Constructor.io Releases logo
Back to Homepage

Releases

Constructor.io Releases

Subscribe to Updates

Labels

  • All Posts
  • Fix
  • Announcement
  • Improvement
  • new
  • This Week in Engineering

Jump to Month

  • May 2025
  • April 2025
  • March 2025
  • February 2025
  • January 2025
  • December 2024
  • November 2024
  • October 2024
  • September 2024
  • August 2024
  • July 2024
  • June 2024
  • May 2024
  • April 2024
  • February 2024
  • January 2024
  • December 2023
  • November 2023
  • October 2023
  • September 2023
  • August 2023
  • July 2023
  • June 2023
  • May 2023
  • April 2023
  • March 2023
  • February 2023
  • January 2023
  • December 2022
  • November 2022
  • October 2022
  • September 2022
  • August 2022
  • July 2022
  • June 2022
  • May 2022
  • March 2022
  • January 2022
  • December 2021
  • November 2021
  • October 2021
  • September 2021
  • August 2021
  • July 2021
  • June 2021
  • May 2021
  • April 2021
  • March 2021
  • December 2020
  • November 2020
Powered️ byAnnounceKit

Create yours, for free!

7 months ago

Attribute-Level Metrics for Attribute Enrichment Now Available!

We’re thrilled to announce an exciting enhancement to our Attribute Enrichment dashboard — attribute-level insights are now available directly within your dashboard!

Previously, catalog-wide metrics provided valuable insights into enriched attributes’ overall performance. Now, with attribute-level metrics, it’s even easier to track and analyze the performance of individual enriched attributes.

Here’s what’s new:

  • Attribute-Level Coverage: See what percentage of your products are enriched with each specific attribute.
  • Attribute-Level Precision: Track the accuracy of AI predictions for each attribute, ensuring your products are enriched with the right data.

How does it help you? Now, it’s even easier to track and analyze the performance of individual enriched attributes — everything you need is now at your fingertips in the form of clear, actionable metrics. For instance, a low Coverage percentage might trigger concerns that some products are missing enrichment, requiring manual review to diagnose the cause. With attribute-level metrics, it’s now clear that if an attribute like "Sleeve length" only has 50% coverage, this is expected since not all products have sleeves.

These metrics give you full transparency into how Attribute Enrichment is improving your product catalog and streamlining your workflow.

The default view of the Attribute Enrichment Dashboard is now set to attribute-level metrics.

Looking ahead, we plan to improve these insights by:

  • Refining the Coverage metric to reflect only the relevant portion of the catalog for each attribute.
  • Introducing filtering by product categories, so you can narrow down your focus and identify opportunities for further improvement.
  • Providing more detailed breakdowns of how enriched attributes impact user engagement, helping you understand their value even more clearly.

For any questions or assistance, please reach out to your Customer Success Manager or contact us at support@constructor.io.

Avatar of authorValentina Nochka
8 months ago

Segmentation by Platform for Per Item Analytics

Building on the positive feedback we received from our customers regarding segmentation by platform for Search Analytics, we are excited to introduce platform segmentation for Item Analytics. Merchandisers can compare Clicks, Add-to-Carts, and Purchases of a certain item between different platforms, including Desktop, iOS Browser, Android Browser, iOS App (if applicable), and Android App (if applicable). 

Judging by where customers shop for a particular item, merchandisers can adjust the item’s presence and focus on platforms that matter the most. A comparison of metrics between platforms provides a better understanding of shoppers’ behavior. It may suggest driving more traffic from one platform to another or adjusting customer journeys on a specific platform.

We are eager to learn how our customers look at the data for making searchandizing decisions. Please, feel free to submit questions and feedback to your Customer Success Manager or contact us through support@constructor.io

Avatar of authorKonstantin Malkov
new
9 months ago

Top Selling Items

At Constructor, our primary goal is to solve for e-commerce sales. Shoppers add products to carts all the time without necessarily buying them. At the end of the day, it’s purchases that really matter. We’ve always used them in our machine learning algorithms, but now we want merchandisers to have greater visibility into them too. As we add data on Purchases throughout the dashboard, we are excited to revamp the Top Converting Items table in the Conversions section of Analytics.

