How do you create a successful online shop?

Depending on the market and target audience, an online shop can take many different forms. Nevertheless, there are common features that make up a successful e-commerce solution. Technology plays an important role, but even more important are good product management, effective marketing measures and, above all, impeccable service.

Author: Roberto De Simone
Date: 01/09/2024
Updated: 04/12/2024

A person climbs to the summit and stretches his arm upwards as a sign of success. The image is created in comic style

Image ‘Success’ created with DALL-E 3

Introduction

Selling products over the Internet is tempting and is also expected by customers. If a customer is interested in a product or already knows the company offering the product, they google it.
The pressure on companies to set up an online shop is great.
What’s more, with a shop system such as Shopify, it only seems to take a few clicks to set up a shop.

As a result, companies have high expectations that an online shop will quickly lead to significant sales growth.

But what does it really take for the shop to be a success?

The prerequisites and the measures required to achieve a high-turnover shop vary from company to company:

A company with a flourishing offline business and an already established brand will certainly find it much easier to be successful than a company that first has to build a brand through its online presence. It is also easier for companies with niche products than for companies whose products have to compete in a saturated market.

However, it is possible to describe the tasks and measures that - to varying degrees - are necessary for a successful e-commerce solution.

Description of the products

Detailed product data and images

An online customer wants to know exactly what they are ordering to avoid future hassle. It is therefore important that all information about a product and its variants (a product with different characteristics such as colour and dimensions) is easily visible in detail. This also includes images, data sheets and product codes.

Compiling the data is often a challenge, especially for companies with a large number of product variants, as the information is not available in the ERP system in a structured form suitable for the shop or is not available at all.
Preparatory work is therefore required to record the products in detail in a product information system (or in the shop system itself) if this is not possible or only possible with difficulty in the ERP system.

This also raises organisational questions: Which IT system should take on the lead role for product management and how should the interfaces between the individual systems be defined?

Good image material (in different environments) is essential for the presentation of the products in the shop. Professional support from a photographer or graphic designer is needed to create this material. (In this context, a market is currently emerging that aims to simplify the creation of image material with the help of generative AI).
Very innovative solutions also allow products to be visualised in personalised environments.

Retail platforms also face the additional problem that all of the product data has to be imported from numerous suppliers and this data is often not available in a shop-compatible format or the export is inadequate. This applies in particular to product variants with numerous attributes (options).
Importing and cleansing the products is a topic in itself, which can quickly lead to high costs and requires in-depth technological expertise. Generative AI can also be helpful here, especially when it comes to categorising the products and completing the data.

Prices with and without VAT

If a shop is aimed at both private and business customers, the shop should offer the option for users to choose whether they want to see prices including or excluding VAT.

Product availability and delivery times

Customers want accurate and reliable information on availability and delivery times. Customers are also used to fast delivery times thanks to the large online platforms. This places high demands on merchandise management and logistics, which in turn has an impact on costs.
The importance of reliable information cannot be emphasised enough in this context. All the measures described in this post are useless if the service is not right in the end! A disgruntled customer will quickly leave a bad review of the shop if the information in the shop does not match reality.
Incidentally, availability and delivery times are also details that Google wants to see in the SEO-relevant structured data.

Information on product returns

Even if the products are described in detail, customers want to be able to return products. Apart from the fact that there are legal requirements, there needs to be a transparent policy for returns that is clearly visible to the customer.
Since recently (June 2024), information on returns at organisational level can also be included in the structured data.

Reviews

Reviews are one of the main decision-making factors for the purchase of a product. The implementation of a review system is therefore an important part of an e-commerce solution. It goes without saying that only genuine customer reviews may be provided. For this reason, it makes sense to integrate an external system into the shop that is recognised by Google.
However, obtaining reviews is a challenge, as only a small proportion of customers are willing to leave one, especially if they are satisfied customers.
In order to comply with data protection regulations, the customer should be asked after completing an order whether they would like to leave a review of the products purchased after a certain period of time.

Reviews are also important for search engine optimization and these can be entered in the structured data.

In addition to obtaining automated reviews or general feedback, personal contact is also recommended in order to gain a better sense of customer satisfaction and optimisation opportunities.

