LOGIC1 recently became the only Google Premier Partner in New Zealand to qualify for the Google Shopping Innovation Awards APAC 2017 so we decided to do a series around how we did it.
Google Shopping – A New Era For Search
Google Shopping is a tough beast, it’s nothing like paid search in many aspects and can be similar to having a mini-website for each product you sell. You need to have good content and images while understanding which products you want to give priority to, for a variety of different scenarios. Optimisation is as much about creating hypotheses as it is about bidding.
While your shopping campaigns will never be perfect, you can still get ahead of the competition with some good foundations by creating a strong feed.
Product Titles & Descriptions
The main obstacle with Shopping campaigns is that you cannot target campaigns via keywords or match type. The keyword queries are connected to your titles and descriptions which means getting those right is crucial in allowing the Google Shopping Algorithm to understand what searches your products should appear for.
The above is a good example, as the query easily triggered the products with the right titles. BUT, I hear you say, people search for many variations of a product – not just name! That is correct, so we would recommend overusing keywords rather than under-using them. What do I mean by that? Well, add in colour, size, type or any other product differentiators – remember the customer cannot touch or feel the product so extra information can help them make that purchasing decision.
Images, as you can imagine for an image based advertising solution, are very important and can be the difference between a user believing your shop to be reputable or not. If you don’t have high-resolution images you might as well not even bother. The example below shows that the store went to a lot of effort to ensure that the title was correct – including size in the title for specific searches but is then let down by a poorly optimised image.
Images though are not just about the resolution. The style, lighting and how they are positioned can make a perfectly targeted ad become practically useless.
Ad Groups vs Product Groups
Too many product groups create a nightmare scenario when you are trying to optimise. The setup of Google Shopping is clunky at best, so having as many campaigns and ad groups as possible allows you more control over bidding as well as budget. Not only that, you can add negative keywords on ad group level which means you can filter search queries towards unique products.
It will take time but creating a well thought our structure will save you hours in the long run as well as driving much better revenue.
As is the case with Google Shopping, you tend to use the same product in a variety of ad groups and campaigns within your account. It is important then to use campaign priority settings to determine which campaign should bid.
For example, if two campaigns share the same product and one campaign has a higher priority versus a medium one then the Google will use the bid from the higher priority campaign first even if the medium campaign has a higher bid. If the higher priority campaign then runs out of budget, the next in line, priority wise, will enter the auction. If the same priority is used for all campaigns then the highest bidder will be used.
The reason you use priority is to filter keywords and products towards specific product groups. So, if a product is on sale or has a better margin but the keyword that triggers the ad, can also trigger a number of similar products.
Negatives
Hopefully from this post you will have realised that control over the account is paramount for success – be it from a bidding perspective or by filtering the best products to show against particular searches. The ability to influence Shopping campaigns is best shown through the use of negatives.
For example, Rossignol have a variation of products – you have generic brand searches, then more specific queries that include the type of product and then finally precise searches for models. These can represent various stages within the users purchase path and you therefore would want to bid higher for certain ad groups.
The above structure should allow you to understand what a user is looking for. The users triggering an ad from 1.0 Rossignol may only be starting their search or looking for a local stockist. In 2.0, we know that the user is looking for something specific and wants to see the latest lines from the brand so they are closer to purchasing while 3.0 could be in the same boat but as they know the name of the model we believe these to be more valuable and bid for those users accordingly.
Summary
Google Shopping is a lot different to any other Google Ads solution and it takes time to master it effectively. It is a fantastic way to sell online but it can be very unforgiving and will chew up your budget if you don’t know how to properly structure your entire account. It’s not unheard of for clients to approach us after the fact, having spent thousands of dollars with minimal return.
As the only NZ Shopping Innovation finalist for APAC, we know how to create excellent feeds, hypothesis and optimisation to drive better revenue for our clients.