The new table now features Top-30 items based on Purchases along with data on Clicks and Add-to-Carts. For a friendlier experience, we enriched the items' descriptions with names, IDs, and miniatures. By clicking on an item, merchandisers can proceed to a single-item analytics page to track its performance over time.

With data on purchases, we want merchandisers to have their top sellers at hand so that they can benchmark the rest of their catalogs and plan new stock. 

We start by showing the top 30 performing items. Future iterations will include the ability to look beyond that and find particular items. 

We are super excited to learn more about how merchandisers look at their product discovery experience through the lens of individual items. Please, submit any questions and feedback to your Customer Success Manager or contact us through support@constructor.io.

Avatar of authorDaniel Fetisov
new
9 months ago

Purchases data for Search and Browse is here

Optimizing e-commerce product discovery experience for ROI is the primary goal of Constructor. We always strive to enable merchandisers to make better decisions with data. We are extremely proud to announce the addition of data on Purchases for Search and Browse analytics. 

Previously, the dashboard communicated the performance of Search and Browse pages through Views, Clicks, and Add-to-Carts. But shoppers add products to carts all the time without necessarily buying them. Unlike Add-to-Carts, Purchases directly correlate with the overall sales revenue. We’ve always used them in our machine learning algorithms, but now we want merchandisers to have greater visibility into them too. By having data on Purchases, merchandisers can track the business performance of Search and Browse pages and make data-informed decisions.

Purchases are available on Search and Browse charts for tracking over time and in tables for tracking particular search queries and Browse pages. For Browse pages, data is available for Categories, Collections, and Facets. 

Tracking Purchases over time for individual search queries and Browse pages is coming soon.

If you have any questions or feedback on Purchases data, please connect with your Customer Success Manager or contact us through support@constructor.io.

Avatar of authorKonstantin Malkov
new
9 months ago

Segmentation by Platform for Search section of Analytics

With our continuous aspiration to enable merchandisers to make data-informed decisions, we are excited to provide Search analytics in splits by platform, including Desktop, iOS Browser, Android Browser, iOS App (if applicable), and Android App (if applicable). Merchandisers can compare conversion rates into Clicks, Add-to-Carts, and Purchases between different platforms over time. 

Judging by where customers shop, merchandisers can focus their attention on platforms that matter the most. Comparison of conversion rates between platforms may suggest driving more traffic from one platform to another or adjusting shoppers’ experience on a certain platform.

We are eager to learn how our customers look at the data for making searchandizing decisions. Please, feel free to submit questions and feedback to your Customer Success Manager or contact us through support@constructor.io.

Avatar of authorKonstantin Malkov
new
9 months ago

Actionable Insights At Your Fingertips

We’re always looking for ways to make the lives of merchandisers easier, and we know that merchandisers are busy folks who often have many activities on the go at once. We also know from speaking with a number of our customers that data-driven decisions remain an important part of their merchandising strategy and that it can be time consuming and sometimes difficult to get meaningful insights. This is where we’re looking to help, with a refreshed homepage and new action-oriented data insights for merchandisers. 

With this update, we’re adding two new insights directly to the homepage: underperforming slotting searchandising rules, and underperforming search and browse category pages. The aim of adding these to the homepage is to make it easy and efficient for merchandisers to understand what and where optimization opportunities exist within their product discovery experience. In this update, we’ve also revamped how we’re notifying users about newly added auto-generated synonyms as well. 

This is just the start, we see the homepage as a real hub for dashboard users to be able to get an overview of what is happening across their product discovery experience with an emphasis on actionable, easy to understand insights (while still keeping, and expanding on our analytics and insights throughout the dashboard for folks to more deeply explore as well). You can expect to see more updates in this experience in the future. If you have any ideas of what kind of information would be valuable here, please let us know either through your dedicated Customer Success Manager, or you can contact us through support@constructor.io.

