How to do ecommerce re-platforming and not lose 20% of revenue?
Learn a few steps that will make switching to a new ecommerce engine not result in a loss of revenue!
How to do ecommerce replatforming and not lose 20% of revenue?
The biggest fear of ecommerce directors
7 points to apply during migration to a new ecommerce
Summary
I have prepared a list of 7 points that will help you switch to a new ecommerce engine without losing revenue.
Many ecommerce directors and owners of large and medium-sized ecommerces fear that the transition to a new ecommerce engine, i.e., switching off the old online store and launching the new one, will result in a decrease in revenue.
Changing an ecommerce engine is indeed a delicate operation that can go against expectations.
However, we should not approach it as if it were the worst task in the world. Plan the process well, and everything will be fine.
In this article, I will address the concerns and objections of most ecommerce directors and owners of large and medium-sized online stores that are considering changing their ecommerce engine.
I emphasize that this is not just about switching to Magento but about any ecommerce engine.
If your business generates, let’s say, 8.400.000 EUR/USD in revenue per year, you certainly would not want to lose several or even tens of millions of Polish złoty when changing the engine.
The risk increases with the size of the business. If you sell at the level of 56.000.000 EUR/USD per year, changing the ecommerce engine is an even more delicate operation.
Observing various ecommerce engine migrations on the market and talking to different ecommerce directors (I conducted about 100 such conversations in 2023 alone), I noticed several of the same elements that are often neglected, and they ultimately lead to a decrease in revenue.
The following points will help you safely avoid the biggest pitfalls.
It’s enough to say that you need to take care of this. I have dedicated an entire separate post to this topic.
So I encourage you to familiarize yourself with it if you want to learn how to take care of a safe SEO migration and not lose a lot of organic traffic when moving to a new website.
If you are in the process of moving from the old ecommerce engine to a new one, unfortunately, you must expect declines in organic traffic anyway.
They do not have to last long, but for a few weeks, the organic traffic will be less. The most important thing is to take care that the drops are as small as possible and last as short as possible.
It’s crucial for the conversion rate of the new store, which means your new site, to be higher than the rate generated by your old website.
This is difficult to achieve, especially when you have a large number of returning users. Returning users usually lower the conversion rate on a new site because they are used to the operation of the old site and move around easily in it. The new site requires them to learn its functions.
When presenting users with a completely new site, they may feel lost. To make their transition easier, you can launch a tutorial on the new website.
Yet, it’s crucial that when designing the new website, you give special focus to the conversion rate. Ensure that the usability of the new service surpasses that of the old one. This necessitates a comprehensive analysis of both quantitative and qualitative data on how users interacted with your previous site.
Equally important is focusing on clear sales communication, that is, the promotion of temporary discounts, and bundle promotions. This is crucial for the conversion rate.
If you design a new site that generates an even several percent higher conversion rate than the old site, you will offset the drop in organic traffic that will invariably occur right after launching the new site. Conversion rate optimization is a broad topic—creating a new site is a great opportunity to achieve a significant increase in its levels. However, you should start thinking about a high conversion rate at the stage of designing drafts of the new store.
Automated testing is nothing other than a script that simulates user actions on your site.
This can include scenarios such as searching for a product, adding it to the cart, choosing delivery and payment methods, and then placing an order.
Such a scenario can be automatically run, for example, once a day. If problems occur, we will receive clear information that something is not working correctly on the site.
Probably not all functionalities of the service will work perfectly. This is why automated tests are needed so potential errors can be detected as quickly as possible.
It is worth checking regularly whether the introduced changes positively affect the main ecommerce processes. To do this automatically and avoid manual checking, it is worth writing automated tests for a few key processes.
Even one automated test can bring many benefits. But it should be a test for a long process: for example, searching for a product, adding it to the cart, logging in, and placing an order.
If you run it once a day, it will provide a lot of information and protect you from surprises when something is not working correctly. Automated tests are a bit like an insurance policy. It is not worth giving it up—especially if your business is large.
As a person who for many years ran Conversion, a company specializing in web analytics, I can say that the most typical mistake is the lack of consideration of implementing the analytical tools in the migration project to a new ecommerce.
It often happens that the implementation of these tools is incomplete or overlooked.
When the service is already running and revenue starts to drop, a difficult question arises: What is not working correctly?
The simplest solution would be to use analytical tools, which would allow for understanding what is happening with users and making the right decisions. Unfortunately, if web analytics are not configured properly before launching the service, it turns out that we do not have the data that would allow us to identify the problem.
That is why it is so crucial for web analytics to be configured properly before launching the service.
It is crucial to set the right alerts that will inform us when certain metrics start to drop below the levels we consider safe (for example, the bounce rate, the conversion rate, or the loading time of the site—of course, divided into mobile and desktop).
For example, if the weekly or daily conversion rate starts to drop by 50%, this is a clear signal that we should look into it because something on the new site is not working correctly.
Tools such as Google Analytics, Piwik PRO, or Adobe Analytics are extremely useful for identifying potential problems on the new site, as well as monitoring how users use it.
They must be configured correctly and provide us with the necessary information. Thanks to this, when a problem arises, you will be able to locate it quickly. The worst situation is if you see revenue falling, but you do not have data from analytical tools that would show which part of the service is problematic (is something not working on the product card, cart, or maybe the login page?)
It is worth preparing a checklist for launching the new site.
The software house or ecommerce agency should have a list of things to check or change in the configuration when going to the production version of the service before launching the new site.
It is important to thoroughly check whether the data for production accounts, couriers, and payments have been properly configured and do not contain, for example, test data (we have seen something like this too).
This is the most common mistake that reveals itself during the checklist review.
Do not forget to transfer and configure all marketing tags.
In this case, Google Tag Manager is often used, which is, in simplification, a CMS for scripts.
If some marketing tool tags have not been transferred or have been implemented incorrectly, when the new production service starts working, we may lose some traffic sources.
Therefore, it is worth taking care not to make a mistake. I have heard several stories where this topic was not sufficiently cared for.
Another point is to ensure that the product database is completed and full.
When implementing a new ecommerce engine, companies often decide to implement a PIM.
Even if they decide to implement a PIM, they should ensure that all product data, slow-moving data, is in this PIM.
Often only basic information is transferred, such as product name, while other data, such as product photos, product descriptions, and technical specifications, are omitted, which again results in… a decrease in revenue. I know that this point seems trivial and it might seem that it cannot be forgotten.
I decided to describe these issues now because I know that with the beginning of the new year, numerous new ecommerce implementations are starting. In medium and large companies, new investment budgets are opening up, often allocated to creating a new online store or B2B ecommerce.
If you are at such a stage, I encourage you to re-analyze these points because they can prove very useful.
If you plan to switch to a new ecommerce engine, implement a new B2B ecommerce, and do not want to lose 20% of the revenue, pay attention to these 7 points.
If you are planning a switch to new ecommerce engines, are considering Magento, or are not sure which engine would be most suitable for you, we invite you to contact us. We’ll analyze your situation and then develop a concept that will be most reasonable for your business situation.
