👩‍💻 How do I build a product finder (using Aiden)?

How do I build a product finder? You could write a book about it - which we did - or an article - which we also did - and you're reading that now.

Making a product finder is not difficult. Certainly not with the best software for making product finder=s. You can read how to get started with Aiden in five simple steps.

Step 1 - Set the goal

A product finder is a tool, not the goal. So before you get started, you need to determine what well the goal is. The purpose of your product finder consists of three parts: a who, a what and a why:

• whom are the people who want to help
• what are the products you advise on and
• why do you want to advise those people?

To make the goal concrete, you can use the following fill-in exercise:

With my decision help, [target group] can easily choose the right [product category] for [key metric].

For example: “With my decision aid, adults can easily choose the right mattress to increase the conversion of our webshop.”

Step 2 - Design the conversation

You now know who you are going to help, where and why. To achieve your goal, you need to understand the context of your target group and what all plays a role when the target group chooses and uses your product.

So the most important thing is that you put yourself in the customer's mind. The questions in your decision aid must be about their situation, wishes and (desired) use, not about your products. Because these kinds of questions are easy for people to answer and are relevant.

To do that, you will first listen. Because before you make an online consultation, it pays to experience an offline version. If you're lucky, you're an expert yourself and you've had these kinds of conversations before. If not, you can listen in with the employees in your own offline store, or those at the telephone customer service. To give this structure, it can help to work a mind map.

For example, this is what our mind map for a headset selection aid looks like:

Conversational design isn't just about asking good questions. It's just about the order of those questions and about keeping the user interested. A good online consultation creates a flow that takes people by the hand, interacts with them and lets go of them a lot wiser at the end. Just like a good offline conversation, in the store.

Have you come up with your good questions? Then consider the best order to set them. We use the example of an imaginary vacation choice aid:

• Start with a question that everyone can answer. This one should be accessible and fun. Are you more of a beach hanger or an active vacationer? '
• First, ask questions with few answer options so that customers get in the mood and so that their brain gets used to making choices. The question 'Where do you want to go on vacation? ' seems simple, but is quite complicated because there are many answer options. Instead, think about: 'Would you like to go on vacation outside Europe? '
• Create branches with in-depth questions that you don't want to bother everyone with. 'Do you have children?” If yes, then: 'Do you think it's important that there is entertainment for the children at the hotel? '
• Provide feedback and additional information while answering questions. 'Good to know: where you enjoy breakfast and dinner at half board, you also get lunch at full board. People like you often opt for half board.

The outcome of your question-and-answer game is product advice. With good advice, the customer understands which products are relevant to him and why. In other words: good advice is clear and traceable (and therefore not a filter on your category page without explaining why these products are recommended).

Step 3 - Match the products

The product database is the beating heart of every webshop. This contains all products and all available information about those products. For example, a product number, description, photo, price and inventory. But also color, weight, dimensions, material and warranty period. So comprehensive product information, but limited to “hard” specifications.

Such product specifications tell what a product is is, not what it is maketh. And buying a product is rarely about the product. Products are a means to an end.

To help customers properly, we therefore need to know which products match which answer (or combination of answers). In other words, who gets which product recommended when? We will arrange this with matching. Matching is linking the answers in your selection aid to the products in your catalog.

In other words, with matching, you create the rules that determine product advice.

In concrete terms, this means that you must indicate for each answer in your choice aid for each product to what extent the product is suitable for this answer. So: is this hiking shoe water resistant or not?

In order to really approach human advice with the advice logic (the advice that customers are used to in-store — and therefore the benchmark against which they test their online experience), it is important to give your decision aid nuanced advice. This means:

• Specifications are binary: true or false, on or off. But good advice isn't binary at all. A hiking boot is not waterproof or not waterproof. Some are splash-proof, others are truly all-weather and others are really sponges. You want to be able to balance and nuance good advice, which is why you need more than “true or false”. You want to be able to indicate whether a pair of hiking boots is “perfectly suitable”, “well suited”, “neutral” or “not suitable” for a walk with a risk of rain.

• The answers that a customer gives are either a requirement or a preferred: the product must fit well, or it would be nice if it fits well. For example, by setting a question as a filter (= requirement), you exclude headphones that do not have noise canceling. If you don't do this (= preference), all headphones will remain available for advice, but those without noise canceling will receive a lower advisory score for this question. If such headphones score highly on all other questions, these can still be the best headphones for that customer.

Products matched? Then you're ready to make your customers happy with the new decision aid.

Step 4 - Put it live

When you start building a decision aid, it is tempting, and dangerous, to want to create the perfect solution right away. One that leads all customers to their “best” in a minimum number of questions for all products. This is a fantastic goal, but too ambitious as a starting point. After all, it requires considerable effort, while there is a good chance that you will miss the point.

Resign yourself to the fact that the first version of your decision aid will never be the best. But also in the fact that the first version will immediately help customers better. In fact, this bar is now rather low. You're taking the step from “not helping people choose at all” to “already helping some people choose”. And that is possible for one product or product category. Or a specific target group. Or a subset of products. With your first version, you're going to make things better than they are now.

So break that barrier as quickly as possible and bring your first decision aid live.

By going live, you ensure that:

• you know what it takes to “complete the circle” (from conceiving to building and putting it live);
• you help people directly;
• you achieve results that generate more internal enthusiasm.

But where do you put your decision aid live?

Doubt can strike customers at various places and at times. So helping customers well means you want to be there in the moment that matters to achieve an uninterrupted customer journey. So place the same decision aid in multiple locations (and multiple forms of integration) in your webshop.

To determine those places, it is good to take a look back at the purpose of the decision aid. Who are you going to help again? With what? And when do they need that help?

For example, post your decision aid:

• As a pop-up on a lister page
• As a fold-out in a lister page
• As a pop-up on a product detail page
• As an attention-getter on a content page

Want to see real examples of real customers? Check this one seven inspiring examples.

Step 5 - Integrate it into your business

Decision aids can be so much more than “a fun tool to increase your conversion”. E-commerce is broken! No one is waiting for you products, but everyone is waiting for the solution to their problems. That's why decision aids work so wonderfully well.

Put decision aids first and take your entire e-business performance to the next level:

• Collect email addresses and let your customers subscribe to your newsletter immediately. For example, see Kees Smit Garden Furniture's decision aids for a live example.
• Get customer insights, for example to support marketing and purchasing.
• Personalize the customer journey by writing data to a CRM or CDP with the decision aid.
• Prevent dead inventory by listing products on the advice page boost.

Step 6 - Make it better

Yes, secretly also a Step 6, because although you can sit back and relax once your decision aid is live (especially if you use Aiden ⚡️ rules), we recommend setting aside some time for optimizations on a regular basis. What questions do people drop out on? And which one, on the other hand, is being responded to very well?

A good goal to set yourself is to make a change to the first version of your decision aid two weeks after the first go live. This doesn't have to be big, but it's important to complete the circle here too. You really want to make yourself aware of the fact that a decision aid is a living organism. And that changes have an impact on results.

And think a step further as well. You deliberately started small so that you could quickly help a limited group of people in certain cases with certain products, but of course you want to expand this. You want to help more and more people. And we will help you with that in turn.

‍