Get in touch

How do I report a caffeine value I believe is wrong, and what happens next?

Send a single, focused email to [email protected] with the product name, the market or country, the exact serving size you used, the source of your number, and the date that source was published; we will acknowledge receipt and tell you whether we need more evidence. A clear photo of the product label or a link to a manufacturer’s nutrition facts page, a distributor’s specification sheet, or a laboratory report speeds verification and will be requested if needed.

Dataset is the open list of drinks and their published milligram values that TrackCaffeine maintains, and the first thing we check is whether the item and serving you report match an entry in our /dataset/ and in our public sources documented on /about/. Include the market (for example, "US", "UK", or the country printed on the label) because identical product names can have different caffeine per serving in different markets; product market is a required field for every report so we can match the right dataset row.

When you report a value, list the serving size exactly as on the product (for example, 240 ml or 473 ml) because our table rows use serving sizes from original sources and a mismatch in millilitres can cause apparent discrepancies; serving size differences are the most common cause of apparent errors. Also include the date of the label or the web page you used, because manufacturers sometimes change formulations and we keep a date-stamped record of each source in the dataset.

If your evidence is a laboratory measurement, include a short methods note (what analytical technique and whether the sample was prepared hot or cold) because laboratory caffeine measurements vary with extraction and reporting method; we prioritise independent lab data when it is clearly documented. If your evidence is a company statement or a nutrition label, include a full-resolution photo or a direct URL so we can archive the source for reproducibility and citation.

After you submit evidence we will validate it against our internal sourcing checklist and reply with one of three outcomes: accepted, needs more information, or rejected with explanation; the reply will include the dataset row identifier and the sources we used to decide. Typical turnaround for straightforward label/photo corrections is 3 to 7 working days and for laboratory claims it is 7 to 21 days; timelines are estimates and depend on the complexity of verification and working hours at Vast Flow, LLP.

How are corrections handled and published?

We verify your submission against original sources and our sourcing policy, update the dataset row if the evidence supports a change, and publish a short changelog entry that names the source and date; we will not change values without a verifiable, citable source. Corrections are applied to the public /dataset/ and to the specific drink page on the site, and our changelog notes the previous value, the new value, the source, and the date of the correction so readers can follow the provenance.

Correction is an editorial update to a dataset entry that replaces or annotates a value when new verifiable evidence becomes available, and we treat corrections as traceable, dated actions in the public record. Every correction entry names the evidence we used and links back to the archived source where possible so future users can inspect the same materials; transparency of provenance is central to our method described on /about/.

We will use the same published rules to reconcile conflicting sources: prefer manufacturer labelling in the market of purchase, prefer independent laboratory measurements that document method, and prefer up-to-date manufacturer specifications over older web pages. If two reputable sources disagree we publish both values with a note explaining the discrepancy rather than quietly picking one, because our priority is to show the evidence and let users see the conflict.

We document every change in the dataset’s public history and the dataset is versioned so users can cite an exact snapshot; full dataset snapshots and the changelog are available from /dataset/. If a submitted correction requires reclassification (for example changing an item from "brewed" to "coffeeShop") we will explain the taxonomy change in the changelog and link to the methodological note on /about/.

How do I request a drink that is missing from the dataset?

Tell us the exact product name, the market, the serving size you want added, and the source of the caffeine number you prefer we use; we will check whether we already have a matching entry under a different name and then add a new row if it is genuinely missing. If you cannot provide a source, we will attempt to find manufacturer labelling or a reliable third-party source and will email you the result and the provisional entry; we do not add entries based on memory alone.

When requesting a new drink, list the preferred serving in millilitres or the discrete serving unit we should record (for example, 473 ml for a US "Grande" or 350 ml for certain cafe sizes) because our entries use explicit serving sizes from the source material and inconsistency in serving size can lead to major differences in milligrams. If you request a size we do not already track we will either create a new size row using a source from the manufacturer or estimate by scaling a documented concentration (for example mg/100 ml) only when an explicit concentration is published and we will label that entry as estimated and cite the source and the scaling math.

If you ask us to prioritise adding a drink we will consider publisher demand and the public interest; we prioritise widely sold commercial products and recurring user requests such as major coffee chains, prominent energy drinks, or packaged beverages. You can help priority by linking to a reliable public source in your request and by indicating whether the item is sold in multiple markets; items sold in multiple countries often need separate rows in the dataset because the same name can have different caffeine content.

We maintain a public index of drinks on the site under /caffeine-in/ and you can search there first; if you find a similarly named entry, quote that row in your request so we can avoid duplicates. If you request a bespoke home-brewed recipe we will offer guidance on how to measure it yourself, and we may link to our calculators such as the caffeine half-life calculator to help you understand retention and timing.

How can I ask about the CoffeeLog app and what can you help with?

For questions about the CoffeeLog app, email [email protected] describing the app build, the device or platform, and the specific problem; we will classify your request and either answer it, triage to the app team, or point you to a support channel. We can help with dataset integration questions, export formats, and how we map dataset entries to app identifiers, but we cannot debug third-party builds or private forks without access and your permission.

