 The pinning of certain products is a useful tool when it comes to marketing and promotion. Now, Findify has made this process even smoother by introducing both an in-pinning search as well as a bulk actions capability - enabling merchants to pin (or unpin) several products at once.  Thanks to it’s advanced AI-powered personalization software, Findify is widely regarded as the best ecommerce tool when it comes to conversion optimization. It’s solutions learn what the shopper is most likely to buy, and reorders product feeds based on each individual’s wants and needs. While this method successfully improves conversion in ecommerce, sometimes additional tactics are required - in order to drive brand collaborations, sell out certain products, or promote a special event, for example. That’s where pinning comes in. Clients can access pinning within their merchant dashboard - go to either Smart Collections or Search, and then select Pinning. Here, merchants can select the products they want to pin, ensuring that their latest sale, newest line, or seasonal offer is always being shown to customers on their site. This function allows for increased visibility of whatever is pinned, without sacrificing the power of personalization - the personalized search results appearing below the pinned product or collection still uses advanced AI and machine-learning algorithms to reorder products based on the preferences of the individual customer. Now, Findify developers have made the pinning process even easier, thanks to the addition of a number of new functions. Pinned items are now searchable, lists of pinned items are now collapsible, and there is an added capability to pin items in bulk. “On the surface these may seem like small additions, but these tweaks will really speed up the pinning process for those using this functionality, making it more convenient for merchants,” explained Findify’s Head of Product Vsevolod Goloviznin.
“We have a lot of merchants who regularly use pinning. Over time, their list of pinned items can become quite long and have in the past become a little difficult for them to navigate - which is why now we have made this list searchable. In order to help these merchants further, we have made the list of pinned items collapsible - eliminating the need to scroll down through the pinned items to get to the non-pinned items. Furthermore, bulk actions will help merchants speed up the pinning process when needing to pin, or unpin, multiple items in one go.” For more information on Findify’s powerful ecommerce tools, including personalization software and solutions such as Personalized Search, Smart Collections, and Recommendations, book a demo here.  In our quest for dashboard control, we have now added the capability to exclude IP addresses - enabling cleaner data and more accurate insights. As part of Findify’s ongoing quest to make merchant lives easier, a new IP Exclusion capability has been released. Clients now have the ability to dictate which IP addresses to exclude from their Findify analytics and personalization data. While on the surface this might seem like a small tweak to the merchant dashboard, it is a valuable one, explained Findify’s Head of Product Vsevolod Goloviznin. “There are many reasons why a client would want to exclude certain IP addresses, but most commonly it is because their employees are visiting the web store a lot to check the products. This becomes a lot of web traffic, and so the behaviour can end up affecting the personalization algorithm as well as slightly skewing the analytics merchants rely on to make data-driven merchandising decisions,” he said. “Up to now, merchants who wanted to exclude IP addresses simply emailed us and we completed this task on their behalf. At Findify, however, we believe in empowering our clients which is why our developers have worked on giving them the freedom to oversee this capability themselves.” Mr Goloviznin revealed the move is part of a bigger plan to give merchants more control within the Findify merchant dashboard, and the ability to tailor Findify solutions even further to their own specific needs. Working with this new feature Clients only have to log in to their Findify merchant dashboard, then navigate first to ‘Settings’ then to ‘Advanced Setup’. Here you will find ‘IP Blocking’. In seconds, clients can easily add as many IP addresses as they wish.
For more information on Findify’s powerful ecommerce tools, including personalization software and solutions such as Personalized Search, Smart Collections, and Recommendations, book a demo here.  Findify’s new Seamless Shopify Collections feature improves the way Shopify merchants work with their Findify collections. Going forward, merchants can manage all their Findify collections directly from the Shopify dashboard - making the process quicker and easier than ever before.Smart Collections continues to be one of Findify’s most popular solutions. Merchants using this technology enjoy collections that are dynamic, personalized, employ unlimited filter capabilities, and perform far better than those created using the static displays offered as a default by many ecommerce platforms. While merchants all agree that Findify Smart Collections is incredibly valuable to them and to their ecommerce business, a recurring request was for Findify to create a more streamlined way for Shopify merchants to leverage their existing collections. Listening to merchants and acting on their feedback has always been a core value at Findify, which is what led developers to create the new Seamless Shopify Collections feature - enabling merchants to work with their collections directly from the Shopify dashboard. “Up to now, merchants have been working with their collections both within Shopify and within Findify. While there have been capabilities for migrating and syncing collections between the two dashboards, our quest for simplicity drove us towards fully streamlining this process,” said Findify’s Head of Innovation Vsevolod Goloviznin. Learn how this new feature works by watching the Findify Explainer Video below, or on YouTube here. Working with Collections now “easier than ever” according to Findify CEOCEO of Findify, Philip Nobel, praised his company’s developers and said it was a key part of the roadmap to get Seamless Shopify Collections rolled out. “This is a very important feature for us, and for our clients,” he said. “Working with Collections is now easier than ever before for our Shopify merchants, with this feature effectively freeing up their valuable time.” Be among the first to try this new feature The Seamless Shopify Collections feature has progressed through two rounds of beta testing so far. A third and final round of beta testing is commencing shortly.
