The world of electronic commerce is moving from websites to social media to mobile apps to chatbots today. Ecommerce is a fast paced, competitive and data driven business. So are you making use of data to stay competitive? This is an introductory article on how to do just that, data driven ecommerce conversion funnel optimization.
80% of your online service visitors are willing to pay more for an enhanced experience, and that experience is the key to converting those to loyal customers.
Today securing new users is 500% more costly than holding on to the one you already have, hence, providing an experience your users will cling on to is well worth the investment.
So what experience do users want? A complete one. Users want an experience that helps them better manage time and increase value to their business and life. Information, discounts and product & service recommendations: and these all need to be relevant to each individual user. Where do you start? Every business has different strengths and weaknesses, so start by identifying which area of business is most in need of improvement.
The best way to improve visitor conversion funnels doesn’t start with making changes, it starts with figuring out where those changes should be made.
We need to know first where and why issues exists in online business. So how to analyze the clients conversion funnels to find where we will get the most value? One step at a time, start with the basics.
First any online business already have conversion funnels. So it is about improving the existing visitor-to-customer conversion funnel by increasing the conversion rate.
So what we need is to set up a funnel reporting by setting up goals. Once goals and reporting are ready, we can make them better and keep on improving. It’s not a one night stand, it’s a loving life long marriage.
Before we proceed, let us touch the basics:
How can we make sure we’re getting the most traffic? There’s no sense in dumbing the capital to attract potential customers if they’re leaving the website before making a purchase. To keep a healthy conversion rate, the website needs easy-to-follow pathways to carry these new visitors to the end goals we want. Buying.
A conversion funnel (AKA sales funnel, e-commerce funnel, website funnel, etc.) is a way of describing the path we have designed for a visitor to move through the website. It’s called a funnel because each layer gets closer to the conversion, with a smaller and smaller group in size.
Do we need to gather emails? Use a conversion funnel leading to an email address form. Do we sell products on the site? Use a conversion funnel that carries users all the way from their first point of entry to the “thanks for purchasing!” page.
The conversion funnels can be as advanced as we’d like. It doesn’t matter what we call each step in the conversion funnel. That’s not what this is about: we’re more interested in the specific steps a B2C customer goes through.
Without clear paths to follow, the website will be confusing for customers to navigate. Having specific funnels or paths allows us to test and optimize these conversion paths, leading to higher overall conversion rates.
Once the steps a customer take is mapped into a conversion funnel, the effectiveness of the steps become easily measurable, helping us to find what needs to be improved. This method of breaking customers’ paths into steps also makes it much simpler to find any issues related to the website conversion rate.
So a conversion funnel analysis shows us where in the funnel we needs to work on: it displays what percentage of users make it from one step to the next, so we can find and improve the non-converting spots through conversion funnel optimization.
We need to find the main problem areas before we zero in on the specific problems. Run a first-page first analysis on the primary steps in the process:
Now, determine which of those primary steps needs some conversion rate work: are we seeing an unexpected drop before they even view an item? Or after they view one? Or after they’ve added it to their cart?
Based on this example: If 1,000 visitors viewed the website and if 344 of those visitors viewed an item. 20% of those 344 people (69) added an item to their cart. Of the 69 who added to their cart, 52.2% made a purchase (36). For every 1,000 people landing on the page, roughly 36 of them made a purchase, making it a 3.6% website conversion rate.
Now we look for issues to improve. Which of these conversion rates should we fix in order to achieve the largest possible improvement in conversions? In this case, we’ll take a thorough look at the steps between adding an item to the cart, and actually making the purchase. If only 52.2% make that final conversion, that means over 45% of our potential customers are slipping away. Looking at global averages, 45% cart abandonment isn’t bad, but how can we make it even better?
Now that we’ve decided on the primary stage we want to work on, we can take a closer look, examining each specific steps taken between adding to cart and purchasing to find which of those needs the most improvement.
Nearly 90% of those that added an item to their cart visited the cart afterward, so that step seems okay. It also seems that once someone has entered their personal details, there’s a strong likelihood of them making it all the way to a purchase. However, there is a surprising drop-off between the steps of visiting the cart and entering personal details. Nearly 30% of the people that make it all the way to viewing their cart are then leaving!
This points to the possibility that the personal details form is in need of improvement. Maybe there are errors in the form. Maybe it’s the introductory text, maybe it’s just too long. How do we find out?
In nature everything grows using law of double negation. Law of double negation is a theorem that states that “If a statement is true, then it is not the case that the statement is not true.” This is expressed by saying that a proposition A is logically equivalent to not (not-A), or by the formula A ≡ ~(~A) where the sign ≡ expresses logical equivalence and the sign ~ expresses negation.
How this can be used is, we first make a hypothesis, then try to disprove it. If we try to prove it, we usually run into confirmation bias. Instead, disproving the hypothesis or double negation: that is if we can’t disprove it, then we’ve found the problem.
Our hypothesis is that people leave the personal details form because of the form’s introductory text. We can test that hypothesis by creating several versions of the personal details form, each with a different text.
For that we will be using A/B testing. A/B testing (bucket tests or split-run testing) is a randomized experiment with two variants, A and B. It includes application of statistical hypothesis testing or “two-sample hypothesis testing” as used in the field of statistics. A/B testing is a way to compare two versions of a single variable, typically by testing a subject’s response to variant A against variant B, and determining which of the two variants is more effective.
If all text perform roughly the same, then we disprove the hypothesis, and it’s time to think of another one. If one text performs significantly better, test it again against different text variants. If it still performs better, then use that text on the website, and start testing other possibilities for improvement beyond text, maybe color or design or presentation or steps.
There should be a clear path for the visitors to navigate from start to finish. No matter how they enter the website, make sure there’s a clear path to get to the page we want them to be. Make sure that every page highlights the end of the funnel. Visitors usually abandon the conversion funnel at every stage for unknown reasons, so get rid of any reasons for leaving by optimizing all the known reasons.
Each step of the conversion funnel should be designed with a call for action, be it getting an email address or getting the visitors to the billing page. Define the action needed for each step, making everything else on the page doesn’t prevent that action: everything on the page should lead toward the call for action.
Make every change justified by a test. Use creativity to come up with hypotheses that can be tested. Even add small changes to the website (colors, fonts, hiding elements) and A/B test them. Remember just because a variation worked best last month doesn’t mean it will work this month. Test again.
If you don’t want to use own A/B testing framework, you can use an online service like Google Analytics Experiments feature, but you’ll still have to dig into the code a bit to set the tests up.
Developing a conversion funnel optimization can be challenging. But when we take the time to understand the visitors and learn what they need in order to become customers, we can optimize the funnel to maximize the number of visitors that ultimately become paying customers.