CoffeeLog is the companion app that some users employ to track personal caffeine intake, and app-related questions often overlap with dataset row identifiers and serving sizes that we control. If your question is about how a specific drink from our dataset appears inside CoffeeLog, include the dataset row URL or the drink page (for example Espresso (single)) so we can check mapping and advise on which dataset value the app uses.

We can explain how the app uses our first-order elimination model for decay calculations and we can provide the math we use for local calculations; ask for the model name if you want the precise implementation details or example arithmetic for a drink such as Cold brew at 200 mg. First-order elimination model is a mathematical model where plasma caffeine declines exponentially with a constant half-life, and our implementation uses a population-average half-life of 5.7 hours (342 minutes) so the formula is remaining = dose × 0.5^(hours ÷ 5.7).

For example, using Cold brew (200 mg) from the dataset, remaining after 6 hours is calculated as remaining = 200 mg × 0.5^(6 ÷ 5.7) = 200 mg × 0.5^1.052631... = 200 mg × 0.482 ≈ 96.4 mg; this arithmetic shows the milligram result at each step. If you need a different half-life for personalisation we will describe how to export the dataset and the formula and how to plug in an alternate half-life in third-party tools such as spreadsheets or in-app settings; our public calculators under /tools/ illustrate the same math.

What can you not answer?

We will not provide clinical or medical advice about caffeine, medication interactions, pregnancy, heart symptoms, or any condition-specific guidance; we will always refer medical questions to a qualified health professional. For general safe limits we cite authoritative agencies, for example the FDA’s 400 mg/day guidance for healthy adults is cited directly from the FDA in the same sentence (FDA, https://www.fda.gov/consumers/consumer-updates/spilling-beans-how-much-caffeine-too-much).

We are an informational publisher and not a medical practice; clinicians provide medical advice and diagnoses, and we will decline to interpret symptoms or give treatment guidance. If you ask about pregnancy and caffeine, we will point you to our informational guides such as Caffeine During Pregnancy and to primary sources like the EFSA opinion when relevant, but we will not personalise the guidance or replace clinical advice (EFSA, https://efsa.europa.eu/en/efsajournal/pub/4102).

We cannot provide legal or regulatory advice about shopping, labelling compliance, or how to contest a product labelling decision with authorities; for regulatory questions you should contact the relevant food authority directly. If you submit a legal takedown request or a court order regarding data we will route it to our legal contact at Vast Flow, LLP and will record the request in our process log; see our terms of use and privacy policy for procedural details.

How do press and data-reuse enquiries work, and do I need permission to use the dataset?

Press enquiries should email [email protected] with publication name, deadline, and a short summary of what you want to reuse; we will provide a media pack, a statement of authorship, and high-resolution assets when available. The dataset itself is published under CC BY 4.0, and you do not need to ask permission to reuse the raw dataset beyond giving attribution as required by the licence (Creative Commons, https://creativecommons.org/licenses/by/4.0/).

CC BY 4.0 is the dataset licence we use, and it permits copying, redistribution, remixing and adaptation provided you credit TrackCaffeine and the original dataset version; the licence text is at the Creative Commons link above. When reusing the dataset in a commercial product please include an attribution line that names the dataset and links to /dataset/ and, if you publish derivative data, include a note of any methodological changes such as scaling or reclassification.

For syndicated feeds or automated scraping, contact us first so we can provide a stable export or API plan; unrestricted scraping can overload the site and slows verification work, and we will prioritise requests that use an approved export. If you want the dataset in a custom format we can often provide a CSV snapshot of a specific subset (for example all Starbucks entries such as Starbucks Pike Place Brewed (Grande)) and we will note the dataset version and the licence to include in your attribution.

How do I reach you about privacy or data requests?

Privacy and data-subject requests should be sent to [email protected] with a clear subject line such as "Privacy request" and include the email address you used on the site; we will respond within the timeframe specified in our privacy policy. If your request involves deletion, export, or correction of personal data, include a copy of a government photo ID to verify identity when required and reference the relevant section of the policy so we can process it promptly.

We publish a detailed privacy process in /privacy/, and that page lists the data we collect, retention periods, and how to exercise your rights under applicable privacy law; consult that page for specifics and cite the section when you email. For formal legal or privacy team correspondence directed to our publisher name use "Vast Flow, LLP" and include a named contact and postal address; we will acknowledge service within three business days and escalate to our legal and privacy leads as required by the policy.

If you prefer a contact by post, write to Vast Flow, LLP with your request and the dataset entry identifiers if relevant; postal notifications are accepted but will follow the same internal timelines as email. For rapid answers about how we use identifiers in the dataset, see /about/ and the public dataset row pages such as Drip coffee and Espresso (double) which illustrate our citation format and source listing.

Publisher contact: TrackCaffeine is published by Vast Flow, LLP and the primary support address is [email protected]; include links or photos to help us handle dataset requests. We try to answer all reasonable queries; if your request exceeds our scope we will explain why and direct you to the right authority or resource such as the FDA guidance on safe daily intake (FDA, https://www.fda.gov/consumers/consumer-updates/spilling-beans-how-much-caffeine-too-much) or the Mayo Clinic caffeine chart for reference (Mayo Clinic, https://www.mayoclinic.org/healthy-lifestyle/nutrition-and-healthy-eating/in-depth/caffeine/art-20045678).