Avatar of authorPing Zhang
Improvementnew
9 months ago

Facet Groups Sorting New Algorithm

Constructor is dedicated to enhancing the shopping experience for our customers' shoppers. Our new intelligent facet group sorting algorithm ensures that the most useful and attractive facets appear first.

What is it? A new algorithm that optimizes the order of facet groups based on various data-driven factors to make product discovery more natural and intuitive for shoppers.

How does it work? The algorithm considers:

  • Popularity: How often each facet is used by shoppers over a recent period.
  • Attractiveness: The attractiveness of the top items within each facet (e.g., facets with options that match products scoring higher on specific pages).
  • Diversity: The distribution of items within each facet to ensure a balanced and diverse set of facets, preventing any single facet from dominating the list.

These principles are combined to calculate the final position of each facet group, using a combination of the popularity score, the average score of top items, and a measure of diversity to determine the overall attractiveness of each facet.

What problem does it solve? Facet group sorting improves the responsiveness and accuracy of facet sorting, aligning more closely with current user preferences and behaviors. This leads to an enhanced shopping experience, helping shoppers find the most attractive products with ease.

Availability: The feature is rolling out to all customers, replacing the current default facet groups sort with this new algorithm.

For more information about this capability, please connect with your Customer Success Manager or contact us through support@constructor.io .


Avatar of authorVardan Aslanyan
Improvement
9 months ago

Introducing Geo-Segmented Boost & Bury Searchandizing Rules

Boost and Bury rules became much more flexible! Now merchandisers can adjust Search and Browse shoppers' experiences according to their locations, such as Countries, States, and even Cities. This can be especially useful for new region-specific product launches and shopper behavior insights. For example:

  • Boost red, blue, and white branded t-shirts for Labour Day (September 2) in the US
  • Boost ready-to-eat meals that cater to urban dwellers in Zurich and Geneva
  • Boost outdoor furniture such as patio sets, garden lounges, and sunshades during the warmer months in Spain, Italy, and Portugal

To use this opportunity:

  1. Navigate to the Searchandizing section of your Constructor dashboard
  2. Add a new Search & Autosuggest or Browse rule
  3. Select Boost or Bury rule and desired items or attributes
  4. Add a new Dynamic Geo User Segment to your setup

New Geo Segmentation opportunity for Boost and Bury rules

For more information about this capability, please connect with your Customer Success Manager or contact us through support@constructor.io.

Avatar of authorDaniel Fetisov
new
10 months ago

Managing multiple indices? New tools to make your life easier

Today our customers are becoming ever more international, operating across multiple regions with multiple production indices. This requires merchandisers to manage multiple production indices, sometimes having to create the same searchandising rules across multiple indices. We have heard from those of you who operate multiple indices that there is a need to copy rules between them. We have released Granular Rule Copying to address these problems, Granular Rule Copying has the ability to copy individual query rules between indices. This feature will be available to both admin and non-admin dashboard users. If you have any questions, connect with your Customer Success Manager or contact us through support@constructor.io.




Avatar of authorPing Zhang
Improvement
10 months ago

Highlighting Results of Boost and Bury in Searchandising Preview

Boosting and burying products by attribute or by a single item allows merchandisers to adapt item rankings dynamically without having to pick a fixed position like they do with Slotting rules. Besides saving time by not having to decide on a certain position, Boost and Bury rules work hand in hand with the Constructor AI algorithms, moving items in alignment with business objectives or other goals for an experience while still delivering on key business metrics.

To provide more transparency to the effect of Boost and Bury rules in real-time, the items in the preview area are now highlighted with a thick gray border, while an arrow in the upper right corner of an item card specifies the direction an item moved (up for Boost, down for Bury).

Boosted item is highlighted with a thick gray border; an arrow specifies the direction.

For more information about this capability, please connect with your Customer Success Manager or contact us through support@constructor.io.


Avatar of authorKonstantin Malkov