Blog

30 Days Of E-Commerce Testing - Day Thirty!

Day 30
Find 4 different types of credit cards and learn what makes them different - e.g number length.

The last day!

last-stock-photo.jpeg

I can just tell you what I know.

VISA cards always start with a 4, Mastercard with a 5. Both have sixteen digit numbers.
AMEX only has 15 digits, and lots of people complain about their high fees.

VISA, Mastercard and Discover are all tied to banks. AMEX isn’t.

That’s all well and good.

VISA and Mastercard offer both Credit and Debit cards (I’m not sure if the others do?)

Fees are generally higher on credit cards than debit cards, and it’s really hard to tell before you’ve been charged whether a card is credit or debit.

I have a great story about my credit card being stolen… but… that’s for another time!

And that’s all!

This has been a great 30 day challenge, but it took me way too long. I need to only do these in a less busy month!

My favourite day was actually day 29 - I learned so much. Thanks for reading!
- JE

 

30 Days Of E-Commerce Testing - Day Twenty-Nine!

Day 29

Research and share how Search Engine Optimisation(SEO) and ecommerce work together.

seo.jpg

Day 29 is the most challenging, but most interesting of the exercises so far.

I had trouble finding resources, but, this guide to SEO on moz.com blows me away. It’s such a good introduction. I really recommend reading it.

I’ve learned a bunch about SEO. Let me tell you.

Links are important
Search engines crawl your site based on links.
If a page is not accessible via a link, then search engines won’t know about it.

In an e-commerce context, imagine you have product pages that are only available via search.
Because they’re not linked, search engines can’t find them!

This is why it’s important to have a nice, easy to navigate, tree structure of products that you can browse to - in categories, etc.

Choosing the right keywords is an art form
You want keywords that represent your business.

So like, if you sell shoes - then “shoes” is a good bet.
But - lots of other organisations sell shoes - it will take a lot to beat the competition for a keyword like this!

A more specific keyword is better - like “purple alligator skin shoes”

But of course, that might be *too* specific and might make it hard for people to find you.

You have to find the right balance!

Implementing keywords is an art form too
Not enough of the same keyword won’t register with a search engine.
Too much of the same keyword will flag up as spam.

Funnily enough, the best thing to do is to use your keywords with some frequency, but in a natural way.
Make sure it’s mentioned enough on the page that it’s clear - that’s the good or service you’re selling.
Mention it in places like the title, meta tags, etc - but don’t go overboard!

Duplicate content is bad

Search engines pick up on sites that have multiple pages with the same content.
Sites may legitimately do this - for example, a regular page, and a ‘print optimised’ version.
But it’s a sign of poor content too, so search engines rank these pages lower.

Avoid duplicate content! But if you must have it, try the ‘canonical’ link tag, so that search engines are aware of it!

Use HTML as much as possible

Rich content like iFrames and Flash can’t be read by search engines!
Try not to use them. But if you must, supplement with text and HTML too - so your great content doesn’t get missed!

I’m so into this right now - looking forward to learning more!

- JE

30 Days of E-Commerce Testing - Day Twenty-Eight!

strugglingforgoodstockphotos.jpg

Day 28
Identify potential ethical issues with the use of ecommerce data. What are some public examples?

Again, I’m going to outsource my answers here. I’ve had to do the research though, so rest assured I’m learning stuff :)

Acxiom.com gives five simple rules for ethical data collection.

Ultimately, I think it boils down to
- know what you’re collecting
- know why you’re collecting it
- take great care when using it!

Umbel.com gives seven examples of cases where data has been treated insensitively or carelessly.

Interestingly, I don’t think this is a new problem, or one particular to E-Commerce.
Stories like these have been happening ever since stores started making mailing lists and sending out personalised marketing material.
E-Commerce has just… exacerbated the problem!

I think the thing to learn from those stories is to think carefully about what you’re doing marketing wise.
There’s room here for some testing of marketing material. What is the worst case scenario? What are the edge cases of people you’re marketing to?

Turns out, marketing needs testers too :)

- JE

30 Days of E-Commerce Testing - Day Twenty-Seven!

Getting there!

anotherstockphoto.jpeg

Day 27
How can we use the analytics data from ecommerce sites to help us test?

This is kind of a follow on from day 26 huh, so let’s try using that as a basis.

The most obvious thing is looking at landing pages.

This is a nice pointer as to where our testing should be prioritised. If most users are landing on one particular page, we should probably focus some testing effort there first, before other less common landing pages.

Conversion rate can be used in the same way.

If we get a high conversion rate from mobile, then this would inform us that we should concentrate testing efforts on mobile first.
If most income comes from returning users, we should concentrate on those flows first.
If we get more income from guests or first time buyers, then we should focus testing efforts there.

Bounce rate

Bounce rate can inform testing in a different way.
If there are spikes in a bounce rate - maybe a page is down?
Or if a particular landing page bounces more than others - maybe there’s a problem with it?
Bounce rate can give off a ‘smell’ of places where there might be a problem - that needs further investigation.

And that’s that! On to day 28!
- JE

30 Days of E-Commerce Testing - Day Twenty-Six!

Day 26
Learn about how users are tracked on ecommerce websites/apps. Blog your findings!

Welp, I’ll just redirect you to a couple of other blogs I looked at: Shopify on web analytics and Kissmetrics on metrics.

Here’s some things I learned.

Study your bounce rate
Bounce rate is the percentage of visitors that come to your site and immediately ‘bounce’ away again.
I’m surprised at the typical numbers being around 50% - that seems high to me.
It’s a good number to have though. If it’s high, once you know it’s high, you can take steps to improve it.

Landing pages
The landing page is the first page a person lands on when they visit your site.
It might not always be your homepage! It could be a promo page, or a product page, or something else.
It’s good to work out what landing pages are most popular, and leverage this information.

Conversion rate
Conversion rate is the percentage of visitors who actually purchase something.
Learning about the sources of your most conversions can be revealing.
Do they come directly from another website, or a search engine?
Mobile or desktop?
What percentage of those conversions are new visitors vs returning customers?

All this data will inform your marketing strategy and, if utilised correctly, will lead to better sales!

- JE

 

AnalysisStockPhoto.jpg