Principle (also known as the 80–20 rule), 80% of your sales come from 20% of your visitors.  This 20 % of your visitors are your best customers. The other 80% are your low-value customers which are the major source of your acquisition cost. For example:

Let us suppose that your target market is the US. So your average customer can be anywhere from the US. Let us also suppose that after analysing one year of data, you found out that customers from New York City are the most profitable:

From the table above you can see that the gross profit is higher in the case of New York City clients. But since the focus has always been on topline figures like orders and sales, the marketing budget is spent on acquiring clients from all over the US.

Yes, your sales are higher when you target all of the US. Yes, you get 3 times more orders when you target the entire country. But at what cost?

There is a metric called ‘Return on Advertising Spend (ROAS)’. It is used to measure the cost-effectiveness of a marketing campaign.

Not many marketers talk about it because it includes “cost” and many of us are committed to focusing only on the topline growth: traffic, conversions, orders, and sales in the name of improving the business bottomline.

But the fact is, we can never truly improve the business bottomline by ignoring the “cost”. So let us calculate the ROAS.

ROAS in case of New York City = $21000/$2000 = 10.5. For every dollar you spend on acquiring clients from New York City, you earn $10.5 in return.

ROAS in case of all of the US (excluding New York City) = $32000/$15000 = 2.13. For every dollar, you spend on acquiring clients from the rest of the country, you earn $2.13 in return.

Your marketing dollar would be better spend, in acquiring more of New York city clients and less of clients from the rest of the country.

As long you will ignore the business bottomline metrics like “cost”, you can never focus on acquiring more of the best clients and less of average/low-value clients.

From the table above we can conclude that the cost per acquisition will remain high if you target the whole of the US through search marketing campaigns or any other ad campaigns. So it is pretty obvious that your total ad spend is going to remain high, for acquiring average clients and you will remain less profitable for the foreseeable future.

Related Post: 7 Powerful methods to improve your Business Bottomline

Getting Geographic data of your best ecommerce clients

As simple and obvious as it may sound, many business owners have no idea where their best clients are coming from. You need to know your most profitable geo-locations.

For an ecommerce website, the most profitable geo locations are those which have generated the maximum amount of sales in the recent time period (last 3 to 6 months max)

“But hang on, shouldn’t you be using metrics like ‘conversion rate’ or ‘average order value’ to determine most profitable regions”, you may ask. My answer is “NO”.

The reason being, I give preference to number metrics (like raw conversions, sales volume) over ratio metrics (like conversion rate, average order value). If you are an avid reader of this blog, this should not come as a surprise to you.

However, if you are a new reader then you need to check out this article to know exactly ‘why’: Here is Why Conversion Volume Optimization is better than CRO

Because of data collection, data interpretation, data reporting and optimization Issues, ratio metrics like conversion rate and average order value are not reliable for taking business/marketing decisions.

So before you take any decision on the basis of ratio metrics like ‘conversion rate’ metric, ask yourself two questions:

1. Is my conversion rate calculation accurate?

2. Is my conversion rate statistically significant?

The conversion rate calculated by Google Analytics is not accurate by default as GA treats every person who has visited your website as your potential client and hence puts everyone in the conversion funnel while calculating conversion rate.

Also, Google Analytics is not going to tell you whether or not its conversion rate is statistically significant or whether the difference in the conversion rates of two or more marketing campaigns is statistically significant.

Similarly, GA won’t tell you whether your ‘Average order value’ is statistically significant.

So just because someone from ‘Wisconsin’ placed a huge order in a single transaction, does not automatically make the region a profitable location to target.

You need to have confidence before you play with ratio metrics like ‘conversion rate’ and I am not talking about the everyday confidence here. I am talking about the ‘statistical confidence’

I am the confidence you need, to play with conversion rates.

Yes, you need a very good understanding of statistics to play with ratio metrics, without getting burned. Only a handful of marketers/analysts go to great lengths to calculate statistical confidence, each and every time, they play with ratio metrics.

Then there are marketers/analysts out there who take marketing decisions on the basis of statistical significance but statistical significance alone.