Payment methods

Business customers

If a shop is aimed at business customers, payment by invoice is the central payment method. The credit check plays an important role here. This process can also be automated, as prominent credit information services have interfaces to their systems.

Private customers

If consumers are the target audience of a shop, the most important payment methods of the target market should be supported; in Switzerland, this is TWINT in particular, which is becoming increasingly popular.

When choosing a payment provider, attention should be paid to the payment methods supported, the transaction costs and the integration options in the shop system. A list of Swiss payment service providers can be found on the federal government’s SME portal.

Support for one-click checkouts, for example with ApplePay, is also recommended. This means that the customer no longer has to enter their contact details at checkout, as these are taken directly from Apple. This is a very convenient function, especially for users who order via mobile phone.

Search Engine Optimisation (SEO)

SEO is a fundamental component of an online shop.

It is the dream of every company that the online shop can be found by googling alone. Unfortunately, the reality is different, even with a fully optimised shop:

Niche products and established brands have a relatively easy time achieving prominent placement. Products in a competitive environment have a difficult time.

Google generally favours prominent shops with a large product range that receive a high number of clicks. It is also said that a shop has authority on a topic.

New shops need time to become search-relevant.

Good and helpful content

Good and helpful content is still the basic prerequisite for a good ranking. It is also important to design the content in such a way that it also contains topics (search terms) that are effectively searched for. Google Trends and the Google Search Console are the first choice of tools that can be used to optimise content. SEO-specialised tools such as Ahrefs or Semrush are also helpful.
There is no real answer as to whether the content should be created with the help of generative AI. I would tend to say no. In the recent past, AI-generated content has led to many highly optimised ‘rubbish’ pages that have achieved top rankings. Google is now counteracting this - see also the August 2024 Core Google Search Update.

If products are imported, the options for optimising the content without great effort are limited.

Accessibility and clear structuring of websites

It is important to structure websites in a semantically clean way (use correct HTML tags) and to make them accessible so that the crawlers can understand them.

Tools such as Lighthouse or screen readers help with optimisation.

Extensive range

An extensive product range is certainly useful for achieving authority on a topic. The more you find in a shop, the more its pages are clicked on, which in turn emphasises the importance of the shop.

Backlinks

Genuine backlinks (referencing of products on other websites) are still important. However, these are difficult to obtain or are associated with costs if you want to persuade a (reputable) third party to report on a product in a post or specialist article and link to it.

Blog/VLog

Writing regular posts with useful tips and tricks about the products on offer is helpful, as posts emphasise expertise on a topic, lead to more pages and in turn generate more clicks. Another idea is to set up a VLog in order to be able to operate the video channel. However, you need to be aware that both of these are long-term measures, the results of which are not immediately visible.

Fast page views (Core Web Vitals)

Google loves fast pages as this leads to a good user experience (see also next section). To this end, Google has introduced the Core Web Vitals values, which are a ranking factor for placement in the search results. The optimisation of a page can in turn be done with the help of Lighthouse.

With all the hype surrounding fast websites, it should not be forgotten that the Core Web Vitals values are only one ranking factor of many. Even if a page achieves values of 100, this does not mean that the page will also achieve top rankings in the search results.

User experience of the shop

An easy-to-use, clear and fast user interface is the basic prerequisite for the acceptance of a shop. If the customer finds it difficult to find the products or cannot immediately find their way around the checkout, they will leave the shop again.

It goes without saying that a shop needs a responsive design. Nevertheless, it is alarming to see how many shops have problems with ordering via mobile devices - especially when it comes to entering data. This often has to do with the fact that during development, the shop is primarily tested on the desktop and not enough on various mobile devices or emulators.

If the shop is aimed primarily or also at private customers, special attention should be paid to its use on mobile devices.

Personalisation

Personalisation is another concept to improve the user experience. This can range from individual product recommendations, prices and offers to a personalised user interface. (The use of Generative AI is again suitable for product recommendations).

Personalisation is also an attempt to achieve closer customer loyalty.

Another form of personalisation is the personalisation of the product itself (if personalisable products are sold in the shop). This includes, for example, the functionality for customers to upload their logo to the shop and place it on the product. Ideally, such personalised products are displayed three-dimensionally with the help of WebGL-based frameworks such as three.js.