Any Findify clients operating on the Shopify platform who would like to be involved in this third phase of beta testing can register their interest by emailing yourfriends@findify.io. For more information on Findify’s powerful ecommerce tool, which includes personalization software and solutions such as Personalized Search, Smart Collections, and Recommendations, book a demo here.  A number of innovative Findify product developments are set to go live in the New Year. Two of these can already be described as gamechangers: Findify’s new Active Sync feature, and the rollout of Findify’s Seamless Shopify Collections.A major new product development, Findify’s soon-to-be-released Active Sync will increase the speed of Findify’s ecommerce assortment synchronization. The feature, due for release in early 2021, is set to become the cornerstone of Findify’s brand new infrastructure - streamlining internal processes and paving the way for additional growth and innovation, and paving the way for Findify’s Content Search solution. “This is all part of our quest to make merchant life as easy as possible and represents a fundamental change in the way we handle huge volumes of data,” revealed Findify CCO, Joakim Amadeus Olsson. “With Active Sync specifically, merchants can rest assured knowing that their product feed - including prices and stock levels - is always up-to-date. Shoppers will never see stale results, only the most accurate and current assortment data. This will be crucial for high pace sales.” “None of our direct competitors offer anything as fast as Active Sync”Findify CEO, Philip Nobel, said he was delighted to get this feature over the line before the end of the year. “Active Sync is an amazing new feature - one which none of our direct competitors have, and which will create great merchant value. Our merchants asked for this, and while we had to rebuild the entire infrastructure to do it, I’m very happy to now have met this need for them. At Findify, we think there is nothing more important than listening to the wants and needs our clients,” he said. “This was a major part of our product roadmap for 2020 and our developers worked really hard to get this done. I’m incredibly proud of them and of what they’ve achieved. It was a big project, requiring an entire stack rebuild, but it is something that will benefit our clients greatly - which means it was absolutely worth it.” Next Steps: Platform Connectors When Findify’s Active Sync is rolled out, it will be immediately available to any Findify client with developer capacity to connect their store to the API. Merchants without developer capacity will be able to benefit from the new feature via Findify-created platform connectors. “In tandem with Active Sync, we have been working on creating a Shopify Active Sync Connector”, revealed Findify’s Head of Innovation Vsevolod Goloviznin. “These two feature will be rolled out at the same time, and further down the line we will look into creating connectors for additional platforms.” For more information on Findify’s powerful ecommerce tool, which includes personalization software and solutions such as Personalized Search, Smart Collections, and Recommendations, book a demo here. To read a preview of Findify’s soon-to-be-released Seamless Shopify Collections, click here.  A number of innovative Findify product developments are set to go live in the New Year. Two of these can already be described as gamechangers: Findify’s new Active Sync feature, and the rollout of Findify’s Seamless Shopify Collections.Smart Collections is one of the most popular Findify solutions - and with good reason. Merchants using this technology enjoy collections that are dynamic, personalized, employ unlimited filter capabilities, and perform far better than those created using the static displays offered as a default by ecommerce platforms. While merchants all agree the solution is incredibly valuable to them and to their ecommerce business, a recurring request was for Findify to create a more streamlined way for Shopify merchants to leverage their existing collections. Listening to merchants and acting on their feedback has always been a core value at Findify, which is what led developers to create the soon-to-be-released Seamless Shopify Collections feature. Coming in early 2021, this feature allows merchants to work with their collections in one place and one place only - within Shopify. “Up to now, merchants have been working with their collections both within Shopify and within Findify. While there are capabilities for migrating and syncing collections between the two dashboards, our quest for simplicity drove us towards fully streamlining this process,” said Findify’s Head of Innovation Vsevolod Goloviznin. “It wasn’t a quick fix, but our developers devoted a lot of time to perfecting this process - and now our new, fully automated “set and forget” method is set to be rolled out very soon, which is great.” Working with Collections now “easier than ever” according to Findify CEOCEO of Findify, Philip Nobel, praised his company’s developers and said it was a key part of the 2020 roadmap to get Seamless Shopify Collections rolled out. “This is a very important feature for us, and for our merchants, and I’m delighted to see it nearing completion,” he said. “Working with Collections is now easier than ever before for our Shopify merchants, with this feature effectively freeing up their valuable time.” Be the first to try this new featureFindify is currently accepting Seamless Smart Collections Beta Testers - a handful of Findify clients operating on the Shopify platform willing to start using this feature in advance of the official launch.
Anyone who would like to Beta Test this feature can register their interest by emailing yourfriends@findify.io. For more information on Findify’s powerful ecommerce tool, which includes personalization software and solutions such as Personalized Search, Smart Collections, and Recommendations, book a demo here. To read a preview of Findify’s soon-to-be-released Active Sync feature, click here.  A fresh redesign of the Findify Merchant Dashboard is set to streamline navigation, increase efficiency, and improve user experience for Findify clients. Delivering a positive user experience is at the very heart of Findify’s business model. Usually, this means delivering a top notch UX for the customers of our clients - online shoppers who interact with our Site Search and Personalization software when they search, browse collections, or view recommendations on their web stores. But the UX of our clients themselves is also high on Findify’s list of priorities, which is why developers at the company have launched a brand new Findify Merchant Dashboard. The new and improved interface sees a number of changes:
 Enabling the new Findify Merchant DashboardTo check out the brand new interface, simply log in to your Findify Merchant Dashboard. At the top of the page, you will see a blue button that says “Switch to New Dashboard”. When you click this, you will now have the new interface. Between now and January 20, 2021, all Findify clients will have the ability to toggle back and forth between the current and the new dashboards, giving them a chance to test out the new interface, and get used to the new look and feel. After this date, the Beta period comes to an end and the new interface will be enabled for all Findify clients on a permanent basis. Positive feedback “We have already had fantastic feedback from clients about the new interface - a lot of merchants have been playing around with this new version of the dashboard and say they really love it,” said Findify’s Vsevolod Goloviznin.
“That is great to hear, considering the aim of the project was to improve user experience by designing a fresh new look and feel, by streamlining the navigation, and by overhauling a few other areas also. For any client who has yet to enable the new interface, I would encourage you to do so.” Any feedback on Findify’s new Merchant Dashboard can be submitted to yourfriends@findify.io. For more information on Findify’s powerful ecommerce tools, including personalization software and solutions such as Personalized Search, Smart Collections, and Recommendations, book a demo here. 
This brand new feature boosts conversion rates for ecommerce merchants by improving UX for shoppers who conduct searches for makes, models, and SKU numbers.
The latest feature released by Findify developers sees the leading Site Search and Personalization software company improve its Personalized Search function even further - this time with improved model search functionality to help shoppers quickly and easily find (and buy!) the products they are looking for.
What is model search, and who uses it?
A model search, also referred to as a SKU search, takes place when a shopper types a product number/product code into the search bar of an ecommerce store.
This search could be written in the form of ‘Product + Model’ along the lines of ‘Iphone 11’, ‘Sony A6000’, or ‘Fujifilm X-T3’, or it could take the more ambiguous form of something like ‘rt9027483’. Merchants most likely to see customers searching using SKU-style parameters are largely those selling cameras, laptops, parts, mobiles, gaming equipment, automotives, and home electronics. 
