Product & answer weightings - how it works

The Help Me Choose app uses Shopify tags to recommend relevant products to your customers based on their answers to the quiz.

In order for it to achieve this, firstly you must have Shopify tags assigned to your products (you can find out how to add tags to your products in your Shopify store here). Then, when you build the quiz you simply input the tags that are assigned to products that relate to a specific answer. You then tell us how important this answer is in relation to the overall recommendations by giving it a weighting between 1 and 100. The Help Me Choose app uses a simple addition logic to provide recommendations so at the end of the quiz, products that have the highest score against their tags will be shown to the user.

Weighting Strategies

Before you start adding weightings to your answers it's worth having a think about your weighting strategy. There are two main strategies our team of expert quiz builders generally tend to employ during our custom build projects; Equal Weighting Strategy and Priority Weighting Strategy.

Equal Weighting Strategy - This is where the answers to every single question are all given an equal weighting. This strategy is used where no specific answer is more important than any other. This strategy is quick and simple to configure but the recommendations provided can be slightly less specific in some instances.

Priority Weighting Strategy - This is where higher weightings are assigned to certain answers and lower weightings are assigned to others. If the answers to one or more questions are more important than others then simply give them a higher weighting, and products with the relevant tags will have a higher score than others and will appear in the recommendations.

For example, imagine you are the owner of a Shopify store that sells lots of different types of handbag and you are using the Help Me Choose app to recommend them to customers. You want to ask what style and then what colour handbags the customer likes in order to recommend a range of perfect handbags. From your sales experience you know that style is far more important that colour preference.

So first you ask “What style handbag do you prefer? with the answers “Tote/Crossbody/Shoulder/Clutch”. Lets assume all of your Tote bags have the tag “tote” assigned to them in your store. So when you add this answer to the Help Me Choose quiz builder you select “tote” from the tags field for this answer and as style is the most important question you are asking (as you want to only recommend tote bags to this particular customer as they have told you that’s what they like) then you would give it a high weighting like 100 for example.

For the colour question, which is not as an important question as style, you would follow the same process, select the tags you have added to your products that relate to colour (blue handbags have the tag “blue” for example) and give these answers a lower weighting such as 75.

At the end of the quiz, the products with the highest score from the weightings against all the tags which are assigned to them will be recommended. In this example all recommended products would be Tote style handbags and if there is enough blue tote handbags then they would all be blue but if not then another colour tote handbag would be shown. You can use any number for the weightings as long as the more important questions have a higher weighting than others. For example you could use a scale of 1-10 if you prefer that to 1-100.

Default Recommendations

At the bottom of the survey page on the quiz builder there is a field where you can add default products for the quiz. This is for when a customer gives a series of answers where you don’t have any products (or enough products for the number of recommended products you have chosen to display) that match exactly what they’re looking for. In this instance we recommend adding some of your best selling items as a default. You can add these products here by clicking on the + add products button and selecting them from your product feed.

Still looking for answers? Get in touch with the team on