So if the increase in conversion rate is statistically significant, then they declare ‘SUCCESS’ in their A/B or multivariate test.

But what if their success is not practically meaningful. What if their results are statistically significant but still trivial or statistically insignificant but still important?

Just because a result is statistically significant, it doesn’t always mean that it is practically meaningful.

So what they are doing wrong here? Well, they are not taking ‘effect size’ into account.

“What is effect size?” you may ask.  Effect size is the measure of the strength of a phenomenon and a decent reminder that ‘a little knowledge is a dangerous thing’

To learn more about the issues with ratio metrics, check out this article: What Matters more, Conversion Volume or Conversion Rate – Case Study

Do you see the issues with ratio metrics now?

In my opinion and experience, any person who is not good at maths and statistics behind web analytics should not be taking business/marketing decisions based on ratio metrics alone.

I avoid taking decisions on the basis of ratio metrics alone, as much as possible even though I have a pretty decent knowledge of statistics.

However when I have to make decisions on the basis of ratio metrics (like in case of calculating cost per acquisition, return on ad spend, ROI, calculating percentages, etc) then I do it the right way, which is via:

• data segmentation
• calculating the statistical significance
• determining the effect size of the results and
• above all looking at the big picture (taking the context into account)

Now back to determining the most profitable locations.

1. Go to the ‘Location’ report (under Audience > Geo in your Google Analytics account).
2. Click on the percentage tab.
3. Select ‘Revenue’ from ‘contribution to the total’ drop-down menu.
4. Select the most recent time frame (e.g. the last 3 months).

You will now see a report like the one below:

From the report, we can conclude that the majority of sales come from Florida. Which also means that our best ecommerce clients live there. So it is prudent to focus on this region.

Now I want to know the demographic details of my best customers so that I can understand them better.

Getting Demographic data of your best ecommerce clients

Demographic data can be age, gender, education, occupation, income etc.

Through demographic data, you can determine how your best customers’ behaviour varies by gender (male or female). For example, do men buy more from your website than women? or do women buy more?

To get the demographic data in Google Analytics reports, you need to enable the demographic reports by going to Audience > Demographics > Overview and then click on the ‘Enable’ button.

Once you have got the demographic details, you can structure your content and marketing campaigns accordingly. For example, promotional efforts, ads, contents, etc. that work well on men, may not work on women and vice versa.

You can use the demographic data, to build and apply advanced segments to your Google Analytics reports and can greatly improve your understanding of the target audience.

Since my best ecommerce clients live in Florida, I created an advanced segment which extracted all of the data for Florida:

Now go to the ‘Age’ report (under Audience > Demographics in your GA account), click on the percentage tab, and select ‘Revenue from ‘contribution to the total’ drop-down menu. Make sure that the Florida segment has been applied to it.

You will now see a report like the one below:

From the report, we can conclude that majority of people from Florida who buy the product are in the age group 35-44 years.

Now go to the ‘Gender’ report (under Audience > Demographics in your GA account), click on the percentage tab, and select ‘Revenue from ‘contribution to the total’ drop-down menu. Make sure that the Florida segment has been applied to it.

You will now see a report like the one below:

From the report, we can conclude that majority of people from Florida who buy the product are women in the age group 35-44.

Well, that is all you can find from Google Analytics at the moment regarding the demographic data.

There is one caveat here. You can’t blindly rely on just the Google analytics data. This is because GA demographic data is based only on the subset of your total site visitors. This data is not available for every single person who visited your website.

If your demographic reports are based on more than 500k visits then Google Analytics is going to sample your data badly whether or not you use GA premium. There is no escape from data sampling issues in GA.

Contrary to this, if your demographic reports are based on only say 10-20% of your total site traffic then you will get inaccurate picture of your visitors’ demographic details.

Remember demographic reports were introduced in Google Analytics only a few months ago. So you should not rely on Google Analytics alone for all of your demographic data needs.

There are tons of tools out there that can provide demographic data of your target audience and in much more detail.