An example of model search in action for Findify client Cyberphoto. The search term ‘nikon d5’ now includes Nikon D500 and D5000 - a full range not available through purely tokenized engines.
How hard is it to build an effective search solution for SKUs?
While getting this type of search right is extremely difficult, it’s well worth the effort. Those searching for products in this way usually know exactly the make and model they want to buy, or at least the exact product range they want, meaning they are shoppers with a very high purchase intent and are, therefore, particularly valuable visitors.
The issue, however, is that a model search is notoriously hard to perfect. “For normal, word-based searches, we’ve built a system, based on Tokenization, where shoppers don’t need to type out the entire word, or to even type their query correctly, and they will still get relevant results,” explained Findify’s Seva Goloviznin. “But model search/SKU searches need to be treated differently. While tokenization is great for word-based searches, there are limitations to this method when applied to model search and SKU search. Up until now, this has meant that shoppers who wanted to search using an SKU had to type their query in perfectly to get the product they wanted to find.” Of course, it isn’t always possible for shoppers to input the SKU exactly right, and merchants shouldn’t expect them to. Perhaps the shopper doesn’t know the full SKU, or they don’t realize there’s a hyphen or space in it… or they simply don’t want to spend time writing the full string. Let’s say, for example, there is a product called an ‘X-1/1’. Shoppers can, and will, search for this in the following ways ‘X-1/1’, ‘X1/1’, ‘X11’, ‘X1’, ‘X-11’, ‘11’, ‘X’. If these searches don’t yield relevant search results, the shopper is at a high risk of churning and taking their business elsewhere, assuming the store doesn’t have what they’re looking for. In other cases, more related to discovery, some shoppers might just want to see all the products in a certain range, so they want to type ‘T5’ and view not only the Lenovo ThinkPad T560 but also the T550 and the T540p.  So what’s the solution?
Up until now, the best solution has been to tokenize the search results. So, in accordance with our previous example, ‘T’ becomes one token, and ‘560’ becomes another token. Longer SKUs would also have subtokens. When a user enters a query, a collection of documents containing these tokens are searched for the best matches.
While this system can work well, it has limitations. Firstly, it is a huge undertaking to create so many tokens, and secondly it creates a large amount of data for our system to process, meaning it will likely take a slightly longer time for search results to populate. Other ecommerce site search providers have attempted to solve this problem with prefix search, however Findify developers quickly realized this was just switching one set of problems for another - this “fix” decreases both precision as well as the overall site experience. This is why Findify developers have implemented a new system, utilizing border-ngrams. This method means there is no need to create an endless amount of tokens, and also means searches take less time as there is less data to screen. The icing on the cake is that Findify’s border-ngram method also makes model search results a lot more relevant. 
With Findify’s model search, queries for ‘microsd’ yield a full range of results, including close ranges of products like microsdx, microsdxc etc.
What is a border-ngram and how is it used?
Put simply, an ngram is a type of probabilistic language method for predicting the next item in a sequence. However a full ngram generation may quickly explode index size, slowing the search down.
A border-ngram, therefore, is a compromise between sensitivity and index size, ensuring relevant results are returned without generating a huge amount of data and adversely impacting speed. Uniquely utilizing this method, Findify’s now improved search solution ensures that whenever a shopper types a search query that contains both letters and numbers, the system will recognize this as a potential model search, activating the border-ngram. This ensures that customers opting for a SKU search will always find exactly what they’re looking for, regardless of misspellings, benefitting both the shopper and the merchant alike. For more information on Findify’s powerful ecommerce tools, including personalization software and solutions such as Personalized Search, Smart Collections, and Recommendations, book a demo here. 
Certain Recommendations widgets require user data to populate. As such, it’s important for merchants to have the ability to implement a fallback mechanism, in the eventuality that the data for their preferred widget is not yet generated. Now, Findify has made this process even easier.
A Recommendations solution is a powerful ecommerce tool that helps merchants drive incremental revenues and higher margins, allowing them to maximize their online store’s upsell and cross-sell capabilities.
And at the core of every great Recommendations suite is the ability to tailor - but without losing the ability to get the Recommendations widget in place quickly and easily. Findify’s Recommendations solution includes a number of widgets, including ‘Hot Sellers’, ‘Recently Viewed Items’, ‘Newest Products’, ‘Frequently Purchased Together’, ‘Customers Who Viewed This Also Viewed’, and ‘Customers Who Viewed This Ultimately Bought’. Now this offering has been improved even further, with the addition of an automated Recommendations Fallback Mechanism. 
A Findify Recommendations widget in action on a product page of Findify client United by Blue.
What is a Recommendations Fallback mechanism?
The aforementioned categories of Recommendations widgets can be split into two main groups - those that require user data to populate, and those that do not.
Take Newest Products, for example. This widget can be applied without personalization and will then be populated based on the parameters the retailer specified, and not on user behaviour. As such, the same products will be shown to all shoppers. If we then look at the Recently Viewed Items widget, this is a good example of the other type, which requires user data to populate. This widget retains information about the current and previous sessions of the individual user, and populates the section with items that that specific shopper has already viewed. So what happens if there is no data for the widget to draw upon? Perhaps it is the shopper’s very first time on the site, and hasn’t viewed any items at all yet. Or perhaps the product itself is new, and there is no data to populate the Users Who Viewed This Ultimately Bought widget, because nobody has yet viewed that product and gone on to make a purchase. In this scenario, it is important that there is a fallback mechanism in place, so that the space occupied by the widget does not lay idle, potentially representing a missed selling opportunity. Previously, merchants using Findify could implement a fallback mechanism directly in the frontend for this eventuality. Now, they don’t have to. They can simply select a fallback option within the merchant dashboard. 
A Recommendations Fallback can be set in seconds within the Findify Merchant Dashboard.
“This is an improvement we are very happy to roll out, and one I think will really help out clients. A fallback mechanism is especially valuable for tailored widgets and, in our quest for merchant simplicity, this addition will make life even easier for our clients to quickly select their preferred fallback option,” explained Findify’s Vsevolod Goloviznin.
