Once the first few foods are recorded, it is easy to wonder whether your baby's list is growing quickly enough. Other parents seem to have a spreadsheet, a freezer full of tiny portions, and somehow three kinds of lentils in rotation.
We measured cumulative distinct foods among eligible early Tummi trackers. The resulting percentiles are benchmarks for app activity, not a score your baby needs to match.
By month 8, the median profile contained 33 recorded foods, while the middle half ranged from 20 to 55. That broad span is the most useful part of the result. It shows how different recorded menus can look, even among families at the same broad feeding stage.
Key Findings
- The median rose from 5 recorded foods at month 6 to 21 at month 7.
- By month 8, the median reached 33 recorded foods.
- The middle half at month 8 ranged from 20 to 55 foods.
- A distinct food record shows presence in a profile, not acceptance or repeat exposure.
- Cumulative food counts differ from both daily food-group diversity and clinical feeding targets.
How quickly does recorded variety grow from month 6 to month 8?
The median early tracker had 5 distinct foods recorded at month 6, 21 at month 7, and 33 at month 8. The progression captures a period when many participating families were adding foods and building a larger recorded menu.
Those values describe the middle profile at each milestone, not a schedule every baby followed. A rising median can tell us that cumulative records grew across the cohort, but it cannot tell us how often each food was served or how much any baby ate.
Source: Aggregated Tummi product data. Early-tracker cohort. Later milestones are withheld when the eligible group becomes too small.
What does the middle 50 percent tell us?
At month 6, the middle half of eligible profiles contained 2 to 12 recorded foods. That range moved to 11 to 35 at month 7, then 20 to 55 at month 8. The median sits inside each band, but it does not represent every family.
The middle range is broader in the older published age cells. At each milestone, some eligible profiles have longer recorded menus and others have shorter ones. Because the age groups are not documented here as one continuous cohort, the analysis does not establish individual growth paths.
What does a distinct food record actually measure?
A food enters the cumulative count when it appears on an eligible baby's recorded menu. Once that food has appeared, another record for the same food does not create another distinct item. The measure is therefore about the breadth of the recorded list.
That list does not show whether your baby accepted the food, swallowed a full portion, tasted a tiny amount, or met it again later. It also cannot distinguish a food that was never offered from one that was offered but never entered into Tummi.
Should 33 foods become a goal for every 8-month-old?
No. The median of 33 is a description of eligible early Tummi trackers, not a required total. Even within the middle half of those profiles, month 8 counts ranged from 20 to 55, so the data itself does not point to one universal pace.
The analysis also does not connect a particular count with growth, nutrition, feeding skill, or another health outcome. If you are concerned about your baby's intake or development, your pediatrician can consider those questions with far more context than a cumulative list provides.
How is this different from the WHO dietary diversity indicator?
The World Health Organization minimum dietary diversity indicator considers food groups consumed during the previous day. Tummi's variety measure instead counts distinct individual foods accumulated across a baby's recorded menu. The measures use different units and answer different questions.
A cumulative list may grow each time a new food is recorded, even if the recent menu remains small. Daily food-group diversity can change without adding anything new to the lifetime list. One result therefore cannot be translated into the other.
Why does repeated exposure matter when interpreting the count?
The USDA repeated food exposure review examines how the timing, quantity, and frequency of exposure relate to early food acceptance. Its scope highlights an important boundary here: counting distinct records does not measure how often a baby encounters the same food.
A profile can add new foods quickly without recording many repeat servings. Another profile can repeat familiar foods regularly while adding fewer new entries. Those histories may produce different variety counts, but this analysis cannot tell us which foods became familiar or accepted.
Methodology and limitations
The analysis uses the existing early-tracker cohort and counts unique foods cumulatively at the end of months 6, 7, and 8. It follows caregiver-entered eating records rather than surveying a representative sample of all babies.
For each age, the chart reports the 25th percentile, median, and 75th percentile. The median divides the eligible profiles in half, while the outer two values define the middle 50 percent of recorded totals.
Each unique food contributes once to the cumulative count. Repeat records for the same food do not increase the number, and no weighting is applied for portion size, frequency, preparation, acceptance, or the amount swallowed.
Month 8 is the latest publishable milestone in this cohort. Later eligible groups were too small for a stable benchmark, so the analysis stops rather than extending the curve or estimating future totals.
Tummi records are caregiver-entered product data from families who chose to use the app. Missing records cannot be interpreted as foods that were never served. The WHO indicator and USDA review provide outside context, but neither converts these app percentiles into clinical thresholds.
The dataset is self-selected because it describes families who chose to use Tummi. The findings describe aggregated app activity and are not population estimates for every family, baby, or country.
Everything shown here is aggregated. No names, individual profiles, free-text answers, or personal meal histories are published.
Conclusion
Among eligible early Tummi trackers, the median cumulative list reached 33 recorded foods by month 8. The middle half ranged from 20 to 55, making variation a central part of the result rather than a footnote.
These numbers describe how recorded menus expanded. They do not measure acceptance, repeated exposure, daily food-group diversity, nutritional adequacy, or the amount a baby ate.
Use the chart as orientation, not a report card. If your baby's feeding, growth, or intake is worrying you, a pediatrician can help interpret the full picture without reducing it to one number.
Frequently Asked Questions
How many foods should a baby have tried by 8 months?
This report does not set a target. It found a median of 33 recorded foods among eligible early Tummi trackers, with a middle range from 20 to 55.
Does a recorded food count as accepted?
No. A record shows that a caregiver entered the food. It does not show how much was swallowed, whether the baby enjoyed it, or whether it was served again.
Why might two profiles have very different totals?
The data cannot identify one cause. Feeding choices, recording habits, repeat servings, and missing entries can all shape a recorded menu.
Why does the analysis stop at month 8?
Later eligible groups were too small for a stable benchmark. The report stops at the latest milestone supported by the available cohort.
Sources
- Tummi Food Team, 2026. Aggregated Tummi feeding records and onboarding responses. First-party product data.
- WHO minimum dietary diversity indicator. A food-group measure based on the prior day. It is different from a cumulative count of individual foods.
- USDA repeated food exposure review. A systematic review of repeated exposure and food acceptance in young children.