For example:

1. Quantcast: https://www.quantcast.com/ (works if the website is quantified and is a high traffic website)
   
2. Compete: https://www.compete.com/ (best for high traffic websites)
   
3. Experian Consumer Insight: https://www.experian.co.uk/marketing-services/solutions/insight/ (best for enterprise clients)
   
4. Google Display planner: https://www.google.co.uk/think/products/display-planner.html
   
5. All government websites which provide demographic, statistics and census data like the U.S. Census Bureau https://www.census.gov/ ,
   
6. Popular Industry blogs and online magazines – industry blogs and magazines can provide a lot of valuable insight about the demographic of your target audience.
   
7. Online survey tools – survey tools like https://qualaroo.com/ is an excellent way to know anything about your target audience. I recommend using online surveys all the time. 24 hours a day, 7 days a week.
   
8. Contests – another good and clever way to collect the demographic details of your clients.
   
9. Your client – if your client has been running a well-established and profitable business for a long time, he/she can provide you with valuable insight into the target audience.
10. Facebook Insight –  works well if you have a large fan following.
11. Follower Wonk – works well if you have got a lot of followers.
12. LinkedIn Analytics – depending upon the industry you are in, you can get a lot of valuable insight from this tool.

 Getting Psychographic data of your best ecommerce clients

Psychographic data can be interest, personality, lifestyle, values, beliefs, liking, disliking, affinity etc.

Once you have enabled the demographic data in your GA account, you can also view the psychographic data in your GA reports. Through psychographic data, you can create better and stereotype-free gender-based marketing campaigns and content.

A stereotypical gender-based campaign can be something like using ‘pink’ colour for the display ads, selling only soft toys to young girls (like there are incapable of playing with any other toy), developing the so-called “girly design” for landing pages, writing cookery/child care/home improvement contents just to target women.

Not all women like the colour pink. Not all women are interested in learning about home improvement and other domestic stuff. But that’s how it goes when marketers have to drop gender into an ad campaign.

You can see this gender-based stereotype imposed on people everywhere from billboard ads, product shelves to TV ads. This ignorance comes from not understanding the diversity of preferences among men and women. Marketers need to move on from the pure gender-based segmentation to psychographic based segmentation.

So if your target audience is mainly women then structure your contents and campaigns according to their liking, personality, and interest and not on the basis of what all the women are assumed to like in general. Not all women are the same. So you can’t have ‘one size fits all ads/contents’ for women. The same applies to men.

Now we will make the following changes to the ‘Florida’ advanced segment:

You will notice here that I have not selected the age group ’35-44’ while creating this advanced segment even when the women in the age group 35-44 buy the most according to Google.

This is because I don’t trust GA demographic data that much. After all this demographic data is just a subset of the overall data. So in order to minimize error, I have deliberately chosen not to segment this data further by age-group, affinity category, in-market segment, and other categories. The more I will segment this data, the more inaccurate my analysis and interpretation will be,  especially if I am dealing with a small data sample.

Now open the ‘Affinity Categories’ report (under Audience > Interest in your GA account), click on the percentage tab, and select ‘Revenue from ‘contribution to the total’ drop-down menu. Make sure that the new ‘Florida Women’ segment has been applied to it.

You will now see a report like the one below:

From the report above we can conclude that majority of Florida women are technophiles, news junkie, avid readers, TV lovers, music lovers, and travel buffs.

Now open the ‘In-Market Segments’ report (under Audience > Interest in your GA account), click on the percentage tab, and select ‘Revenue from ‘contribution to the total’ drop-down menu. Make sure that the ‘Florida Women’ segment has been applied to it.

You will now see a report like the one below:

In-Market Segments are the segments in which people are most likely to be ready to purchase products or services.  So according to the report above the majority of Florida women who are most likely to make a purchase are from the consumer electronics and real estate market segments.

So you can pretty much assume that my target audience is tech-savvy and probably use multiple devices to browse the web.

However, all of the psychographic data you have acquired so far is still a hypothesis as it is based on just one tool called ‘Google Analytics’. In order to confirm your hypotheses, you must collect data from multiple data sources and integrate them before you draw any conclusion.