“In our opinion, the best fallback mechanism is usually Hot Sellers, but of course merchants can make that call for themselves very easily now within our dashboard. The whole thing only takes a few clicks now, which is great.” Discover more how Findify can improve your salesRequest a demoWithin the analytics side of Findify’s merchant dashboard, retailers can view how the widget is performing, but also how the fallback option is performing, and data like how many times the fallback was rendered. “We would advise our clients to check their analytics regularly. Yes, it requires setting aside a portion of time each week or month to go through the stats, but it’s such valuable data that’s being generated. It’s a great way for merchants to gain some truly actionable insights. For more information on Findify’s powerful ecommerce tools, including personalization software and solutions such as Personalized Search, Smart Collections, and Recommendations, book a demo here. 
New one-way synonym capabilities are set to make shoppers’ search results even more relevant - helping merchants increase revenue and reduce churn.
In the world of ecommerce, site search is everything. An effective search functionality is often the only difference between a sale and a lost customer - if shoppers can’t find what they’re looking for, they don’t have the opportunity to buy it.
One aspect of a truly hard-working site search lies in the engine’s synonyms capability. Two-way synonyms are good...
Take, for example, a furniture store. A shopper goes to the store’s website and searches for ‘couches’... and no couches are shown. Why? Because the store lists their couches as ‘sofas’.
Likewise for electronic stores and ‘multi-function’ vs ‘all-in-one’ printers, and retail stores battling with ‘pants’ vs ‘trousers’ and ‘t-shirts’ vs ‘tees’.
This functionality may seem simplistic, but a recent study shows the site search of more than 70% of online stores don’t feature proper synonym capabilities. Findify, a leading provider of site search and personalization solutions, includes synonym capabilities at the core of it’s search AI-powered engine, and has newly improved this offering to go one step further with the introduction of one-way synonyms. An example of a regular synonym, also called a two-way synonym, means that ‘pants’ = ‘trousers’ and, therefore, ‘trousers’ = ‘pants’. A search for either query will always return the other also. While this is essential and is, apparently, more than what 61% of ecommerce stores currently feature, the next level of this is a one-way synonym for use in situations where term one equals term two… but term two doesn’t equal term one. ...but one-way synonyms are even greater
In an electronics store, for example, someone searching for ‘phone’ should also see results like ‘Google Pixel’ and ‘Samsung S10’. However, someone searching for ‘Samsung S10’ does not want to see phones of other makes and models.
Likewise, in fashion, a search for ‘summer clothes’ should return results like ‘dresses’, ‘skirts’, and ‘t-shirts’, however you don’t want a search for ‘dresses’ to return skirts or t-shirts, or any results that are not actually dresses. In general, one-way synonyms are usually used on more generic terms in order for shoppers to find items that have been given quite specific titles and descriptions. This means search results are more relevant for the customer who, if finding what they were looking for, is more likely to make a purchase. Managing synonyms within the Findify dashboard
Merchants can control their site’s new and improved synonym capabilities via the Findify Merchant Dashboard.
First, click on ‘Search’. Then, included in the menu on the left, you will see both ‘One-way synonyms’ and ‘Two-way synonyms’. In the field provided, write what word you would like to create a synonym for, and then write the desired synonym in the corresponding field. Hit ‘Save’ and your synonym will appear in the below list of ‘Synonym Groups’. While the synonym is now created, merchants need to push the changes live on their site. This is easily done by migrating back to the homepage of the Findify Merchant Dashboard, and then clicking on ‘Sync, Status & Usage’. Then, in the left-hand-side menu, select ‘Product Sync’ and hit the blue ‘Sync Products’ button.
For more information on Findify’s powerful ecommerce tools, including personalization software and solutions such as Personalized Search, Smart Collections, and Recommendations, book a demo here.
Read more news about our productExperience the Findify effect
A new visual preview function for ecommerce Recommendations is set to improve processes for Findify clients and help them create effective upsell and cross-sell opportunities quicker and easier.
Findify’s popular Recommendations ecommerce tool is trusted by hundreds of merchants who want to improve their store’s upselling and cross-selling capabilities.
Now, the leading site search and personalization software provider has made it even easier for clients to work with Recommendations, by implementing a new visual preview function. The main benefits of this new feature are increased control and heightened simplicity - making it quicker and easier for merchants to configure widgets that are in line with their overall strategy. “With our new preview function, merchants can now experiment with their Recommendations and see what will populate under which parameters, without having to set the widgets live on their site first,” explained Findify’s Vsevolod Goloviznin. “Previously, our clients would set up their preferred Recommendations widgets within their merchant dashboard, whether it was ‘Hot Sellers’, ‘Recently Viewed’, ‘Those who viewed ultimately bought’, or any of the other options, but they were not able to see exactly what products would populate in the widget until it went live on their ecommerce site. This new feature will give them a lot more control and make the process simpler for them.” The feature works in two different ways, depending on what type of Recommendations widget the merchant wants to preview. For those that do not need any product input to populate, such as ‘Newest Products’ and ‘Hot Sellers’, the preview loads as soon as the merchants clicks ‘Preview’. For those that require specific product input, such as ‘Customers Who Viewed This Ultimately Bought’, ‘Customers Who Viewed This Also Viewed’, and ‘Frequently Purchased Together’, merchants can click ‘Preview’ and then search for or select a specific product from a dropdown menu. This means they can see what will appear in the Recommendations widget for each individual product. 
Viewing Recommendations Previews for specific products.
While Mr Goloviznin says that, on the surface, this might seem like a small addition, it will really speed up the creation of effective Recommendations widgets, making life more convenient for merchants.
“I think it’s a great little feature, worth celebrating - especially since it was a request from our merchants. We believe in focusing innovation around areas that are relevant for other users and urge everyone to share their needs with us." 
Findify’s new Recommendations Preview function in action on the Merchant Dashboard.
What are ecommerce Recommendations?
A Recommendations solution allows retailers to maximize their online store’s upsell and cross-sell capabilities. It uses available space in certain areas of an ecommerce store (such as the home page, cart page, product pages, or even any custom page) to offer shoppers additional items they might want to buy.
Findify’s popular Recommendations solution is a powerful ecommerce tool which includes categories such as ‘Hot Sellers’, ‘Recently Viewed Items’, ‘Newest Products’, ‘Frequently Purchased Together’, ‘Customers Who Viewed This Also Viewed’, and ‘Customers Who Viewed This Ultimately Bought’. 
