Charts

The Overview tab contains 5 chart categories to help you analyze your dataset. By default, only 4 charts are shown. If you wish to see the other charts, you must select them in 'Select charts'.

Click 'Select Charts' to open the selection menu. You can either select all charts in a category by ticking the chart category as a whole, or you can select individual charts. Once you have applied your selection, click the Select button to generate the charts. By default, the year range for all charts is the past 25 years. You can create custom year ranges as well. You can also reset your date range with the 'Reset' button.

All charts

Each chart has buttons located in the upper right corner. With these buttons, you can modify the size of the chart, export the chart data in various formats (such as JSON, XLS, CSV, PNG, or PDF).

To quickly gain insights from the charts, click the blue Chat button located in the upper right corner. Odin automatically generates descriptions for charts and analyzes data to help you understand the key insights. Please note that all charts are subject to the filters you set. Each time you update your filters, all charts will be updated accordingly.

For some charts in the Technology category, you have the option to choose how the data should be displayed. You can choose between "Average," which shows the average values based on all patents that have been published up to that year, or "Year-By-Year," which shows the values only for patents that were published in that year.

The improvement rate chart shows how rapidly a technology area is improving. Every area of technology is trying to optimize its performance and cost over time. It turns out that the rate at which this happens can be predicted with metrics from patent data. The improvement rate is the percentage of extra performance per Dollar that can be expected from this area of technology every year. This metric was devised together with MIT and numerous papers have been written on the subject (ask us if you are interested!). The easiest way to think of the improvement rate is as interest on a bank account. If you have a really high interest rate on your bank account, your money balance will increase tremendously fast. Improvement rates work exactly the same. Technologies with a high improvement rate improve rapidly, as that percentage or performance per Dollar is added each year. A technology's improvement rate metric can be estimated based on the cycle time and knowledge flow metrics. Cycle time measures how quickly a technology area produces new generations of itself, and knowledge flow measures how much of a step forward those new generations represent. A technology with a short cycle time and high knowledge flow will have a high improvement rate and vice-versa. In and of itself, a certain improvement rate percentage does not say whether a technology will be successful or not. We can only compare a technology's improvement rate to those of competing technologies to see whether a given improvement rate is comparatively high or low. The fastest improving technology out of a set of competing technologies typically becomes the dominant solution, but in order to determine whether a given improvement rate is high or low, we have to compare it to other competing technologies.

The cycle time chart shows how quickly a technology area produces new generations of itself. Technically, we measure the 'median age of backward citations' for all patent families in your dataset published in a certain year. If you think about what a patent is from a high level, you will see that each patent contains the same structure. First you describe what your invention is, then you explain which prior inventions you are improving over, then you argue for why yours is better, and finally you provide a long description of exactly how it works. Your patent has a date, and so do the inventions that you are improving over. Since all patents have dates and the related prior inventions are cited, we can calculate the median distance in years between your invention and those you are improving over. If you scale this up to all inventions belonging to an area of technology for each publication year, you get the median number of years it took to come up with the next iteration of technology. It turns our that this number strongly correlates with how rapidly technologies improve. Technologies with short cycle times, e.g. that take quick steps/iterate fast, improve rapidly, and vice versa. Just as with the improvement rate, a certain cycle time cannot be determined to be high or low in isolation, we must always compare it to competing technological alternatives.

The knowledge flow chart shows how impactful inventions from a certain period are. Technically speaking, we measure the average number of times a patent family in your dataset gets cited by later inventions within the first 3 years of publication. Generally speaking, important inventions are immediately important and get cited much more than inventions that are not important. Therefore, we can use citation numbers as an indication of how important or influential inventions have been. Knowledge flow also strongly correlates with how rapidly technologies improve. Technology areas that get cited a lot generally improve faster, as a citation is an indication that the knowledge that has been created is useful enough to develop upon. Whereas cycle time measures how quickly technology areas take steps forward, knowledge flow measures how large the step was. Technologies that take quick big steps forward improve rapidly and vice versa. Just as with cycle time and the improvement rate, a certain knowledge flow value cannot be determined to be high or low in isolation, we must always compare it to competing technological alternatives.