Since psychographic data is a subset of demographic data, whatever tools you use to get demographic data can be used to acquire psychographic data.

Understanding the behaviour of best ecommerce clients

Open the New vs. Returning Report (under Audience > Behavior in your Google Analytics account)

From the report above we can conclude that Florida women who make the maximum purchases from the website are returning visitors. So getting repeat visits is very important for this website and this is possible only through user engagement.

Now open the Engagement Report (under Audience > Behavior in your Google Analytics account) and apply two more advanced segments called ‘visits with conversions’ and ‘visits with transactions’.

You will now see a report like the one below:

From this report, we can conclude that I need to hold my target audience for at least 180 seconds or 3 minutes on my website to get any possibility of conversion and sales.

Any visit duration which lasts for less than 3 minutes is not profitable for my client’s business as it won’t likely result in sales. This analysis again highlights the importance of developing strong user engagement with the target market.

WHAT’S INSIDE: My Step-By-Step Blueprint For Generating Record Sales And ROI By Leveraging Analytics Data.

Learn To Generate Record Sales And ROI By Leveraging Analytics Data. (150 pages)

Yes I want the ebook

Understanding how your best ecommerce clients are being acquired

Open the ‘All Traffic’ report under ‘Acquisition’. Click on the percentage tab and Select ‘Revenue’ from ‘contribution to the total’ drop-down menu.

You will now see a report like the one below:

From the report above we can conclude that the best ecommerce clients come from Google organic search.

Now in order to determine how effective the acquisition strategy is, click on the ‘Comparison’ tab on the reporting interface. You will now see the following similar report:

So compared to the site average, the performance of Google organic search in terms of generating sales is highest. So once again we can conclude that the best ecommerce clients come from Google organic search. So SEO is very important and profitable for the website.

Understanding the type of contents being consumed by your best ecommerce clients

If you can determine the type of content being consumed by your best clients, you can produce more of such content to increase your sales.

One of the most important and yet not so obvious differences between a low converting website and a high converting website is the volume of profitable content on the site.

Create ‘Profit Index’ custom report in your Google Analytics account and then apply the ‘Florida Women’ advanced segment to it.

Note: Use ‘Page Path Level 1’ as your primary dimension while creating the profit index custom report.

Profit index is a database of all those web pages on your website which have the capacity to trigger conversions and/or transactions.

I have explained in great detail about creating profit index in this article: Optimizing Contents for Sales and Conversions through Profit Index. Read this article if you are not sure about how to create a profit index for your website.

Once you have the profit index report with the ‘Florida Women’ advanced segment, you can determine the type of contents being consumed by your best ecommerce clients.

Click on a content category to dive deeper and determine which contents are being consumed in that category. You need to develop more of such content to increase user engagement and sales.

Understanding the type of products being purchased by your best ecommerce clients

Sell what is selling and sell even more. There is no point in spending time and money on products which people rarely buy. Focus on the products people buy the most, especially your best ecommerce clients.

Open the ‘Product Performance’ report (under Conversions > Ecommerce in your GA account) with the advanced segment ‘Florida women’ applied to it:

From this report, you can determine the type of products being purchased by your best ecommerce clients.

So now you know your best clients and you also know the product they buy the most. The next thing that you have to do is to promote these products among your best clients and laugh all the way to the bank :)

Some caveats you need to be aware of

Understand that this is an oversimplified guide for finding and engaging with best ecommerce clients. I have not taken many factors into account like seasonality, attribution modelling, customer lifetime value, ZMOT, etc.

I have not really used many data sources for my analysis. In my analysis, there is almost always a clear winner. This may not be the case with you.

Maybe you have multiple profitable locations. Maybe in your case, both men and women generate an almost equal volume of sales. Maybe you need to dive deeper and determine which cities in Florida are most profitable. May be targeting Tampa women will generate more sales than targeting Orlando women.

There can be so many situations that it is impossible to factor them all in just one blog post. But hopefully, you now have a better understanding of finding and targeting your best ecommerce clients.

Cheers!

Another article you will find informative: 7 Powerful methods to Improve your Business Bottomline