A Findify Recommendations widget in operation on the site of Findify client Happy Socks.
Merchants can select to show these widgets on their site without having to manually curate the products shown.
Furthermore, merchants can even freely add Merchandising Rules to tailor what's shown, without losing the automatization. For example, as a fashion store you can tailor a "complement the look" by removing results from the present category, or even selecting what categories should be featured. The algorithm will then select most relevant products from this defined pool of products. This solution also includes access to advanced search analytics which enable merchants to better understand the behaviour of their shoppers, giving them access to extremely valuable, and actionable, insights. For more information on Findify’s powerful ecommerce tools, including personalization software and solutions such as Personalized Search, Smart Collections, and Recommendations, book a demo here. Read more news about our productExperience the Findify effect
New labelling capabilities are set to streamline the way Findify clients using an API integration track their Recommendations widgets.
During the last few years, the growth of PWAs, headless, and fully customized frontends have increased the usage of API integrations. While this development simplifies a lot of processes, it also puts demands on changing how systems are configured and labelled.
At Findify, we are aware of how important it is for merchants to easily access relevant information, which is why we are constantly tweaking and refining our own analytics offering - giving our clients the most actionable insights possible. Our most recent development in this area sees a change to the way we label certain Recommendations widgets within our popular Recommendations solution, meaning merchants can now more easily track certain aspects of the solution.
“As more and more clients use API integration and are accessing the Recommendations API directly instead of via widgets created in the Findify merchant dashboard, it has not been fully straightforward to track the performance of such Recommendations in our analytics,” explained Findify’s Vsevolod Goloviznin.
Acting on feedback received from clients, Findify’s development team came up with an easy solution - allowing for the inclusion of an API label to be used later on in the analytics when accessing the API directly. Mr Goloviznin said this would make the tracking process that bit easier for a lot of retailers. “I think it’s a great little feature, worth celebrating - especially since it was a request from our merchants. We believe in focusing innovation around areas that are relevant for our users and urge everyone to share their needs with us." For more information on Findify’s powerful ecommerce tools, including personalization software and solutions such as Personalized Search, Smart Collections, and Recommendations, book a demo here. Read more news about our productExperience the Findify effect
This brand new feature makes it easier for retailers to organize their Recommendations widgets, improving workflows and increasing clarity within their Findify merchant dashboard.
It may be a slightly overused quotation, but the old adage “Time is Money” still rings true today - especially in the world of ecommerce.
At Findify, where our solutions have been trusted by more than 1,800 ecommerce clients, we are extremely aware of the precious commodity of time and strive to streamline processes for our clients wherever possible. As such, our latest feature seeks to improve the way our clients work with Findify’s Recommendations solution, an ecommerce tool which boosts the upselling and cross-selling abilities of online stores.
An example of a Recommendations widget as shown on the site of Findify client Happy Socks.
Up to now, merchants have been working with Recommendations in the following way: they log into the Findify merchant dashboard, migrate to the Recommendations tab, then create, name and configure their new Recommendations widget.
While the current process of setting Recommendations is simple and straightforward, some confusion can arise later should the retailer create more and more widgets. This is largely due to the fact that the widget’s title is often generic, and the widgets become hard to distinguish from each other. Acting on feedback received directly from merchants, Findify has now solved this issue by adding a new ‘Notes’ section where the merchant can describe each widget and the context in which it was created.
“This is a simple addition to our Recommendations product, but it’s a very important one which will make life easier for a number of merchants using the recommendations solution. We think it will prove to be extremely effective,” explained Findify’s Vsevolod Goloviznin.
“We also think merchants should experiment with this new feature. They can write not only what the widget is for, and what filters it has, but also their thought process at the time, the results they aim to achieve with it, and even go back later and add in the actual results they got. This detail will really help the merchant in the future when assessing campaigns, and will also help to improve clarity in case a handover needs to take place - whoever takes over will more easily understand the solution if individual widget descriptions are included.” Clients will find the new feature under the Recommendations section of their Findify merchant dashboard. 
Where to find the new Recommendations notes section within the Findify dashboard.
“It is also important to note that adding or changing notes in this section will not affect how the Recommendations widget is presented to shoppers,” added Mr Goloviznin.
“But what it does do is improve the way in which our clients work with, and interact with, our solutions - saving them time, improving their workflow, and making them more productive.” For more information on Findify’s powerful ecommerce tools, including personalization software and solutions such as Personalized Search, Smart Collections, and Recommendations, book a demo here. Read more news about our productExperience the Findify effect
We’ve all been there - browsing an ecommerce site, selecting our preferred item, and then finding out it doesn’t ship to the country we want it to. It’s a frustrating experience, and one that greatly disrupts the shopper journey.
In the past, merchants have solved this issue by creating multiple, localized stores - one store for the US, one for the UK, and one for France, for example - with each store offering a regional-specific experience in terms of products, currency, language, and spelling. But this solution does not marry well with another growing ecommerce trend - the single URL solution. This hugely popular method is exactly what you think it is - it’s one single URL for your website that you direct all your traffic to, regardless of the customer’s geographical location. There are huge benefits to this path - having a single URL ensures higher Google search ranking and better SEO, not to mention the substantial saving on time, effort, and resources when not having to create & manage multiple stores. When it comes to managing the assortment, merchants have traditionally had to choose between either providing tailored, localized experiences, or reaping the benefits of a single URL solution. If going with the latter, the workaround was usually to redact some products shown in the product feed - so if a UK customer searched for “jeans”, their resulting product feed would show all products, but with the products unavailable for purchase in the UK either blurred, faded out or carrying a sticker like “Currently Unavailable in Your Area”. As you can imagine, this creates an extremely poor user experience. Human nature often dictates that we want what we cannot have, so if a shopper sees an item in their search results that is unavailable, they will most likely be frustrated and annoyed by that. 
A typical example of a store showing unavailable items
An ecommerce bolt of lightning 
Findify, a leading site search and personalization software company, has a rich experience of international commerce from working with more than 1,800 merchants from almost all the continents of the world (Antarctica still pending), but it wasn’t until teaming up with Reach, leveraging the complementing technology and know-how, that a solution was developed.