This chart shows all published patent families in your dataset, and breaks them down by status and publication year. There are 4 statuses that can be shown for patent families. In-force at least 1 member has been granted and is currently enforceable. Pending all members are still in the application phase. Inactive all members have lapsed. Unknown we have insufficiently reliable data from the patent office to assign a status. Technology areas that are emerging will show an upward overall patenting trend. Emerging technology areas will consist mostly out of Pending and In-Force patent families.

This chart shows all published patent families in your dataset, and breaks them down by status. It gives an idea of the overall maturity of the technology field you are analyzing. Mature or declining technologies will have lots of inactive patents, as in many cases the inventions will have already lapsed, and emerging or immature technologies will lean more towards pending and in-force patent families.



Shows how quickly the technology in question will double, triple, etc. its performance per Dollas. If you know how a technology performs today, and at what cost. You can use the TIR forecast chart to find out when a technology will supersede another. In the TIR forecast chart, you can set starting cost and performance figures and add other technologies to the chart. Once you have added all technologies, their improvement rates, and their starting performance and cost figures, you can create a forecast of when technology A will disrupt technology B.  In settings, you can enter performance and cost targets for your technology area in 'Technology'. You can also set the current performance and cost of competing technology areas under 'Segments'. The percentage shown next to the 'Technology' settings is the last known improvement rate from the improvement rate chart. This chart is very useful for forecasting but extremely sensitive to which values you put in. Garbage in = garbage out. If you want to use this chart for decision-making, you have to make sure that the current performance and cost figures you are putting in are reliable. If you only have a hunch, putting in multiple scenario's with error margins might be useful.

In the below example, the solid blue line represents the incumbent technology which is improving at a 19.7% rate. The disruptive technology, labeled segment 1, has a much higher improvement rate of 50% but starts much lower because it is still emerging and has a much higher cost for the same performance (typical of emerging technologies). With these starting performance and cost figures, given the higher improvement rate of the emerging technology, it will likely outperform the incumbent technology by 2031. You can also add multiple technologies at once. If you need help with setting starting cost and performance criteria feel free to get in touch with our Research Team.

These charts give insight into the organizations that are active in your technology area.

For organization charts, it is possible to select up to 25 top organizations based on the cumulative number of patent families per organization.

This chart shows you the top patenting organizations in your dataset. It gives insight into which organizations are most aggresively pursuing innovation in this area. We count the total number of patent families associated with this organization in your dataset and order them from most to least patenting organizations.

This chart shows you the top patenting organizations in your dataset but shows their portfolios as a trend over time. It gives insight into which organizations are most aggressively pursuing innovation in this technology area.

This chart shows you the top patenting organizations in your dataset but segments their portfolio by family status. There are 4 statuses that can be shown for patent families. In-force at least 1 member has been granted and is currently enforceable. Pending all members are still in the application phase. Inactive all members have lapsed. Unknown we have insufficiently reliable data from the patent office to assign a status. Generally speaking, organizations that are actively pursuing innovation in a technology area will have a significant amount of pending or in-force patent families. If the vast majority of an organization's patent families are inactive, they likely have decreased interest in the technology area.

Organization quality charts give insight into the quality of innovation coming from organizations present in your dataset. These charts display improvement rates, cycle time, and knowledge flow for various organizations. Organizations can be sorted based on the number of patent families they own or by metric values.

The improvement rate per organization can give insight into which organizations are innovating the most in a technology area. Identical to how we calculate the improvement rate for the overall technology area (see improvement rate under technology charts), we can also do so for individual organizations. For all patent families in your dataset belonging to an organization, we calculate cycle time, and knowledge flow, and estimate a rate of improvement per organization based on those metrics. There are a few things to take into consideration when using these insights for decision-making. The smaller the portfolio, the higher the risk of outliers affecting the improvement rate. If an organization only holds 4 patent families in a technology area, and one of them is an extreme outlier, this can have a strong biasing effect on the improvement rate metric. Generally, the overall improvement rate for a technology area is more reliable for forecasting, and the improvement rate per organization can be used to determine which organizations should be studied in more detail to observe important innovations. However, take this metric with a grain of salt if that organization's portfolio is very small.

