Blog › Redbubble guides Browse all Redbubble articles →

Redbubble keyword research: a search-bar method that needs no volume data

Redbubble keyword research, done honestly, is manual: you type candidate phrases into Redbubble's own search field, read what comes back, write it down, and repeat. There is no search-volume figure anywhere on this page, because Octozia holds no search-volume data for Redbubble or any other marketplace. The method below is built to work without one.

Start by accepting you have no volume number

Most keyword guides open with a number. If you are shown one and told it is Redbubble search volume, treat it as an estimate until whoever is showing it says where it came from. Octozia's tools — a bulk uploader, a free tag generator, a trademark pre-check — report no search volume, no sales rank, no demand metric of any kind.

What you can observe first-hand costs nothing but time: the wording that recurs across listings on a results page, whatever the search field offers while you type, and the kind of products a query surfaces. Recorded consistently and dated, those beat a precise-looking number of unknown origin.

The search-bar method, step by step

  1. Write five seeds. A seed is the plain-language thing your design is of, in a buyer's words rather than a designer's: "roller derby cat", not "dynamic feline linework".
  2. Expand each seed letter by letter. Type the seed, a space, then each letter of the alphabet in turn — 26 probes per seed, 130 across five seeds. Watch the field as you type and copy any completion it offers verbatim: exact spelling, exact word order, plurals included. If it offers nothing, that is a result too.
  3. Run the search and read the titles. Press enter and read the first screen of results. Write down the words that repeat across designs by different sellers. Repetition between independent sellers is weak evidence that those are the words people type — weak, but first-hand and free.
  4. Note what the page is full of. Apparel, stickers, wall art? The product mix behind a phrase tells you what the demand is for, which usually matters more than how big it is.
  5. Look at the controls rather than assuming them. If the results page offers a way to sort or filter, record what the options are called and what visibly changes when you apply one. Do not assume a control exists because a guide said so.
  6. Log every phrase with a date. One row per phrase: where it came from, its character count, the product types you saw, and the date. The date is the part everyone skips and the part that gives the exercise value.
  7. Re-run the same seeds later and compare. Two dated snapshots of five seeds say more about movement than one undated snapshot of fifty.

Turning the phrase list into 15 tags

Redbubble's help centre states the limit plainly: "You can use up to 15 tags, with 50 characters each" (help.redbubble.com, page updated 22 January 2026). Fifty characters is the ceiling on each individual tag, not a pooled budget you can spend on one long phrase. In practice the slot count is the constraint you feel first. Planning is therefore an allocation problem: how do you spend fifteen slots without burning twelve on synonyms for one noun?

A starting allocation, where the five groups add up to 4 + 4 + 3 + 2 + 2 = 15:

Group Slots What goes in it
Literal subject4What the design depicts, in the words that recurred in listing titles
Buyer or gift framing4Who is buying it and why: the recipient, the hobby, the relationship
Style or medium3Visual descriptors you actually saw used, not ones you invented
Occasion or context2Only if a real one exists; leave these empty rather than padding
Verbatim completions2Any phrasing the search field completed for you, copied exactly

Treat that as a way to keep your slots diverse, not as a ranking formula. The help page quoted above gives counts and character limits, not weighting, and this article is not going to guess.

What this method cannot tell you

  • How much demand exists. A crowded results page might be a durable market or a collapsing fad. It looks identical either way.
  • Whether a phrase is protected. A free trademark search exists at tmsearch.uspto.gov, and running your phrase through it before you publish costs nothing. None of this is legal advice, and a clear search result is not permission — if real money depends on the answer, ask a lawyer. Octozia's trademark pre-check ($6.99/month, or $49.99 lifetime) automates the lookup, not the advice.
  • What you keep. Ranking is not the last step. Redbubble's platform-fee help page (updated 31 July 2026) sets the platform fee on monthly earnings at 50% for Standard, 20% for Premium and none for Pro, with Standard and Premium capped at $150 / 150 EUR / 150 GBP per payment period, and a separate excess-markup fee above a 20% product markup.
  • Anything about the design. Octozia has no cover design, no listing-copy generation and no sales-rank data. Keyword work will not rescue art nobody wants.

Does the same phrase list work on TeePublic?

The list transfers; the rules do not. TeePublic's own tagging guide says to "Use 6-15 total tags", that main tags "are weighted heavily in search", and that you should not add redundant tags because "Repeating the same word will hurt your search results". It also describes tags rendering blue or grey, where grey ones are "too specific and do not have the same power" and may turn blue once indexed.

So on TeePublic you can legitimately stop at six strong tags, and you get a visible colour signal to react to. Treat that signal as TeePublic-specific. The seed list and the search-bar discipline are the reusable part; there is a fuller walkthrough in TeePublic keyword research.

Where Octozia fits, and where it does not

The Redbubble tag generator is free and needs no account. It expands a seed into tag candidates so you edit a list instead of staring at an empty field. It does not know what sells, because we do not have that data. It saves typing, not judgement.

The Redbubble bulk uploader is a Chrome extension that pushes a prepared tag set across a batch of uploads — the point at which a researched list stops being a spreadsheet and becomes listings. Pro is $6.99/month or $69.99/year, with a $49.99 lifetime licence. The annual plan loses to lifetime on price alone: $69.99 is more than $49.99, so there is no case for it. Against monthly, lifetime breaks even at $49.99 / $6.99 = 7.15 months — seven months of monthly is $6.99 x 7 = $48.93, $1.06 short of lifetime, so month eight is where lifetime wins.

The trial is 3 days and requires a card, and it applies to a first subscription only: if you have subscribed before, you are charged at checkout rather than trialled again. Sales are final once billed — there is no refund window, so decide during the trial.

One caveat you should hear from us. Redbubble spells this out directly in its Community and Content Guidelines, last revised 17 August 2026: using a bot, scraper, or any other automated method to upload content without the company's written sign-off is listed among the examples of a policy violation that may warrant account deletion, and the guidelines say such an account is "immediately and permanently disabled". The same page sets a hard ceiling on volume: "you may upload up to 30 works per day to your Redbubble account", counted across every account you operate, and says that creating extra accounts to exceed it gets all of them closed. No enforcement action naming a form-filling extension has been published as far as we can tell, and that absence is weak evidence rather than clearance. It is your account and your call; the longer version of the argument is in the extension comparison.

FAQ

What is the search volume for a Redbubble keyword?

We cannot tell you: Octozia holds no search-volume data for Redbubble. If another source gives you a number, ask how it was measured before you plan a quarter of uploads around it.

How many tags does Redbubble allow?

Per the help centre's most recent revision, dated 22 January 2026, you get fifteen tag slots total, and each individual tag can run up to fifty characters. The fifty-character limit applies to each tag separately, so the number of slots is usually the binding constraint rather than the length of any one tag.

Does the free tag generator replace the search-bar method?

No. It expands a seed into candidates quickly; the search bar is where you check whether those candidates match what people type. Generate, then verify, then cut the list down to your fifteen slots.

How often should I repeat this?

We have no verified cadence to recommend. Consistency matters more than frequency: keep the same five seeds, date every row, and re-run on whatever schedule you will actually keep. The comparison between two dated passes is the finding.