A topic marked Exploding can be worth investigating. It is not, by itself, a reason to change your content plan or launch a product. Before acting on that label, look at the numbers, time period and source that produced it.
Shane’s Topics organizes search-interest or volume readings you enter or import. It calculates change, draws a small chart and helps turn a topic into a research outline. The labels describe your recorded values. The tool does not fetch live search volume or predict future demand.
Understanding its calculation makes the dashboard more useful, especially when a dramatic percentage comes from a small starting value.
The calculation uses the first and last readings
For a nonzero starting value, total change is the difference between the latest and first readings, divided by the first reading, multiplied by 100.
Suppose you enter the fictional sequence 12, 18, 27. The last value is 15 higher than the first. Dividing 15 by 12 and multiplying by 100 gives 125%, so the tool displays +125% and marks the topic Rising.
The middle reading helps show the path on the chart, but it does not change that first-to-last calculation. Neither the growth label nor the percentage tells you why the values moved.
Know the label boundaries
The current implementation uses these thresholds:
- Exploding: Growth of 500% or more.
- Rising: Growth of at least 50%, but less than 500%.
- Stable: Change from −15% up to, but not including, +50%.
- Cooling: A decline greater than 15%.
Those names are categories for observed change, not a statistical test. A topic that rises 49% is still labeled Stable, while one that rises 50% is Rising. The one-percentage-point difference does not create a sudden change in the topic’s business value.
The displayed percentage is rounded to a whole number, while the label uses the unrounded calculation. Near a boundary, check the original readings rather than relying on the rounded figure alone.
At the lower boundary, a move from 100 to 85 is −15% and remains Stable. A move from 100 to 84 is −16% and becomes Cooling. Reviewing the underlying values is more informative than treating each label as a decision rule.
A small starting value can create a large percentage
Consider two invented examples measured in the same units over the same period:
Topic A moves from 1 to 6. Its increase is 5 units, but its percentage change is 500%, so it receives the Exploding label.
Topic B moves from 100 to 350. Its increase is 250 units, while its percentage change is 250%, so it receives the Rising label.
Topic A has the larger percentage. Topic B has the larger absolute increase. Neither fact alone tells you which subject deserves an article, investment of time or new offer.
Check the size of the starting value, the actual readings and how closely the topic matches your customers’ needs. These examples illustrate arithmetic; they are not observations of real search trends.
The same label can hide a different shape
The sequences 20, 25, 30 and 20, 80, 30 both finish 50% above where they started. Both are labeled Rising.
The first sequence increases across each observation. The second contains a spike followed by a decline. A first-to-last percentage cannot distinguish those paths, so inspect the chart and return to the original data when the shape raises questions.
A longer series can provide more context, but more points are useful only when they represent comparable observations. Do not add guessed values just to make a chart look complete, and do not enter zero as a substitute for an unknown reading.
Handle zero baselines separately
Percentage growth from zero does not have the same calculation as growth from a positive starting value. Shane’s Topics marks a sequence that starts at zero and ends above zero as New data, with New shown in place of a percentage.
For example, 0, 5, 10 is New data rather than a very large percentage increase. A sequence that starts and ends at zero receives 0% and the Stable label, even if an intermediate reading is higher.
That is another reason to inspect the full series. The label summarizes the endpoints; it does not summarize every movement between them.
Keep the comparison fair
Before ranking topics against each other, check four things:
- Source and units. A relative-interest score and a count of searches are different measurements. Do not compare their raw values as though they were interchangeable.
- Geography. Keep regions consistent or make the difference explicit in the research question.
- Time period. A three-month change and a three-year change cover different windows. Use comparable start and end periods when comparing growth.
- Observation order. Enter periods and values oldest to newest, with one reading for each period. The calculation uses their supplied order rather than rearranging the series chronologically.
The Sort by latest volume option orders topics by the last number you provided. It does not convert different sources into a common measurement. Use it within a comparable group of records, and retain source details in your notes.
Turn the signal into a research question
Choose Create brief on a topic card to open a research outline that includes the observed change, period and source. Use it to investigate what might explain the movement: a seasonal pattern, a short-lived event, a changing customer question or a collection difference.
The generated outline is a starting structure. Check the original evidence, look for information that challenges your first interpretation and decide whether you can offer something useful to the audience.
Open Shane’s Topics with a small set of comparable readings. Read the label, inspect the curve and write down the question you need to answer before acting.