Skip to content

What is a product recommendation quiz?

It is most often used where products are chosen by skin type, hair type, scent or shade rather than by size or price: skincare, haircare, fragrance and makeup. This page covers what the term means, how it works, how it differs from tools it is confused with, and when it fits.

Updated

Other names

The same thing under other names

Product finder, matching quiz, shop quiz, guided selling.

The same tool is also called a product finder, a product finder quiz, a product matching quiz, a shop quiz, or guided selling. The names vary by industry and platform. What they have in common: the shopper answers questions, and the answers lead to products from the shop’s own range.

How it works

Questions, sorting and a recommendation

Every product recommendation quiz works in three stages. What differs from one to the next is how each stage is designed.

  1. Questions

    The shopper answers a handful of questions, usually by tapping. A question is only useful if the answers point to differences between your products. The number of questions is a balance between being short enough to finish and giving enough information to sort the shopper.

  2. Sorting

    The combination of answers is matched to a type or profile defined in advance. The mapping from answers to types is best designed backwards from the product range. If the types and the products do not line up, the recommendation will miss.

  3. Recommendation

    The products that belong to that type are shown, with the reason. The reason is what separates a quiz from a list of suggestions. If the shopper cannot read why this product was chosen, they are as undecided as before.

Often confused with

How it differs from filters, search, chatbots and personality quizzes

These tools overlap, and a shop can use several. They answer different questions.

ToolWhat it starts fromWhat the shopper doesWhat comes back
Product recommendation quizThe shopper’s answers to a few questionsAnswers questions; does not need to know product termsOne or a few products from the catalogue, with the reason
FilterAttributes the shopper already understands, such as size, price or typeNarrows a list by choosing valuesA shorter list of products; the shopper still picks
Search boxA word the shopper typesNeeds to know what to call the productA list of products that match the words
ChatbotFree-form messages from the shopperTypes a request in their own words and waits for a replyDepends on the bot: an answer, a link or a handover to a person
Personality quizQuestions about the shopper as a personAnswers for fun or self-descriptionA label or result about the shopper; it may or may not lead to a product
Behavioural recommendation engineWhat the shopper viewed or bought, here or on similar shopsNothing; it works in the backgroundProducts shown as “you may also like”, with no stated reason
  • Quiz or engine: a quiz works on a first visit because it uses what the shopper says now. A behavioural engine needs history before it improves. They can run together.
  • Quiz or filter: a filter assumes the shopper knows which attribute matters. A quiz asks about the shopper instead and does the matching.

When it fits

It helps in some shops and not in others

Many similar products and unsure shoppers.

The same tool behaves differently depending on the range and the visitors. These are the points to check before adding one.

It tends to fit when

  • There are many products and shoppers are unsure which to choose.
  • Products are chosen by skin, hair, scent, shade or purpose, not just price.
  • A large share of visitors are first-time visitors.
  • You want to give the advice a good assistant gives in a shop, online.

It tends not to fit when

  • There are only a few products and little to choose between.
  • Shoppers choose almost entirely on price.
  • An equivalent route, such as a live adviser, already works well.
  • Visitors leave before they would answer a question at all.

Claims and wording

The words on a quiz screen count as claims

Quiz wording falls under the same claims rules.

In cosmetics and similar categories, there are rules on what a product may be said to do. Those can apply to the questions, the type names and the reasons on a quiz screen, as well as to product pages. One cautious design is to sort answers into types and recommend from the type, rather than promise an outcome for the shopper’s body. Wording is easier to settle at the design stage than to correct afterwards.

Not legal advice

This is a general description, not legal advice, and not a judgement on any particular wording. Check your own wording against the rules that apply to your shop.

How the Rekotto claims check works →

Measuring it

Decide what counts as a result before you add one

Starts, finishes, stops and clicks. Agree on orders first.

A quiz does not show whether it worked just by being installed. The basic measures are how many shoppers started, how many finished, which question they stopped at, and how many went from the result to a product page. Orders usually sit in the shop’s own cart and analytics, so agree how a quiz-assisted order will be counted before you start.

How Rekotto’s numbers are measured →

In Rekotto

How Rekotto does it

The one place on this page that names the product.

Rekotto is a Shopify app that runs this kind of quiz for skincare, fragrance, haircare and makeup shops. It sorts the shopper’s answers into a profile and recommends from your own catalogue only, with the reason attached. It does not promise outcomes.

The quiz is hosted for you, so there is no theme code to edit. You start from a ready-made question set, adjust it, preview it and publish. A word check flags risky terms and stops you publishing until you reword them. It does not clear a claim, and the wording stays your responsibility.

Available for UK and EU shops on the Shopify App Store.

Questions

Questions about product recommendation quizzes

What is a product recommendation quiz?

A short set of questions a shopper answers on a shop’s website. From the answers, the shop recommends one or a few products from its own catalogue, usually with the reason attached. It is meant for shoppers who do not know which product to choose.

Is a product finder the same thing?

Usually, yes. Product finder, product matching quiz, shop quiz and guided selling are used for the same kind of tool. The names differ by industry and platform.

How is it different from a filter?

A filter narrows a list by attributes the shopper already understands, such as price or size, and the shopper still chooses. A quiz asks about the shopper, does the matching and returns a recommendation with a reason.

How is it different from a search box?

Search needs the shopper to know what to type. A quiz needs the shopper to answer questions about themselves, so it works when they cannot name the product.

How is it different from a chatbot?

A chatbot takes free-form messages and answers however it is built to. A quiz has a fixed set of questions and a defined mapping from answers to products, so the same answers lead to the same recommendation.

How is it different from a personality quiz?

A personality quiz returns a label about the person, and may not lead to a product. A product recommendation quiz exists to end at a product from the shop’s own range.

How is it different from a behavioural recommendation engine?

An engine infers from what shoppers viewed or bought, so it improves as history builds up and gives no stated reason. A quiz uses what the shopper says now, so it works on a first visit. They can be used together.

How many products do I need for it to be worth it?

The count matters less than whether shoppers are unsure how to choose. With a few products that are easy to compare, a quiz adds a step. Where products are chosen by skin, hair, scent or shade, more products usually means more uncertainty.

How many questions should it have?

Enough to sort shoppers into types that match your products, and few enough that they finish. More questions allow finer sorting but lose more people along the way. Watch the question where shoppers stop, and adjust.

Can I use one for cosmetics?

Yes, but the wording on the quiz screens is subject to the same rules on claims as the rest of your shop. Sorting answers into types and recommending from the type, rather than promising an outcome, is one cautious design. This is not legal advice.

How do I measure whether it works?

Look at how many shoppers started, how many finished, where they stopped, and how many clicked through from the result to a product page. Orders usually sit in your own cart and analytics, so agree how to count quiz-assisted orders before you start.

Should I build one myself or use an app?

It depends on who designs the questions and maps answers to products, and who keeps them up to date. Building it is one job. Adjusting questions and changing products carries on afterwards. If you can own that upkeep, either route works. If you cannot, choose something that handles it with you.

Give every visitor an answer, on your own store.

Install it from the Shopify App Store, start from a ready-made question set for your shelf and preview it before it goes live. Billing runs through Shopify, and you can stop at the end of any month.