“Reach has a great team, with experts in international trade. Matt Steinbrecher has a lot of ideas, and the capabilities of their platform made the development easy,” said Findify CCO, Joakim Amadeus Olsson. “Reach’s popular ecommerce tool gives international shoppers a frictionless local currency experience with an API that opens up relevant local pricing and payment methods - essentially showing customers the correct currency based on their geographical location, even with merchants using a single URL solution.” It soon became clear to both parties that Findify and Reach were well placed to work together on the issue of product feeds frustratingly showing customers items of geographical irrelevance to them. Two state-of-the-art solutions working together to increase ecommerce revenue Reach's Matthew Steinbrecher
“When consulting with our merchants on international strategy through a single website, native currency and payment methods are just one piece to the puzzle when we’re talking about hyper-localized shopping experiences,” explained Mr Steinbrecher.
“Oftentimes geo-targeted merchandising is a huge part of the global strategy, as factors such as shoppers wanting to focus on different collections based on the season, country specific brand restrictions, and/or inventory availability. Partnering with Findify solves a critical part of our clients international strategies, allowing merchants to further localize the shopping experience.” “We are very happy with the fruits of our labour and the expertise of Reach's team,” added Mr Olsson. “Working together, we were able to solve the issue - now, with both Findify and Reach installed, merchants can ensure that none of their customers will see products they cannot have. In optimizing the product feed in this way, merchants are boosting conversion by providing better customer experiences, and providing shopper journeys that are superior to their competitors.” On a basic level, the solution works like this: Reach’s script reads where in the world the customer is, based on their IP address, in order to show the correct currency for that region. Findify’s software then accesses this extracted information, to determine which sales cluster the customer is included in. This could, of course, be a country, but most commonly a region: North America, The Nordics, Europe, etc etc This information is loaded only once, on the landing page of the ecommerce website, and does not need to be loaded again - meaning it is already there when shoppers get to the product pages, causing no additional loading times and maintaining the current rapid site speed enjoyed by Findify merchants. While this is the solution to the aforementioned problem of merchants showing unavailable products in their feeds, the method has a secondary benefit.  Findify's Joakim Amadeus Olsson
“This could also be used on an operational, merchandising level. We know many merchants have different tactics and collaborations in different regions, so with this method merchants could also boost certain products to specific markets,” explained Mr Olsson.
“Above all, we at Findify pride ourselves on showing customers only the products that are the most relevant for them - using our unique personalized search algorithm and overall personalization software, and now also utilizing geo-data. So we’re immensely happy that we were able to work with Reach in solving a problem that so many merchants were experiencing. It’s been a fantastic collaboration which is going to benefit countless merchants and shoppers across the globe.” * Findify, an advanced AI solution and personalized search ecommerce tool which uses machine-learning algorithms to change the order in which each individual shopper sees each product, creates an optimized and personalized journey for online shoppers. For more information, book a demo here. * For more information on Reach, a leading provider of ecommerce currency conversion solutions, go to www.withreach.com. Check out the latest Findify clientsExperience the Findify effect
Trusted by more than 1,800 merchants across the globe (six continents and counting!) Findify is a veritable linguistic expert. From Hebrew to Japanese, Hungarian to Norwegian, experts at Findify have molded their Site Search and Personalization tools to work across more than 20 languages.
Not all languages were created equal, however. Many languages are intensely complex, requiring excellent decompounding standards. Swedish, in particular, is a good example of this. While Findify is today known as a global provider, the company was founded in Sweden. As a result, a large segment of the Findify client base is still currently located in this area of Scandinavia. Like most smaller languages, Swedish has far fewer resources online than more widely spoken languages such as English, Spanish, French and German, and so there are relatively few open-source solutions that properly analyze and handle the Swedish language. In true Findify fashion, the company decided to build their own scalable solution. Rather than a specific fix, only applicable for Swedish, they came up with a methodology with broad applicability, and perfected Swedish in the making. English search
When you search for a “hand cream” in English, typically your search query is piped through a set of normalization filters:
Why is English different from other germanic languages?
Germanic languages have the idea of “compound word”, when multiple dependent words are merged together in a long super-word. This behavior can also be observed even in English (example: basketball), but it’s not as frequent as for Swedish. So “hand cream” becomes “handcreme”, which makes the English-style search flow a bit more complicated:
The general idea behind the compound word analysis in search is to leverage the spell checker dictionaries to split the words into parts and then index them as usual. There are even some ElasticSearch supports for German, but not for Swedish. But still, even for German the splitting was quite aggressive and could split a word like “skyscraper” into “sky+crap”, which could result in quite unexpected search results. Academic approach
There is quite a lot of academic research happening around the Swedish language, so we checked out a couple of the most cited articles about decompounding in Germanic languages. Most of them can be summarized into a single algorithm:
Which seems quite trivial from the first sight. Scoring the split candidates
The most simple approach will be to score these candidates by the word frequency, so for su+nshine, sun+shine, suns+hine, sunsh+ine it’s quite clear that the sun+shine is the most probable one.
But this approach quickly showed that there are quite a lot of false-positive splits like for the word “thermometer” which should not be split at all, but thermo and meter are very frequent words! So instead of writing too many quirks for the algorithm, we decided to trust the machine and build the ML algorithm, which will attempt to build the best scoring rule out of the training data. Machine learning Swedish compounds
To ask the machine to learn something, at the first step you need a lot of examples, training data. For Swedish compounds, we found two datasets suitable for it:
The main problem with these datasets is that they had only positive samples for the training (so like “this word is compound”). We can use all other words as negative samples (like “all other words are non-compound”), but it’s not always true: other words can also be compound, but not yet marked as compound. But as we had no other data on this step, we still trained the model as is, using the following input data:
So the model can try to build the (hopefully) perfect splitting rule based on the data from the grey part of the table above. After the training for each split the model emits a number - the probability of likelihood that that split is correct.
The problem is that the resulting model is not always completely sure that split is possible. See the histogram of score distribution below: 
As you can see, when the score is close to 1.0 (or to 0.0), the model is almost sure that this split is correct (or not possible). But what with all these values in the middle?
Leveraging human power
The more good data you have for the training, the better the model. So to further improve it we generated 3,000 split candidates in which model was not really sure and asked a couple of native Swedish speakers to help it in labelling the dataset.