The cycle time per organization can give insight into which organizations produce new generations of technology the fastest in your dataset. Identical to how we calculate the cycle time for the overall technology area (see cycle time under technology charts), we can do so for individual organizations. For all patent families in your dataset belonging to an organization, we calculate their cycle time by organization. There are a few things to take into consideration when using this insight for decision-making. The smaller the portfolio, the more risk of outliers affecting the cycle time metric. If an organization only holds 4 patent families in a technology area, and one of them is an extreme outlier, this can have a strong biasing effect on the cycle time metric. Generally, the overall cycle time for a technology area is more reliable, and the cycle time per organization can be used to determine which organizations iterate the fastest. However, take this metric with a grain of salt if that organization's portfolio is very small.

The knowledge flow per organization can give insight into which organizations produce the most impactful innovations in your dataset. Identical to how we calculate the knowledge flow for the overall technology area (see knowledge flow under technology charts), we can do so for individual organizations. For all patent families in your dataset belonging to an organization, we calculate their knowledge flow per organization. There are a few things to take into consideration when using this insight for decision-making. The smaller the portfolio, the more risk of outliers affecting the knowledge flow metric. If an organization only holds 4 patent families in a technology area, and one of them is an extreme outlier, this can have a strong biasing effect on the knowledge flow metric. Generally, the overall knowledge flow for a technology area is more reliable, and the knowledge flow per organization can be used to determine which organizations produce the most impactful innovations. However, take this metric with a grain of salt if that organization's portfolio is very small.

These charts give insight into where in the world organizations in your dataset are focusing their patenting efforts. Each chart can be displayed by Country or Continent and show the top 'X' patent family (from the top 5 up to the top 25) holding countries/continents in your dataset.

This chart shows the top patenting organizations in your dataset and breaks down their portfolio by country or by continent. It gives insight into where the organizations in your dataset are focusing their patenting efforts geographically. Patents only provide protection in the country they are held in. In other words, if your competitor only holds Japanese patents and you operate in the United States, there may not be a problem.

These charts give insight into where in the world patent families are being filed and held in your dataset.

The families per geography map shows the top patenting countries or continents in your dataset and displays it as a heatmap. It gives insight into where in the world innovation in your technology area is being pursued.

The families per geography pie chart shows the top patenting countries or continents in your dataset and displays it as a pie chart. It gives insight into where in the world innovation in your technology area is being pursued.



This chart shows the top patenting countries or continents in your dataset over time and displays it as a line chart. It gives insight into where in the world innovation in your technology area is being pursued over time.

This chart shows the top patenting countries or continents in your dataset and breaks the total portfolios down by patent status. There are 4 statuses that can be shown for patent families. In-force at least 1 member has been granted and is currently enforceable. Pending all members are still in the application phase. Inactive all members have lapsed. Unknown we have insufficiently reliable data from the patent office to assign a status. It gives insight into where in the world innovation in your technology area is being pursued and breaks the numbers down by status.

Tracking charts show you how tracked values have changed over the selected time period. Select the time period you would like to visualize and the charts update automatically.

This chart shows the difference in counts of pending, in-force, inactive, and unknown status patent families for your selected period.

Shows the top patent publishing organizations in this dataset over your selected period. Rank ordered from most patent families published to least.

Shows which organizations appear in your dataset for the first time during your selected period and displays how many patent families were published. Serves as an easy way to spot organizations that are innovating their way into a new technology area.

This chart shows the top patent-holding organizations in your dataset and breaks down their portfolio by country (or by continent) over your selected period. It gives insight into where the organizations in your dataset are focusing their patenting efforts geographically.

This chart shows in which countries or continents new patent families are being published over the selected period and displays it as a heatmap. It gives insight into where in the world innovation in your dataset is occurring over the selected period.

This chart shows in which countries or continents new patent families are being published over the selected period and displays it as a bar chart. It gives insight into where in the world innovation in your technology area is being pursued over the selected period.

This chart shows which countries or continents are seeing patenting activity for the first time.