In August 2026, according to INSEE, prices for fresh produce rose 2.9% month-over-month and 5.9% year-over-year. Over the same period, overall food prices rose 0.4% month-over-month.
At first glance, these indicators provide a snapshot of market conditions. But an aggregate trend does notreveal whereprice movements are actually concentrated. It tells us even less about which retailers have adjusted their prices, on which products, through which distribution channels, or to what extent.
An average can mask very different trends
This diversity becomes even more apparent when we look at the product level.
The latest data from INSEE show that in August 2026, agricultural producer prices rose by 2.4% year-over-year. But behind this average, the disparities are considerable: +18.0% for fresh vegetables, with increases reaching +52.5% for leeks, +43.1% for tomatoes, and +34.6% for lettuce.
These figures refer to farm-gate prices, not retail prices. Nevertheless, they illustrate how an aggregate trend can mask very different dynamics across different products.
For a pricing team, the challenge is to take it a step further: to observe how these changes play out among competitors, which products are affected, and where the discrepancies become significant enough to warrant further analysis.
Relying solely on an average trend can therefore cause you to miss the movements that are truly important for your positioning.
From the global market to the product level
For pricing teams, the challenge is to gradually drill down into the data. The analysis can begin at the category level and then be refined by product , brand, format, distribution channel, retailer, or geographic area.
This level of detail makes it possible to answer much more operational questions:
In which areas does our positioning differ from the market?
- Which competitors have recently changed their prices?
- Do the discrepancies observed apply to an entire category or only to certain items?
- Is a change a one-time occurrence, or does it reflect a genuine trend?
- Which products actually require a decision?
Multiple levels of granularity to combine
A pricing analysis can be refined according to several complementary dimensions. Product-level granularity makes it possible to identify the specific SKUs where price variations are most concentrated. Competitive-level granularity distinguishes changes specific to a particular retailer from broader market trends. Geographic-level granularity reveals discrepancies between catchment areas or store networks. Finally, temporal granularity makes it possible to distinguish between a one-time variation, a short-term promotion, or a lasting price repositioning.
These factors do not necessarily need to be analyzed in the same level of detail. For products that are highly comparable or that contribute to the brand’s price image, more frequent and detailed monitoring may be warranted. Conversely, secondary products can be monitored less frequently.
Granularity is also a matter of frequency
Granularity is not limited to the product, competitor, or geographic level of analysis. It also applies to time.
A single snapshot provides insight into price levels at a given moment; a sufficiently detailed historical record, on the other hand, allows us to understand price dynamics. The more volatile a market is, the more critical the frequency of data collection becomes.
A short-term promotion, a price reduction that is quickly reversed, or a competitor’s price adjustment that occurs between two data collection periods may go unnoticed if data collection intervals are too long.
The appropriate frequency therefore depends on the level of volatility observed and the strategic importance of the products being monitored. Some categories may require daily monitoring—or even more frequent monitoring—while others can be managed at longer intervals.
Not all price fluctuations are signals
Increasing granularity does not mean reacting to every observed price movement. An isolated price fluctuation may be due to a one-time promotion, temporary unavailability, a local decision, or simply an adjustment with no lasting impact.
Conversely, when several trends converge on the same category, persist over time, or involve multiple competitors, they may indicate a more structural shift in the market.
The challenge, therefore, lies less in detecting every fluctuation than in distinguishing price noise from the signal that is actually useful. To characterize a price movement, several criteria can be analyzed in combination:
- its range, in order to identify truly significant deviations;
- its duration, to distinguish a one-time movement from a persistent trend;
- the number of competitors affected, which makes it possible to assess whether the phenomenon is isolated or widespread across the market;
- the strategic importance of the product, including its share of sales, its margin, or its impact on the retailer’s price image.
This analysis allows pricing teams to focus their attention on the variations that truly warrant further analysis or a decision.
Example: The Same Average Increase, Very Different Decisions
Let’s say a category shows an average increase of 4 percent. At this stage, the pricing team might be tempted to consider a comprehensive review of its pricing strategy.
But a more granular analysis may reveal a very different reality: the majority of items remain relatively stable, while a handful of products account for most of the increase. By factoring in the competitive dimension, it also becomes clear that some retailers have absorbed part of this trend, while others have passed it on to a greater extent in their prices.
The question, then, is no longer simply, “Should we adjust prices in this category?” but rather, “Which products should we actually adjust, to what extent, and what potential impact will this have on our positioning?”
Data is only valuable when it enables good decision-making
The goal is not necessarily to collect more and more data. It is to transform a mass of pricing information into actionableinsights .
In particular, this requires having a reliable product-matching system, monitoring the right retailers and distribution channels, measuring significant discrepancies, and identifying trends that truly deviate from the defined rules.
The rate monitoring system then changes its function.
It is no longer just a way to determine whether a competitor is more expensive or less expensive. It is becoming a system that allows you to Identify, contextualize, and prioritize pricing decisions.
The more granular the tracking becomes, the more critical data quality becomes. Comparing two similar but not truly equivalent products can artificially create a price discrepancy and lead to misinterpretation. The reliability of matching comparable products is therefore an essential prerequisite.
The timeliness of the data is just as important. A change observed too late can lose much of its operational value, particularly in environments where promotions, stockouts, or competitor adjustments change rapidly.
Finally, granular data is only truly actionable when placed in context: price history, the distinction between regular and promotional prices, availability levels, and the strategic importance of the product. It is this contextualization that allows an observed discrepancy to be transformed into a truly relevant decision.
From Observation to Management
In a market where prices can change rapidly, the ability to detect a change is no longer enough.
We need to understand where it occurs, what its scope is, how it affects competitive positioning, and whether it actually requires action.
It is precisely this shift from a broad overview to a more detailed analysis that transforms market data into a decision-making tool.
Because, when it comes to pricing, the real question is no longer just: “How are prices changing?”
But:
“When are changes significant enough to warrant a decision?”
Primary source: INSEE, Consumer Price Index – August 2026.