Confidence-building measures

Certifications

Certifications of all kinds with the corresponding labels, which are prominently displayed in the shop, help the customer to gain confidence in the shop.

Compliance documents

The shop must fulfil the compliance requirements applicable in the target market. The relevant documents must be clearly written and easy to find. Compliance and certifications are usually linked.

Contact options

The customer must be able to contact the shop operator to obtain support in the event of questions or problems. Ideally, this should be possible by telephone. The contact options should be visible at first glance.

Advertising

As can be seen from the previous explanations, a new shop that does not offer niche products will not be visible by googling. It must therefore be advertised.

We generally recommend working with an online advertising agency that is familiar with the placement of advertising on all relevant channels.

Collaboration with comparison portals or marketplaces is another way to publicise the products. Data exports must be provided for these, which can be time-consuming depending on the product types and requirements of the portal.

In addition to advertising on online channels, traditional advertising such as mailings, trade fairs, advertisements in trade journals (websites) and the associated placement of trade articles should not be forgotten.

Advertising is expensive and you need to be aware of this before you even think about launching a shop in a competitive environment.

Analytics

A shop requires constant optimisation. This is only possible if you have a detailed overview of search queries, page views, user behaviour, sales figures, customers and competitors.

The Google Search Console is the most important tool for determining which search criteria are used to find the shop.

The integration of analysis tools in the shop allows the user behaviour of visitors to be determined. However, it is becoming increasingly difficult to harmonise this with data protection. The use of server-based analytics is therefore becoming increasingly important.

Segmented sales figures can usually be taken from the shop or ERP system.

In a competitive environment, price comparisons with competitors are of crucial importance. If the prices are not competitive, the customer will buy elsewhere. Automated solutions can also be developed for price comparison.

The development of an analytics platform should also be considered, with the help of which data from various sources can be summarised and presented in a dashboard. In this context, Microsoft Fabric or Google Cloud Looker with Big Query should be mentioned as possible platforms.

The use of Generative AI

In addition to the aforementioned use of Generative AI for importing products, creating image material and for product recommendations, Generative AI can also be used for product searches within the shop. The use of LLMs (Large Language Models) and the vectorisation of product data allow products to be found even with less precise search terms, as is possible when googling.

As already mentioned, I would refrain from using generative AI for the creation of posts or product descriptions, as this could have a negative impact on SEO.

The use of generative AI is ideal for the localisation (translation) of product descriptions and posts. Services with APIs such as Google Translate or DeepL Translate, which can be used to automatically translate content, have been available for this purpose for many years. The quality of translations has steadily improved thanks to the new AI-based technologies.

The use of chatbots for customer support is tricky as long as the hallucinations cannot be completely controlled. However, concepts such as RAG (Retrieval Augmented Generation) or a broad context window with context caching help to reduce hallucinations. The AI model is extended with the user’s own data.

Generative AI, on the other hand, can be used for data analysis. The prominent analysis platforms already mentioned offer this functionality.

Interfaces to the ERP system

Interfaces to the ERP (Enterprise Resource Planning) system primarily increase productivity in the back office. Product mutations, including price changes, which are automatically transferred from the shop system to the ERP system or shop orders which are transferred directly to the ERP system are particularly worthy of consideration.
Customers can thus benefit from the latest data on stock levels and delivery times.

Here too, organisational questions arise as to which system takes the lead and where which tasks are carried out.

The integration of ERP and shop systems is usually rather complex and also depends heavily on the systems used and their interface options (APIs).

Communication between the ERP and shop system should take place via a message broker, such as RabbitMQ, Apache Kafka or via a cloud-based solution such as Google Cloud Pub/Sub.

Summary

Setting up and running a shop is not a sure-fire success. It requires constant observation and adaptation to changing market conditions.

The shop must be constantly updated in terms of content, product range and prices. The customer should also be offered an exciting user experience by enticing them to visit the shop regularly with changing offers and new, useful specialist articles.

The shop must also always be kept up to date in terms of technology. Security aspects with associated updates, changing technologies or changing requirements of big tech companies require continuous adjustments.

Of course, new technologies also constantly open up new possibilities for functionality, which in turn lead to a better user experience.

It goes without saying that all of this involves significant effort!