Innovation performance: how to compare two launches carried out in different contexts?
An innovative product launched in the spring by a national distributor is showing higher sales volumes than a product introduced in the fall at a few regional depots. Can we conclude that it is more efficient? Not so fast.
To compare the performance of innovations , raw volumes alone are insufficient. Time period, coverage, availability, target customer base, and promotional support can significantly alter the results. Without contextualization, the risk is drawing the correct conclusion… from a flawed comparison.
Two launches, but rarely the same conditions
In the foodservice industry, two innovations rarely follow the same path. One benefits from immediate widespread distribution. The other starts in a limited number of outlets. One arrives during a period favorable to its category. The other must find its place out of season or with a less receptive clientele.
The differences may include:
the launch date and the observation period;
the number of distributors and depots concerned;
the actual availability of the reference;
the types of end users served;
the promotional operations put in place;
the pricing conditions;
the packaging or the sales unit;
commercial activity carried out on the ground.
Comparing only cumulative sales is like placing two products on the same finish line without looking at where each started.
Let's consider a hypothetical case. A new sauce is listed in a national network at the beginning of the summer season. A second product is launched a few months later in several regional depots, without any specific promotion. If the first product generates higher sales volume, this result may reflect broader coverage , a more successful season, or better availability. It does not, however, prove greater adoption on its own.
Why sell-in volumes can skew the comparison
During a product launch, the first orders placed by the distributor provide a useful signal. They indicate the quantities delivered and the initial product placement. However, they do not necessarily describe its actual sales volume.
A high sell-in volume may indicate stock buildup. Conversely, a more cautious start may be followed by regular sales and a gradual expansion of product listings. These two scenarios then tell different stories.
This distortion of performance between sell-in and sell-out can lead to overvaluing an innovation heavily loaded at launch or underestimating a reference that is still not widely distributed, but already dynamic in active depots.
Inter-warehouse transfers add another layer of complexity. A product delivered to one warehouse can be resold from another. Without accounting for these movements when they are available, local performance risks being attributed to the wrong warehouse.
To measure the actual reception of an innovation, it is therefore necessary to complement the view of incoming flows with the outputs from warehouses to end users .
Comparing the performance of innovations within a constant scope
The first step is to make the launches as comparable as possible. This requires defining a common scope of analysis before looking at the results.
Rather than comparing two cumulative sales figures since their respective launch dates, we can compare their first weeks or their first full months on the market. This approach allows us to observe the trajectory of each product at the same stage of development.
Next, consistent parameters must be defined: the same distributors when possible, warehouses with similar profiles, comparable markets, and periods subject to similar seasonality. When these conditions cannot be met, the differences must be explicitly documented.
The question is therefore no longer simply: "Which product sold the most?" It becomes: "Which product achieved the best results considering its length of presence, its coverage, and its accessible market?"
Select indicators that neutralize contextual differences
A sound analysis relies on several complementary indicators. The total volume remains useful, but it must be accompanied by information explaining how that volume was compiled.
The average volume per active repository helps, for example, to compare a widely distributed innovation with one whose listing remains limited. The evolution of this volume over time also makes it possible to distinguish a simple launch spike from a more consistent trend.
The proportion of deposits actually generating sales provides another piece of information: is the referencing actually activated in the field? An innovation may be present in the distributor's files without producing sales in a part of the network.
Analysis by end-user type , when available, also clarifies where the product finds its market. A product may have a lower overall volume while achieving better results with the customer base for which it was designed.
Finally, the growth curve deserves as much attention as the cumulative effect. An innovation can start quickly and then slow down. Another can progress more slowly, as sales teams present it and retailers integrate it into their offerings. These profiles call for different decisions.
Create a context sheet for each launch
To avoid hasty interpretations, a simple method is to associate each innovation with a context sheet . This should specify the chosen start date, the period analyzed, the distributors involved, the depots covered, the target markets, and any promotional activities.
It is also useful to record any events that may have disrupted sales: temporary unavailability, gradual increase in demand, changes in packaging, or incomplete product listing. If information is unavailable, it is better to state this than to compensate with a hypothesis presented as certain.
This document provides sales management, key account managers, and marketing with a common foundation. It prevents one team from focusing on orders, another on sales, and a third on a different timeframe.
Transforming analysis into concrete business decisions
A contextualized comparison should lead to action. If an innovation performs well in the repositories where it is active but remains under-distributed, the priority may be expanding its search engine optimization (SEO ). If it is widely available but under-sold, the focus should instead be on examining its suitability for end users, its positioning, or its marketing efforts.
The analysis may also reveal regional contrasts. Some deposits may show regular adoption while others generate almost no output. Regional managers can then focus their visits on areas where action seems necessary or on those whose potential remains insufficiently exploited.
For Category Managers and marketing teams, sales segmentation helps verify whether the innovation is reaching the target customer base. It also allows for the evaluation of an activation by comparing results before, during, and after the campaign, provided that the timeframe and scope are sufficiently consistent.
What sell-out data changes in the analysis
Sell-out data can provide, depending on the information provided by each distributor, insights into product references sold , volumes , warehouses , and end-user profiles . However, its availability and level of detail must be verified on a distributor-by-distributor basis.
The main challenge then becomes reconciling disparate data sets and maintaining common definitions. A management solution like KaryonFood can address this by centralizing and harmonizing sell-out data from distributors. This provides teams with a consistent foundation for monitoring product launches, comparing different regions, and sharing a common understanding of performance.
The value also lies in discussions with the distributor. An innovation deemed weak based solely on total volume may actually be experiencing good turnover in active warehouses. Conversely, a seemingly successful launch may depend on a very broad reach without any real local momentum. These observations allow for discussions to be based on actual sales rather than impressions.
Compare to decide, not to establish a ranking
The performance of innovations is not simply a matter of designating a winning product. A useful comparison must explain the differences, identify the favorable conditions, and determine the most relevant action for each benchmark.
Before the next product review, start by checking three things: the period covered, the actual coverage, and the sales per active warehouse . This simple review is often enough to transform a misleading classification into a usable business analysis.
Comments