Serca Review (2026): Pricing, Features & Honest Verdict

Serca is an AI workspace for video creators that keeps scripts, references and channel memory in one composition. Good ideas, unusually honest positioning, and not a single published usage limit anywhere on the site. Best for solo YouTube and short-form creators. Price: free forever, or $14/mo for Creator Pro. Rating: 7/10.

What is Serca?

Serca is a planning and scripting workspace built for people who publish video, from a two-person team of Shivam Singh and Tejesh Bhalla. It launched on Product Hunt in July 2026. The core unit is a “composition,” which holds everything for one video in one place: the hook, the script, the b-roll list, the references and the thumbnail brief. An AI assistant sits inside that composition rather than in a separate chat window, so when you ask it for a better cold open it already knows what the video is.

The pitch it makes against every other AI content tool is worth quoting because it is the opposite of the usual one. “AI can’t make videos for you. And it shouldn’t,” reads one section heading, followed by “end-to-end AI content isn’t the future, the world doesn’t need more of it.” Serca positions itself as handling the optimisation work, meaning hooks, structure and retention, while you stay responsible for the creative. In a category currently racing to generate finished slop, taking the opposite position is both a differentiator and a defensible product decision.

What are Serca’s key features?

Compositions instead of six open tabs

Everything for one video lives in a single composition, and you can hold a whole series in one project. You can flip between a linear document view and a board view on the same source material with one click, which suits the way scripting actually goes: brain-dump in a mess, then organise without retyping. Focus mode strips the interface down to the writing pad. The AI is reachable inline, on the board and as a side tab, so you are not context-switching to ask a question about the thing in front of you.

Retention-first AI

Serca says its model is trained on high-retention videos and built around retention specifically rather than general writing quality. The practical output of that is format awareness: it argues a Shorts hook and a long-form hook are different problems, and that a vlog cold open and a video essay cold open are not interchangeable, so it writes for the runtime, niche and platform each time. Whether the training claim holds up is a separate question we get to below, but the format-specific behaviour is a sensible thing to build for.

The Vault

Paste a link to a video you admire and Serca breaks it down: hook formula, retention beats, transitions and CTA insight, laid out minute by minute. You can then pin that analysis to a composition and ask the AI to base your outline on it, and the marketing is careful to say it actually does that rather than gesturing at it. Every analysis you run also joins a reference shelf the assistant draws on later, so the tenth breakdown improves suggestions more than the first did. This is the most useful feature in the product and the hardest for a general-purpose chatbot to replicate.

Memory that learns your channel

Serca builds a profile of how you write and applies it to every suggestion. The demo shows three learned traits: a style note that you cut qualifiers and tighten middles, a hook preference for curiosity gaps containing a specific number, with a record that you rejected every list-style hook offered, and a pacing note that you shift beats every 38 to 55 seconds and lose retention past 70, sourced from six of your strongest videos. Whether it reaches that level of specificity on a real channel is something only your own usage will tell you, but the design intent is right: the value compounds as you use it.

An assistant that sees the work

The distinction Serca draws between its AI chat and a general model is context. Ask a chatbot for a hook and it starts from nothing. Ask Serca and it has the composition, the pinned reference video, your rejected hooks and your pacing profile. In the demo, the assistant reads a note buried on line 14 of an outline and pulls it forward into the hook. That is the sort of thing that only works when the AI and the document are the same surface.

How much does Serca cost?

Serca has a free tier and one paid tier. The free plan is described as free forever with monthly limits, and it includes the basic script workspace, some AI help, limited Vault saves and Memory usage. There is no trial countdown, which the site makes a point of. Creator Pro is $14 a month and unlocks the full workspace: scripts, Vault, Memory, web research, AI chat and completions, with higher limits across the board.

The $14 is a discounted price. The pricing card shows $19 struck through with $14 beside it, labelled “for limited time only.” No end date is given, and no counter runs down. Everywhere else on the site, including the hero subtitle, the closing call to action and the footer, $14 is stated as the price with no mention of $19. Treat $19 as the number you may eventually pay and $14 as a launch rate.

Against the field this is cheap. Taja AI’s pricing page publishes $19, $49 and $99 tiers. vidIQ’s plans page renders a free tier alongside $39 a month for Boost and $120 for one-on-one coaching. Serca undercuts both, though it is also doing a narrower job than either. There is no annual option, no team plan and no seat pricing, so a two-person channel buys two subscriptions. Note also that serca.io/pricing returns a 404; pricing lives as a section on the homepage only.

What are Serca’s usage limits?

