Alpaca marketing: how a developer brokerage built a 696 page library of tutorials, case studies and AI trading guides

How Alpaca markets a brokerage API to developers: 279 blog posts, 417 Learn tutorials, partner case studies, MCP and AI guides, and what to copy.

The Convrt teamUpdated 12 min read17 sources
696blog and Learn pages read
417Learn articles
26AI, agent or MCP titles in 2026
The short version

Alpaca is a brokerage that sells through APIs, so its marketing is written for developers. We read all 696 pages in its blog and Learn library: 279 blog posts from January 2018 and 417 Learn articles from June 2018, both running to November 2026.

Learn does the teaching. Trading API is the topic on 279 of the 417 Learn articles, 173 carry a code block, and output peaked at 108 articles in 2021 and 114 in 2022 before the posts got longer instead of more frequent.

The newest work follows AI. Titles naming AI, agents, MCP, LLMs or GPT went from 7 in 2025 to 26 in 2026, with an MCP server launch, Claude and ChatGPT connectors, and guides on keeping an evidence trail for AI orders.

Alpaca does not market a trading app to the public. It sells the plumbing under other apps, and its buyers are developers and fintech founders. So its content is code, tutorials and partner stories. This breakdown covers what Alpaca is, the story by the dates, the two libraries, the Learn topics, the 2021 and 2022 surge, how a Learn tutorial is built, the code in the tutorials, the community writers, the Broker API case studies, the paper trading launch, the turn to AI, the MCP server launch, the MCP or CLI guide, trace before trade, options education, and the publishing rhythm.

What Alpaca is

Alpaca sells to builders. The About page puts it in one line: Alpaca is a brokerage platform providing access to financial markets through developer-friendly APIs. The people who read its content are developers, quant hobbyists and fintech teams, not first time investors.

The same page gives the size. Alpaca says we've helped our partners open over 7 million brokerage accounts across 40+ countries, and that it is supported by over 61,000 monthly active developers. It is also clear on what comes next: We are actively working on new product offerings such as global stocks, index options, and more.

Alpaca at a glance
  1. 017M+Brokerage accounts opened by partners.
  2. 0240+Countries those accounts span.
  3. 0361,000+Monthly active developers.
  4. 04279Blog posts read.
  5. 05417Learn articles read.
  6. 061,292Median words per Learn article.
Alpaca at a glance, from the About page and the blog and Learn library (read 19 Nov 2026).

“Conversation helps define the request, and a command helps repeat it.”Alpaca, Learn guide on MCP or CLI, 5 Nov 2026, read 19 Nov 2026

The story by the dates

The first post was a puzzle. In January 2018 Alpaca ran a coding challenge and reported that fewer than half (39 percent) of those who submitted answers were able to get the correct answer. The winners were moved up a waitlist: Today we’re releasing the emails of those who solved the challenge and are sitting on their royal thrones right at the front of our waitlist. The tone was set early, Makes sense. Alpaca is about algo trading.

The later dates follow the product. December 2018 brought the Paper Trading API, April 2021 the Broker API, November 2025 an official MCP server, and October and November 2026 connectors for ChatGPT and Claude. The newest Learn article in our read is dated 19 November 2026.

The story by the dates
→ 01 Jan 2018 Coding challenge and waitlist
→ 02 Jun 2018 First Learn article
→ 03 Dec 2018 Paper Trading API
→ 04 Apr 2021 Broker API
→ 05 2022 114 Learn articles, the peak
→ 06 2024 15 options guides
→ 07 Nov 2025 Official MCP server
→ 08 Oct 2026 ChatGPT connector
→ 09 Nov 2026 Claude connector
The story by the dates, from the post dates and titles (read 19 Nov 2026).

Two libraries, blog and Learn

Alpaca runs two libraries with two jobs. We read 279 blog posts from 26 January 2018 to 18 November 2026 and 417 Learn articles from 8 June 2018 to 19 November 2026. The blog carries news, launches and partner stories. Learn carries tutorials.

