FAQ: All

Question 1

What is attribution for an online store?

Accepted Answer

Attribution describes how a conversion is assigned to each of the channels involved in the purchase funnel. Through attribution, we can assess the effectiveness of individual communication channels and determine their role across the entire customer purchase path. Before buying, customers reach our online store through many sources: paid search results, price comparison engines, social media — which is why attribution models exist, used to assign a share of the conversion to specific marketing channels. We often limit ourselves to checking the effectiveness of only the last source before purchase. In reality, however, analyzing the entire path the user has taken is extremely important. Why is it worth analyzing different attribution models rather than only the defaults visible at first glance in various marketing tools? We may mistakenly cut investment in traffic sources that don't appear to convert, but actually serve as the first touch with our store or reinforce the customer's purchase decision (for example, inspirational content on our social media). By cutting the budget for these channels, we may make sure the customer never even hears about our offer.

Question 2

What are the most common mistakes when implementing web analytics?

Accepted Answer

The most common mistakes stem from an incomplete understanding of what web analytics actually is. It often happens that we don't fully know what we are really measuring and how to draw business-relevant conclusions from it. Many e-stores deploy only the basic Google Analytics tracking code, which measures only the pages users visit, or they misinterpret metrics (bounce rate is one example). It also frequently happens that a lot of information is collected, but its quality is questionable: e.g. due to incorrectly firing tracking codes on the page or because employee traffic was not excluded.And what if you do know exactly what you measure on your store and you're confident in the data quality? When data lacks context and is not properly presented, it can be hard at first glance to discover interesting relationships — that's why it's so important for the entire e-commerce team to engage in producing coherent, readable reports or dashboards.

Read more on this topic in the article on our blog!

Question 3

How to plan a budget for online campaigns?

Accepted Answer

There are several key elements that let you estimate an effective budget for an online store: among them are CPC rates (cost per click) — that is, how much we pay for a click on our ad — which is the most commonly used billing model for online campaigns. We can check average rates for our industry in the Google Ads keyword planner, among other places. 
The next factor is the conversion rate: according to many industry reports, the average conversion rate in e-commerce is 2%. Knowing how much we pay on average per click (acquiring a user), the average order value, and how many customers we need to reach a satisfactory revenue level, we can estimate the amount needed for paid advertising. 
In addition to such simple calculations, it's worth remembering that e-commerce online campaigns are not only the ones aimed at conversions (purchase). You also need room for top-of-funnel activities aimed at building brand awareness and customer recognition. Part of the budget should also be reserved for tests (new ad creatives or target groups).

Question 4

Cross-sell vs up-sell — what are the differences?

Accepted Answer

Cross-sell encourages customers to buy related or complementary products. The goal of such activities is to show the customer related items (e.g. extra memory for a phone) so they buy them from us as well, increasing order value. It is a very effective tactic that exposes customers to a wide range of the catalog: we can show users products they didn't know we had on offer. Up-sell is a tactic of showing customers a more expensive product than the one they're currently looking at — a product with better specifications and more features. Both strategies are often used e.g. on product pages ("Customers also bought" or "You may like" sections), in the on-site search, or at the cart level. Cross-sell and up-sell can also be applied in remarketing campaigns instead of standard product reminders the user has already seen.

Question 5

What is a product feed?

Accepted Answer

A product feed is a file, usually in XML or CSV format, that contains information and specifications about the products in your online store. Beyond names and prices, it also includes descriptions and attributes. With such a file you can publish your offer for example:
in price comparison engines,
in product campaigns such as Google Shopping, which appear at the top and on the right side of Google search results,
in Facebook campaigns, including remarketing campaigns showing the user products they recently viewed on your store, or presenting a product catalog to a recipient who has not yet visited the store.
 You can build a feed yourself e.g. in Google Sheets or use a built-in feature on your e-commerce platform. Bear in mind that each provider may have different requirements for the file (e.g. image size or format). It's important to keep your product feed up to date and not show, say, incorrect prices for your products.

Question 6

How to measure the effectiveness of e-commerce campaigns?

