Why price monitoring is not enough and your business needs product mix monitoring.

In this article, we present a new approach to price monitoring via frequent scanning of product mix. This method not only takes into account the static product mix on competitive websites, available at the start of the price monitoring process, but also product mix changes.

According to multiple already implemented projects, the product mix monitoring method increases price monitoring efficiency by up to 99%. Thus, it achieves the most accurate price monitoring available on the market.

The benefits of using product mix monitoring.

Businesses would agree that the main purpose of price monitoring is to maximize profit via a better understanding of competitive pricing. Effective price monitoring requires two elements:

  1. accurate product and pricing data and
  2. an algorithm to dynamically recalculate product prices.

Unlike traditional price monitoring approaches, which only effectively monitor around 50% of competitive products, frequent product mix scanning is able to detect 90% to 99% of matches.

This practicly doubles the efficiency of pricing decisions.

Where does the additional 49% come from?

Traditional price monitoring approaches employ product matching at the start of the monitoring process. However, at this point, there are on average only around about 50% of matching products available on the competitors’ websites.

Furthermore, when the exact number of matches is not known in advance, 50% is usually taken as an estimate. For example, if price monitoring covers 100 SKUs on 10 competitor websites, 100 х 10 х 50% = 500 SKU are usually taken as an initial estimate for the scope of price monitoring.

When price monitoring is repeated over the following days, competitors not only receive updated prices, they also receive updated product mixes, new products, recategorized products, and deleted products according to their assortment.

If product matching was carried out only at the very beginning of the process, price monitoring would only cover products that were present at the time of the initial matching (green color on the diagram below). Products that were not present in the product mix at the time of matching (red color on the diagram) will be excluded from price monitoring because they were not detected and matched at the start of the process.

This is why the traditional price monitoring approach covers, on average, only up to 50% of a competitors’ offers.


Moreover, in the traditional price monitoring approach, the identification of added products takes place irregularly and often ad hoc. Due to the changes (new products added, but not monitored; existing products are moved within the product mix) in the product range of monitored products, the efficiency of monitoring quickly decreases from the initial 50% to 30-40% within the first 2-3 months after the launch of the price monitoring process. Additional manual and selective product matching can help increase the efficiency of price monitoring back to about 50%.

The irregularity and selectiveness of manual product matching are caused by a very labor-intensive process. It is impossible to imagine a single person or a team of people, who monitor multiple websites daily with thousands of products and manage to identify product mix changes. For this reason, repeated product matching (rematching) is almost never part of the traditional product price monitoring approach.

Why is the frequent product mix scanning method twice as effective?

First, it is worth mentioning that price monitoring based on the frequent product mix scanning approach has already been tested on multiple projects in collaboration with established price monitoring services.

Second, frequent product mix scanning performs price monitoring on a regular basis to keep price monitoring more up to date. This approach establishes a new mandatory procedure: regularly scan competitors’ websites to identify changes in the assortment.

Comparison of the frequent product mix scanning method (green color on the diagram) vs. the traditional product price monitoring (red color on the diagram)

Comparison of the frequent product mix scanning method (green color on the diagram) vs. the traditional product price monitoring (red color on the diagram)

As a result, frequent product mix scanning doubles price monitoring accuracy (green area on the diagram).

By following the frequent product mix scanning method, price monitoring accuracy can reach 99% and can remain at this level. The following three rules should be followed:

  1. Frequent crawling of a competitor’s website collects the target product range.
  2. Automatic product matching recognizes products based on defined rules.
  3. The process is performed repetitively. It continuously provides updated product prices for the range of active products.

Frequent automatic product mix scanning, unlike manual matching, is able to automate the daily routine of processing large amounts of similar product data. This method can definitevely identify products and match them. No manual matching, even using the largest teams, is capable of ensuring such efficiency.

When implementing frequent product scanning, our recommendation is to start with weekly scanning and then gradually increase to daily scanning if donor websites allow it. The daily scanning ensures the highest monitoring efficiency (up to 99%). 1% is always left for possible errors.

What is frequent product mix scanning based on?

Frequent product mix scanning is based on the automated product matching technology developed by MarketMixer.

Automated product matching supports continuous scanning, identification and rematching of competitors’ products to keep an always-current view of competitors’ product landscape.

Advantages of the frequent product mix scanning method

1. Improves data accuracy for decision making

Automated data collection and frequent updates provide new opportunities for data-driven decision making. Decisions rely on fresh data, collected daily, compared to data, collected a month ago.

2. Enables more accurate pricing

It is well known that as early as 2013, Amazon has been updating product prices 2.5 million times a day to better respond to demand and to maximize revenue. By using frequent product mix scanning, your business can become one step closer to the market leader.

3. Reduces manual work and human errors

It is clear that errors in data cause problems for retailers. The automated frequent product mix scanning process reduces human errors by at least 75%.

4. Compatible with any existing price monitoring technology

Automatic product matching, powered by MarketMixer, complements any existing price monitoring technology. There is no need to replace existing price monitoring technologies. Most modern systems can be integrated.

5. Handles unlimited data volume and frequency of updates

Cloud data processing handles unlimited volumes of data and employs as many servers as required for parallel processing.

6. Facilitates assortment planning

By using a non-exact matching approach to match similar (based on identical product attributes), but not exactly identical products, it becomes possible to monitor the prices of products that are not yet present in your own assortment, but which would make sense to include.

7. Indicates competitors’ product delivery dates

By analyzing the peaks in a competitor’s product introductions, it becomes possible to better understand the dates and frequency of product deliveries to your competitors.

Known issues

Similar to any other method, the frequent product mix scanning has its shortcomings. We are aware of them and are working to further improve the approach.

1. Initial setup takes time

Setting up automatic product matching requires the creation of data models and related algorithms. This initial setup takes time early in the process but is fully compensated for during regular operations. The initial setup enables the quick and continuous addition of new data sources and products to be monitored.

We are aware of this issue and keep on improving the method to reduce the initial setup time and ensure a faster process for including new products in the monitoring

Comparison of manual and automatic data processing

Comparison of manual and automatic data processing

2. Cost of initial setup

Initial setup requires the involvement of experts in data and machine learning. Expert time has its costs. We are constantly improving and standardizing these processes and procedures to automate them and to make them independent of manual labor.


Starting at 400 euro per month.

Go top