We also conducted a dataset expansion in a way that if sun+shine is a compound, then all the forms like sun+shines, sun+shining are also compounds. After we carried out this set of improvements, the model was much more reliable at detecting the proper splits: 
But still a single question was not clear: Which threshold value can be defined as a proper level of confidence that the split is correct?
Choosing the threshold
As formally we’re doing a binary classification here, which emits some confidence score between 0 and 1, we still need to choose the best threshold so we can assume that any one above it is classified as compound.
And after we chose the threshold, all the classification results can be split into four categories: 
And the numbers can look like this:
So the number of correctly classified compounds out of all compounds is called precision. And recall is the number of correctly classified compounds out of all words classified as compound. By moving the threshold up and down you can balance between precision and recall.
There is also a way to compute the best precision and recall combination called the F1 Score. For our case it was around 0.97, which resulted in a threshold of 0.88. Word forms
But being able to split (or not split) the word into its parts is still a first step for proper Swedish language support:
So to properly support Swedish, we also need to have an extensive dictionary of word forms, so for the word “shining” we can know that the lemma (normal) form is “shine”. For Swedish, we used the same Folkets/Talbanken datasets, and also SALDO morphological dataset for all the word forms, which further improved our ability to properly search Swedish. Results
For one of our swedish merchants, Apohem, we were able to deliver search results for complicated queries:
In the screenshot above, for example, a search for “huvudvärkstabletter” (head pain pills) is correctly decompounded to the parts. And even if there is no exact match for the word “huvudvärktabletter” in these products, we still were able to match products having separate words “huvud”, “värk” and “tabletter”.
This piece was written by Findify’s Machine Learning expert Roman Grebennikov. Findify is a leading provider of AI-powered site search and personalization software. For more information on Findify’s powerful ecommerce tool, which includes solutions such as Personalized Search, Smart Collections, and Recommendations, book a demo here. Read more news about our productExperience the Findify effect
Findify is now utilizing the power of Delta Testing to ensure clients receive the highest level of proactive technical support following their implementation of Findify’s Personalized Search, Recommendations, and Smart Collections solutions.
Developers at Findify have recently upgraded the way in which they monitor the performance of Findify solutions - a change which is set to strengthen the provider’s client support service even further.
While Findify was already using automated testing to carry out this function, the leading site search and personalization software provider is now taking this to the next level by using automated Delta Testing to increase the speed with which they identify and deal with technical issues. This improved method of technical support is complementing existing security practices such as top issue tracking protocols and having developers on call round-the-clock who are notified by any system disturbances. What is a Delta Test?
The term delta stems from its usage in mathematical equations, where the symbol for delta is used to denote a change in a variable.
Put simply, a delta represents a difference between two values - a change of some sort. In practical terms, this could be a change in the welfare of a patient whose condition is improving or worsening, a change between the current results of a laboratory test and the previous result of the exact same test, or a change in the value of an asset, for example. The method can be used in almost any situation in almost any industry. Why is Findify using Delta Testing?  Anders Gullander, Head of Product
As a leading site search and personalization software provider, Findify solutions are used by hundreds of clients, and tens of thousands of shoppers, each and every day.
If any aspect of the service is not performing exactly as expected, Findify developers need to know about it before it has any major impact. However, with so many clients using Findify solutions, it would be impossible for them to manually go through each store and check for discrepancies on a daily, weekly, or even monthly basis. That’s where Delta Testing comes in. On a basic level, this method essentially sees Findify experts telling the test what results it expects to see. The test is then run on a continuous, round-the-clock basis, checking the services being provided, and any abnormal change (or delta) is automatically flagged with the Findify team. “We are using this in many instances, but let’s take a product feed as an example,” said Findify Head of Product, Anders Gullander. “Let’s say this product feed usually has around 100 items in it. Maybe that number occasionally goes up a little, goes down a little. These changes are considered normal, and are not flagged. But what if suddenly there are only 30 products showing? This is something that would be flagged by the Delta Test, which would then immediately alert Findify developers.” What does this mean for Findify clients?
While Findify developers are always on alert for any issues that might arise, the implementation of Delta Testing means this support has been strengthened even further.
This means any issues, or anything that could potentially become an issue, are identified immediately, ensuring a Findify developer can look into the instance and quickly fix anything that needs fixing. “Delta Testing is a fantastic next step for us, as we felt it would improve our already strong quality control and automated testing capabilities,” said Mr Gullander. “This is a great system because it takes measurements over a long period, continuously assessing trends, and identifies issues before they even become issues. Also, Delta Testing doesn’t just assess in terms of black or white, broken or working, it also monitors shades of grey - instances where something is not working, but also instances where something is working as well as it could.” Delta Testing is now in use by Findify developers, who say it has proven extremely successful so far and has already been a key factor in improving the support they offer to their clients. For more information on Findify’s powerful ecommerce tools, including personalization software and solutions such as Personalized Search, Smart Collections, and Recommendations, book a demo here. Read our most recent product updatesCheck out the latest Findify clientsExperience the Findify effect
This brand new feature makes it easier for online shoppers to find the products they are looking for - also when searching using numerical SKU-style parameters.
The latest feature released by Findify developers sees the leading Site Search and Personalization software company improve its Personalized Search function even further - this time with improved numerical search functionality. As the most popular Findify solution, with an algorithm powered by the latest in AI and machine-learning developments, the company’s Search solution is constantly being improved, tweaked, and optimized - for the benefit of all merchants using the valuable ecommerce tool to increase revenue, boost conversion, and improve the overall user experience on their website. The new numerical search functionality now encompasses SKU-style look-ups. What is an SKU and how can Findify help customers who search for them?
A Stock Keeping Unit, or SKU, is a unique code that is assigned to a product to identify it, and helps merchants keep track of inventory.