Accepted Answer

Although the first metric that comes to mind for e-commerce campaign effectiveness is ROAS (Return On Ad Spend), it's worth remembering that it's not the only correct or universal metric. Yes, it tells us how much revenue every złoty spent on campaigns generated, but revenue is not always the best reflection of effectiveness. A more comprehensive view of the data may lead you to insights such as:

a campaign generated high ROAS, but it was a clearance sale on your site — you sold products at a very low margin (campaign ROI, Return On Investment, was low), and some customers will only return when you offer similar prices,
 a campaign generated high ROAS but was based on users who already added something to the cart (a remarketing campaign), so they had a higher likelihood of converting — it's hard to compare its results with a campaign targeting customers who don't yet know your brand,
 a campaign generated low ROAS but brought loyal customers who signed up for your newsletter (which can be considered a micro-conversion) — you can later activate them practically for free via email.

Another factor is that ROAS will not be the best effectiveness indicator for every e-commerce campaign. For marketing activities to be effective, we have to align them with the level of the purchase funnel.

Campaigns aimed at building awareness will rarely be a success in revenue terms — judging them by ROAS is therefore not a great idea. However, this kind of activity, with well-prepared messaging and a clear value proposition, is a good lead-in to building a solid purchase funnel.

Question 7

What is the fake doors experiment?

Accepted Answer

The fake doors experiment (literally "fake doors") consists of presenting end users with a mock-up of a finished feature (e.g. a new function in an online store) that has not yet been built. The goal is to determine whether and what percentage of users would click that feature, which provides evidence that the feature is (or isn't) wanted on the site and should (or shouldn't) actually be developed.

Question 8

What is the difference between conversion and micro-conversion?

Accepted Answer

Conversion (the completion of a goal desired from the site owner's perspective) can be split by the level of business goals. The main e-commerce goal is the transaction, while a micro-conversion can be a newsletter signup, adding a product to the cart, or starting a chat with a consultant on the site. Measuring micro-conversions in e-commerce is justified for several reasons: 
 they help us understand customer behavior on the site better: they reflect what stage of the funnel users are at — people who already added a product to the cart are usually closer to a transaction than those who were on the site briefly without any interaction,
 they can simplify campaign optimization in Google or Facebook: when it's hard to reach the minimum conversion volume that helps ad systems optimize, it's worth using interactions that signal a higher likelihood of conversion and that can be collected in much greater numbers in a much shorter time.
we can better identify optimization opportunities on our store: measuring more than just the purchase helps answer the question of how users who bought differ from those who haven't yet.

Question 9

What is the purchase funnel for?

Accepted Answer

The purchase funnel (or sales funnel) concept lets us visualize the process a user goes through from first contact with the brand to making a purchase. For online stores, the funnel in its simplified form has four levels and visualizes the stages of customer acquisition:

1. Customer interest in the offering and brand awareness building
2. Convincing the customer to buy
3. Sale
4. Customer loyalty

To increase the effectiveness of your sales activities remember that optimization activities should run on every stage. A customer rarely decides on the first visit — before a transaction they compare products in other stores or consider a specific variant. The last stage — loyalty — should not be skipped either; acquiring a new customer is usually more expensive than encouraging an existing one to buy again.

Question 10

What is marketing automation and why is it important in e-commerce?

Accepted Answer

Marketing automation is the automation of marketing activities through dedicated software. The software can identify and monitor user actions and then optimize a range of marketing and sales processes. With marketing automation tools we can:

increase basket value (by presenting dynamic product recommendations or using cross-sell / up-sell mechanisms),
 reactivate dormant customers (sending information about new collections or promotions based on transactional data),
 recover abandoned carts (sending a reminder to complete the order).

When is it worth considering marketing automation? Check whether your online store has:
 the resources needed to execute the plan,
 a defined customer acquisition strategy,
 sufficient user volume visiting your site,
 a list of needs and goals to be reached with the tool.

When choosing marketing automation software you should account for stages such as:
 defining brand touchpoints across the customer journey that you can use to reach the customer more effectively,
 defining the goals that should be achieved with the marketing automation software — outcomes should be measurable (e.g. average basket value growth) and profitable (try to estimate implementation cost vs potential ROI),
 defining what data from IT systems can and should be synced with the tool: most marketing automation systems can aggregate data stored in CRM or ERP. It's worth understanding how synchronization will run, to fully tap the tool's potential.

Another key step is testing the system. Investing resources (people and money) in a complicated, clunky tool may produce the opposite of the intended effect: instead of automating sales and marketing processes, you complicate them. Before the final pick, validate tool usability, the quality of available training materials, and the level of customer support.

Ecommerce, Magento 2 and AI concepts explained