Nobody knows, and that is the single biggest problem with the product as it is currently sold.

Read the free plan and you get “limited scripts and AI usage each month,” “basic workspace, Vault, and Memory access,” and “limited Vault saves and Memory usage.” Read Creator Pro and you get “full Creator Pro limits,” “more scripts, Vault saves, AI chat, and completions” and “higher limits for creators shipping consistently.” Every one of those is a comparative with nothing to compare against. There is no number anywhere on the page: not a script count, not a Vault save count, not a token or credit allowance, not a monthly cap of any kind.

We have written up several products recently that price in a unit they never define. Serca goes one step further and does not name a unit at all. That leaves a buyer unable to answer the only question that matters at the point of upgrade, which is whether $14 buys enough for the way they work. If you publish three videos a week, “more scripts” does not tell you whether you will hit a wall on day nine. The workaround is simple but annoying: run the free tier until you hit its ceiling, note what the ceiling was, and only then decide whether the paid multiple is worth $14. Do not upgrade on the strength of the word “more.”

Can you verify Serca’s retention claims?

No, and you should read them accordingly. The homepage leads with “+162% retention forecast” on a demo draft, shows a “retention forecast, 38%, same as your last 9 cold opens” widget, and elsewhere claims the model is trained on “thousands of viral videos” and “millions of high-retention minutes” across YouTube, TikTok, Shorts and Reels. Further down there is a before-and-after comparison of “six tools open, zero scripts done” against “one composition, three scripts shipped,” with a 45-minute figure and a 1.6x multiplier attached.

None of these come with a methodology, a sample size, a date range or a source, and the worked example running through the page is a MrBeast-style hotel video used as illustration rather than customer data. A retention forecast is a very hard prediction problem, and a percentage presented without any statement of how it was measured is a marketing number rather than an analytical one.

This is not a reason to avoid Serca, and the underlying product does not depend on the forecast being accurate. But if the retention percentage is the thing selling you, be aware you are buying an unaudited claim from a two-person team that launched six weeks ago. The features you can actually verify, which are the composition model, the Vault breakdowns and the memory profile, are the ones worth paying for.

Who is Serca best for?

Use Serca if you are a solo creator or a two-person channel who scripts before shooting, publishes often enough that the workflow is the bottleneck, and currently has that workflow spread across Notion, a notes app, a spreadsheet of reference links and a chatbot. The Vault breakdown feature alone is worth the free tier, and $14 is a low enough number that the decision does not need a spreadsheet. The anti-slop positioning is also a real filter: if you want a tool that writes the video for you, this is deliberately not it.

Skip Serca if you need team collaboration, since there are no seats and no shared workspace pricing. Skip it if you need to know your limits before you commit, because the site will not tell you. Skip it if you shoot unscripted, since nearly all the value sits upstream of the camera. And be realistic about the risk profile: two makers, six weeks live, no named customers and no annual plan to lock in the current price.

What are the best Serca alternatives?

vidIQ is the established option for YouTube specifically. Its plans page renders a free tier, a $39 a month Boost plan and $120 a month for one-on-one coaching, alongside enterprise. It is stronger on keyword research, posting times and channel analytics, and much weaker on the actual writing surface. Pick it if you want data about your channel rather than a place to write.

Taja AI is closer to Serca’s territory, publishing tiers at $19, $49 and $99 a month. It leans toward packaging and metadata, titles, descriptions and thumbnails, where Serca leans toward the script and the structure. If your problem is that your videos are good and nobody clicks them, Taja is aimed at your problem rather than Serca’s.

TubeBuddy remains the incumbent channel management toolkit and covers bulk operations, A/B testing and optimisation across a large feature surface. Its pricing renders client-side and we could not read a figure off the page, so check it directly. It is the broadest of the three and the least focused on writing.

Final verdict: is Serca worth it?

Serca gets the shape of the problem right. Scripting a video is not a chat problem, it is a document problem with references attached, and building the assistant into the composition rather than beside it is the correct call. The Vault, which turns a video you admire into a minute-by-minute structural breakdown you can pin to your own outline, is the feature we would actually use. And refusing to generate finished video in a market rushing to do exactly that is a position with some spine behind it.

What holds it back is not the software, it is the sales page. Every limit is a comparative, the paid price is a discount with no stated expiry, the headline retention figures carry no methodology, and there is no annual plan to lock anything in. The free tier is the answer to all of it: it is free forever rather than a timed trial, so you can find the ceiling yourself and decide from evidence rather than adjectives. Start there, run a real video through a composition, and upgrade only once you know what you are actually buying.