We hear you! There are many misconceptions for you to deduct: You don’t have the time and resources, site search is not where you want to go.
- Your catalog is incomplete – Youâve heard of GIGO (âGarbage In Garbage Outâ) – so whatâs the point?
- You are understaffed – You do not have additional resources to dedicate to manage search on your eCommerce storeÂ
- This requires massive IT resources – which means youâll have to give up on other projects to ensure your search is converting more and continuously
Through this blog, we would walk through these concerns and try to dissect what has changed, where does the machine âkick inâ and how should you look at the ROI (Both in revenue and man-hours) of such a project.
Site Search has evolved. And it is going to keep evolving. We have designed the new Unbxd self onboarding product to be as seamless as you onboard your new mobile device. By sharing your catalog and your analytics, we can immediately analyze and predict what would be the impacted lift simply by turning our AI-powered search ON for your online store, and this is only the beginning. We at Unbxd are bringing The Future of Commerce Search to you with as much ease, simplicity and sophistication!
Itâs all about automation.
Just like in construction, every good building starts with a solid foundation. We start with your existing product feed.Â
- The first objection we hear from our customers is – âI know my data is incompleteâ. In an ideal world, we would all like to have every piece of metadata available for our shoppers to search on. Understanding this limitation, we begin our journey by mining what you DO HAVE. Our ML process runs through your catalog and looks for coverage (i.e what fields have the most data). These insights will allow us to fix and fine-tune relevance and give priority to fields that do have data in them.Â
- Second, we can now go ahead and mine how your customers are searching for your catalog. The aggregated data of your last yearâs worth of query, allows our ML to understand a few things
- Keyword Variation – Word break/Word Joint – Unbxd Spell Check ML identifies when to split and join the keywords using its Word Break algorithms, then generate the list of related keywords and spell checks to rectify them.Â
- Example:Â
- You sell Jackets with annual revenue of $500,000:
- Query volume is 1,000,000Â
- The conversion rate of 10%Â
- AOV of $5Â
- Unbxd realizes you had many more searches, for the same product, different way, with color! âBlue jacketsâ – revenue @ $75,000
- Query Volume – 300,000
- Conversion rate – 5%
- AOV – $5
- You sell Jackets with annual revenue of $500,000:
- Example:Â
- Keyword Variation – Word break/Word Joint – Unbxd Spell Check ML identifies when to split and join the keywords using its Word Break algorithms, then generate the list of related keywords and spell checks to rectify them.Â
2. Query Clustering – Longtail is your oldest nemesis, even if you were able to control the âheadâ or your queries, customers by default will always find another way to describe the same product.Â
How do you rank the products correctly if you do not have enough data? Unbxd identifies similar queries using its string matching algorithm. Among these similar queries, the head queries are identified and the scores of the head query are attributed to similar queries. For example, see the image below, as âtopsâ comes in different shapes and sized, search can utilize all interactions in order to rank sub interactions
This is just the tip of the iceberg. The search queries vary in their type, syntax, frequency, seasonality, recency, and even the way in which each shopper searches in the native language. Our AI and ML models ensure that each query is well construed and understood to derive the most contextual reference and conclusion out of the search query. Only to ensure the most relevant results for the shoppers.
Itâs all about Relevance – Whatâs in a query?
The picture above is from conceptNet, an open-source project to understand concepts for ML. While you can see there are many related concepts we can derive, there are limitations. It is up to our ML to then decide on which are truly related based on YOUR relevance. Letâs take our customers as an example:
1. A search for tee produces the following 560 results, where we are showing Tees, tanks & t-shirts
2. A Search for tshirt produces the same number of results, but we can see the product ranking is different
There are three things happening here: First, we modified the âtshirtâ queryâ to match T-shirt via our extensive spell check. Second, through contextual understanding, we have mapped tee, tank & t-shirt as a similar product. Third, we now rank the products differently on a QUERY level based on your clickstream data.Â
Â
There is a fourth dimension to this, Personalization. As we power multiple workloads across your site, we also take a look at every moment on what would we recommend for you. The intersection between our recommendations and relevancy will determine what we would show your consumer as top products.
Note, up until this point, there was no business user intervention required. We have automated for you the following tasks:
- We mined your catalog to understand what coverage of data is available
- We mined your usersâ query log for the last year to understand if that coverage matches their intent
- We enriched your catalog with synonyms
- We have defined the importance of the relevance of your fields
- We will autocorrect 90% of your customerâs misspellings (and do it up to six worlds at a time) moving forward
- We will solve for any stemming, lemmatization in queries moving forward as well.
Once all of this will be done by our AI and ML engine, you will implement our analytics, and we can start doing the following:
- We will identify trends and adjust our auto-suggest accordingly
- We will automatically rank every search result and category page based on product performance
- Our recommendations widgets will be able to provide a personalized recommendation as well as âviewed also viewedâ and âbought also boughtâ strategies
From reading and interpreting your catalog to fine-tuning the relevance and ranking the most relevant products higher on the product listing pages and the category pages, we ensure that the end-to-end product discovery journey is automated for an eCommerce store and a delightful and seamless experience for end shoppers!
To put impact visually, letâs take a look at the average query classification we have seen in our customers before and after implementing the Unbxd solution!
On the left, is where you are Today. Moving to the right, not only automation kicks in, but also a massive reduction in the work you, and your team, will have to put in:
- We will recognize and return about 86% of the queries (vs 48% before) by classifying Queries correctly to the products and attributes
- An immediate impact will be a sharp drop in zero results query – our customers usually see about a 75% decrease in null results
- Another impact would then be your team would no longer to manage massive redirects effort – We usually see a 60% drop there
- As our spell check is super advanced – we can significantly minimize that number as well, saving your team from a vicious cycle of synonyms.Â
With that, all that is left is to deploy unbxd to your site. We provide a full set of APIs, SDKs and services offerings to get you ramped and live for holidays. We have onboarded customers in as fast as 21 days.
Not to mention, If you are on Magento or Hybris, click deploy, you’re already there!!!Â
Contact sales@unbxd.com if this is of interest. Once youâll see what self onboarding can do for you, this will be a no brainer.