To give an example, a paperback edition of the first book in the Lord of The Rings Trilogy could have this SKU: LOTR-1-PB. Things get a little more complicated, however, when searching for items with similar, or complicated, SKUs. Take computer gaming company Nvidia for example. The company could have a product with the SKU of: Nvidia gfx1080. “Previously, Findify search would take this term and treat it like a word, meaning searching just for ‘gfx’ or just for ‘1080’ wouldn’t find the product,” explained Anders Gullander, Findify Head of Product. “Now, we split it into parts, so that it’s easier to search, improving our text analyzer to break down these SKU-style strings. Searching just for ‘gfx’ will find all products with gfx in it, including ‘Nvidia gfx1080’ and ‘Nvidia gfx960’ etc. Searching for ‘gfx1080’ or ‘gfx 1080’ will show exact match products first, followed by partial matches. If there are no exact matches it will fall back to the usual expanded search, as it did before.” 
An example of numeric search functionality on the website of Findify client Shopping Express.
While this new functionality may seem basic, it is actually quite a complicated matter behind the scenes.
Mr Gullander explained there is a lot of complexity involved in breaking down not only words, but also numbers, and teaching the system to understand the endless ways and combinations that a product title could be written into a search bar. After extensive testing of the new feature, which involved invaluable feedback from merchants, the experts at Findify have reported this improved numeric search feature greatly improves the search functionality for shoppers conducting SKU-style searches. “This is a really great new feature, which I think a lot of merchants, and online shoppers, are going to benefit from. Our customers that are already using it see a great improvement,” said Mr Gullander. “We have a lot of big projects in the works at the moment, but it’s great to still have the capacity to roll out these smaller, but extremely useful, features which will help our clients to continue to sell more online.” For more information on Findify’s powerful ecommerce tools, including personalization software and solutions such as Personalized Search, Smart Collections, and Recommendations, book a demo here. Read our most recent product updatesCheck out the latest Findify clientsExperience the Findify effect
The brand-new default merchandising feature makes it easier for merchants to optimize sales by pushing products in all locations on their website in line with their strategy (e.g. including items that are on sale or that have higher margins) WITHOUT sacrificing the power of personalized product rankings.
Smarter Selling Made Simple
While there are many differences between online and brick-and-mortar stores, some aspects can, and should, remain the same.
Say, for example, you want to order 100 red dresses… but you accidentally order 1,000. What do you do with the excess? In a brick-and-mortar store, you would start heavily merchandising the dresses - maybe putting a sign in the window, maybe creating an offer around the item, and definitely placing them in prominent areas of the store so they can sell quicker. Giving these items the same treatment online can be trickier, depending on the solutions you use, but it can (and should) be done. Using solutions without personalization (using AI to learn what your shoppers want, then reorder your products in real time to appeal to each individually) you can perhaps feature the dresses prominently on your website. But this is risky - you could lose sales if the dress doesn’t appeal to shoppers who might have purchased other items, had you pushed them instead. Use a solution with personalization and you likely won’t be able to push the red dresses at all - the algorithm will continue to optimize only the products your shoppers are most likely to buy. This is better than the previous option, as sales are optimized, but it still doesn’t help you move those dresses. This limitation is exactly what prompted the developers at Findify to create the new Default Merchandising Rule feature. With this capability, you can easily push the products you most want to sell, WITHOUT sacrificing the power of personalization thereby improving your conversion rate without any manual coding or tweaks. How does the Smarter Selling aspect of the Default Merchandising Rule work?
'You can use this capability in a multitude of scenarios. You can apply any dimension to the field, utilizing any property that best fits your strategy - including the boosting of products with higher margins, and those that are on sale, for example.
With the Findify dashboard, you simply go to ‘Merchandising Rules’ and click either 'Create Default Rule' or ‘Edit Default Rule’. You can then select the ‘Rule Type’ and the ‘Variants’, input a ‘Value’, and hit save. Deleting the rule is equally simple. Again go to ‘Merchandising Rules’ and click ‘Edit Default Rule’. Then press the trash icon, and hit save. The rule is now deleted. Anders Gullander, Findify Head of Product, further explained that retailers who already utilize Findify solutions may wonder what the difference is between this new feature called Default Merchandising Rule and a feature Findify already has, called Merchandising Rules. “While a standard Merchandising Rule applies to a defined search term, the Default Merchandising Rule applies to ALL search terms. This is great if you want to create a rule that spans across every single search term,” he said. “A Default Merchandising Rule does NOT affect any of the standard Merchandising Rules that you have created. The standard rules will always take priority over the default. This ensures that you are not promoting items at the expense of maximizing sales.” Another great aspect of this new feature is that you can apply it not only to search, but also to recommendations and smart collections. As these are three different shopper journeys, the optimal strategy would be for a merchant to create default merchandising rules across all three solutions. “This really is a fantastic new feature, which I think a lot of merchants are going to be happy with, and will use a lot,” said Mr Gullander. “It’s very exciting to be a part of a team which can focus not only on developing the big solutions, but also on rolling out smaller features like this one which helps make our clients’ lives easier, and helps them continue to sell more online.” - To read more about our clients, click here. For more information about how Findify personalization and optimization can increase revenue for your ecommerce business, book a demo with us here.  While major breakthroughs in product innovation are always welcome, the smaller steps - the tweaks that make life a little easier - should not be overlooked.
Have you ever created a merchandising rule and pushed it live, for example, only to realize you then needed to go back and change something? Findify’s newest preview feature fixes this potential problem by allowing users to view their merchandising rule in action before it is published. The update can be found, in the Findify dashboard, under Merchandising > Merchandising Rules. Once you have tailored your merchandising rule to you needs, you can then hit the new ‘Preview’ button at the bottom right of the screen. This allows you to see what the rule will look like for your customers once you publish it. “In talking to a few of our clients, we realized some were having issues with the creation of merchandising rules. They would create one, push it to live, check it, and then notice an element they wanted to change,” explained Anders Gullander, Findify’s Head of Product. “So instead of having them go back into merchandising on their dashboard and make the change, we decided to implement this new preview feature. It’s a change that will make our clients’ lives that bit easier, and that’s important to us.” While the Findify team continues to work on creating innovative new solutions, it is also working on improving the current offering, thus ensuring clients continue to receive the best in smart search solutions. Is there anything you’d like to ask us about the new feature? Is there any other feature you’d like us to implement? Email us on yourfriends@findify.io. |
Discover more how Findify can improve your salesRequest a demo
CategoriesCollections Interface Linguistics Merchandising Recommendations Search Shopify Sync Technical Support |
|