Product Matching Farm for Retailers

Product Matching for Price Comparison

One of the most important aspects of competitive data is product matches. Matches are made when attributes of a product (e.g. picture, product title, product description) are compared to products that other retailers sell.

The Types of Matches

When enough attributes are determined to be the same, they are considered a "match" and are collected to be used to compare a retailer with market peers in regards to prices.

Two main types of product matches:

Identical matches
Identical matches – Exact matches are made when the products compared have the exact same features
Similar matches
Similar matches – Products that are more-or-less the same but have one or more characteristics, such as color, that make them slightly different from each other

How are matches made? Two main methods:

Manual matching is done visually by a data entry team. Manual matching produces very high-quality matches which are extremely accurate, but the process of collecting them is expensive and slow.
Automatic matching is made via advanced algorithms rapidly searching through a large product database for matching product attributes. Though automatic matching is significantly faster and cheaper than manual matching, it is more prone to error due to issues such as mismatching taxonomy.


Through manual processes with a professional team, about 600 matches can be made per day at an accuracy rate of about 90%.
Automatic matching can produce about 1000 matches per hour, but in some categories like grocery items, the accuracy of these matches is quite low.

Together, automatic and manual matching can be used in conjunction to create a hybrid matching process. Using both manual and automatic matching together can create a large assortment of very accurate product matches.

How other matching methods can fall short

Several factors can severely impede the product matching process. The most common issue revolves around the accuracy of matches. Certain industries and categories of products are harder to make accurate matches for than others.

For example, laptops, which are usually always sold with the same product description, picture and title, have high matching accuracy rates — even with automatic matching, accuracy can be as high as 90%. However, other categories that are often sold under differing titles, descriptions and pictures, matching accuracy falls to rates as low as 30% via automatic methods.

The issue of accuracy in regards to matching can severely impact the rest of the pricing process.

If a company is selling 50 thousand products which are priced using competitive data that has a matching accuracy rate of 70%, 15 thousand products are likely to be priced incorrectly. This is why in our product matching process, accuracy is the highest priority when providing businesses with competitive data. Through a combination of manual and automatic matching, Competera provides an accuracy rate of up to 98%.

140,000 data points with an accuracy rate of 98%

Retailers that are equipped with accurate product matching data such as ours can make drastic improvements in their pricing strategy. RDE found out the benefits accurate data has to offer using 140,000 data points from Competera.

80% sales growth in two months

Previously, RDE’s’ in-house product matching process was tedious, required numerous excel sheets, and often needed time-consuming rechecks due to errors. Using Competera’s competitive data, their pricing strategy was optimized according to market demand more accurately, resulting in an 80% growth in sales in just two months.

Experience the benefits of high-quality matches

Try out Competera’s competitive data to experience how high-quality, accurate product matches could revolutionize your pricing strategy in the retail market