When setting up a touchpoint, you are asked to select a recommendation strategy. In order for you to choose the right one, it is very important you know what every single one does at a basic level. Following will be a list of all recommendation strategies, the way they work and tips for when to use them. It is also possible to use multiple strategies in a single touchpoint. You can find out more about this at the bottom of this article.

  • Let the algorithm decide

This option will allow the algorithm to decide the recommendation strategy for you. It does this by looking at all known data about both this specific visitor as well as any other previous customer journey to get the best possible result. This option is generally best used when you are not sure which selection is best. Having the algorithm decide is pretty much always better than wildly guessing by yourself.

  • Most relevant

When choosing this option the system will look at the visitor’s general profile (history, current url etc.) and suggest the most relevant content. This option is good when you want to simply promote the most relevant items based on the general information about the visitor.

  • Most popular

This option makes sure the profile sees the most popular items. This does not always mean these items are best fitted to this individual profile, but it takes some of the profile's preferences into account. It makes sure the popular items, still relevant to the profile, are sufficiently shown to remain popular. This is best used when you have some very popular products and want to go all-in on these already popular products.

  • Most relevant up-sells

By selecting this option, the system will make sure to promote up-sells as much as possible. This option is great when you sell multiple tiers of the same product. A good example here would be a software company that sells a basic, intermediate and pro version of software. The system will always look to push a higher tier.
Note: this recommendation strategy only works on content item pages, since you're up-selling related to the content item someone is looking at. Content item pages should also be registered correctly.

  • Most relevant cross-sells.

This option will make sure the visitor’s recommendations are based on the items he is currently interested in. Let’s say someone is looking at a mouse. It might be interesting to the visitor to also buy a new mousepad or even a keyboard. This is best used when you sell a lot of products in a similar category and want to make sure they buy products that go well together.
Note: this recommendation strategy only works on content item pages, since you're cross-selling related to the content item someone is looking at. Content item pages should also be registered correctly.

  • Most relevant cart-based

By selecting this option you're recommending products based on the customer’s shopping cart. It will look at the items that are currently in there and base it’s recommendations on that. You can use this option when you want to push items based on the current in-cart items, this is mostly used for recommendations on the shopping cart page itself.
Note: this recommendation strategy only works when the shopping cart script is correctly implemented on your website.

  • Most relevant based on last purchase

When this option is active the last purchase this profile has done will be heavily involved in deciding which items are recommended. This option is best used when the previous purchase heavily determines someone’s interest. An example could be a visitor being shown Apple products after buying an iPhone.
Note: this recommendation strategy only works when the conversion script is correctly implemented on your website.

  • Most relevant based on category

Selecting this option will make sure your visitor sees the most relevant product for them, within the category they are visiting. This is useful if you want to serve the most relevant content within the category the visitor is looking at.
Note: this recommendation strategy only works on category pages. These category pages should also be registered correctly.

  • Most popular based on category

This option is very similar to the previous one. The difference is that it will not take the visitor’s relevancy into account but instead simply recommends the most popular items within the category. This is very useful when you want to make sure the most popular items within a category keep getting pushed to visitors.
Note: this recommendation strategy only works on category pages. These category pages should also be registered correctly.

  • Last viewed

When this option is selected, the strategy will look at the items your visitor looked at last and simply display those items. This option can be used when you really want to push that one item your visitor was looking at.
Note: content item pages should be registered correctly for this to work.

  • Last purchased

This option is very similar to the last viewed option. The difference here is that it will not look at viewed items but instead purchased items. This option is very helpful when you have a lot of repurchases on your website and want to show the items profiles previously bought.
Note: this recommendation strategy only works when the conversion script is correctly implemented on your website.

  • Abandoned cart

While this option is active, the system will keep an eye on the visitor's shopping cart. If the visitor leaves the shopping session while there are still items in the shopping cart. These abandoned items will be promoted next time this visitor comes to your website. It's also possible to show them during the same session. This option is very helpful when there is a lot to do on your website (reading blogs, watching videos, etc) causing your customers to potentially forget about going to the checkout.
Note: this recommendation strategy only works when the shopping cart script and the conversion script are correctly implemented on your website.

  • Abandoned browser

This option is very similar to the "abandoned cart" option. The difference is that it is not required for the items to have been in the shopping cart. The customer has to have visited the items without making a purchase. This option can be used in a very similar sense to the abandoned cart option.
Note: this recommendation strategy only works when the content item view script and the conversion script are correctly implemented on your website.

  • Favored items

With this option you let the visitor do most of the work for you. They can decide which items they favor and the strategy will output these items. You can use this option when you want to put extra emphasis on the visitor' favored items.
Note: you need to send the favored items to our tracking script for this to work.

  • Based on user favorite items

This option is very similar to the previous one. The difference here is that the algorithm will not be pushing the favored items themselves but instead focus on items that might be interesting based on said favored item. This option is very helpful when you want to promote items based on favored items but don’t want to push the favored item itself.
Note: you need to send the favored items to our tracking script for this to work.

  • Based on user location

This recommendation strategy will look at the location of the customer and show the nearest content item. You need to add the longitude and latitude of the content item for this to work. This can be very helpful when location matters a lot. A good example here would be a hotel service (with multiple hotels in the country) recommending the closest hotel to the visitor’s current location.
Note: add three attributes to the content items ("longitude", "latitude" & "range") and add the longitude and latitude and range of the content item to it.

Using more than 1 strategy in a touchpoint

You may not always want to narrow down your touchpoint to just a single strategy. At the same time there might be some strategies you definitely don't want to use. You can manually select which (and how many) strategies you want to use in your touchpoint. The selector looks like:

In this example we have chosen to use 4 different strategies. The algorithm will attempt to use them starting at the top and working to the bottom. This means that it will first try to base its recommendations on the abandoned cart. If this proves unsuccessful it will try most relevant cart based. It will then try most relevant based on last purchase and finally most relevant. The system will work its way down until it has found one that is successful or until it runs out of options. You can change the order by dragging and dropping.

If you set up a touchpoint with 4 slots and there is just 1 item in the cart of the profile, it will continue with the next recommendation strategy and keep adding content items. So it could be the case that the touchpoint recommendation will contain 1 item of the profile's shopping cart, 1 relevant item based on this cart, 1 item recommended based on their last purchase, and 1 other relevant item for the profile.

Note: It is usually best to end with a strategy that is always applicable. This makes sure the touchpoint always shows something.

Did this answer your question?