Together they peaked in 2021 and 2022. There were 137 pages in 2021 and 146 in 2022, then 50 in 2023 and 55 in 2024, rising again to 122 in 2026. The blog alone went from 42 posts in 2025 to 91 in 2026.

Pages per year, blog and Learn
2018 36
2019 30
2020 54
2021 137
2022 146
2023 50
2024 55
2025 66
2026 122
Pages per year, blog and Learn together. 2026 runs to 19 Nov (read 19 Nov 2026).

The Learn topics

Learn is organised around the product. Trading API is a topic on 279 articles, How-to Guide on 122, Algorithmic Trading on 96, Community & Partners on 82 and Crypto on 70.

Each topic is a search term a developer would type. Python appears on 40 articles, Options on 33 and Market Data on 26, so a reader looking for one exact task lands on a page built for it.

Learn articles per topic
Trading API 279
How-to Guide 122
Algorithmic Trading 96
Community and Partners 82
Crypto 70
Broker API 50
Python 40
Investing Basics 36
Options 33
Learn articles per topic. An article can carry more than one topic (read 19 Nov 2026).

The surge, then longer posts

Learn grew fast and then changed shape. There were 108 Learn articles in 2021 and 114 in 2022, then 24 in 2023. Crypto alone had 61 articles in 2022.

When the count fell, the length rose. The median Learn article ran 721 words in 2022 and 560 in 2023, then 2,508 in 2024, 3,964 in 2025 and 2,877 in 2026. Fewer, deeper guides replaced many short ones.

Median words per Learn article by year
2018 1,251
2019 1,146
2020 1,501
2021 1,217
2022 721
2023 560
2024 2,508
2025 3,964
2026 2,877
Median words per Learn article by year, counted from the 417 articles (read 19 Nov 2026).

How a Learn tutorial is built

The 0DTE options guide from November 2024 shows the pattern. It opens on a scene: Imagine a trading scenario where decisions and outcomes unfold within hours, sometimes even minutes. Then a box titled Before You Begin lists the guides to read first.

The headings are the questions a reader would search, such as What Are Zero Days To Expiration (0DTE) Options?, What Are the Pros and Cons of Zero Day to Expiration (0DTE) Options? and Is There a List of Securities That Offer 0DTE Options? Each answer starts with a plain definition: A 0DTE option can be defined as any contract that is being traded on the same day that the contract expires.

How a Learn tutorial is built
→ 01 Scene A trade that plays out in hours
→ 02 Before You Begin The guides to read first
→ 03 Definition What a 0DTE option is
→ 04 How it works The mechanics
→ 05 Pros and cons The trade offs
→ 06 Strategies Worked examples
→ 07 Conclusion The short version
How a Learn tutorial is built, from the 0DTE options guide, 21 Nov 2024 (read 19 Nov 2026).
What the 417 Learn articles carry
An image 417
An h2 heading 332
A code block 173
Alpaca code library named 86
A YouTube video 46
An FAQ heading 36
A Before You Begin box 35
What the 417 Learn articles carry, counted from each page (read 19 Nov 2026).
The community writers
  1. 0158Distinct Learn bylines.
  2. 02160Learn articles under the Alpaca Team byline.
  3. 0382Community and Partners articles.
  4. 0444Of them in 2021.
  5. 052018An intern’s robo advisor series.
  6. 06SixStudents behind Agent M, 2026.
The community writers, from the Learn bylines and topics (read 19 Nov 2026).
Broker API case study posts per year
2021 6
2022 4
2023 14
2024 5
2025 18
2026 37
Blog posts tagged Alpaca Broker API Case Study per year (read 19 Nov 2026).
The paper trading launch
  1. 01FreeTo use, open to anyone globally.
  2. 02No accountNo brokerage account needed.
  3. 03Any countryNo need to be a US resident.
  4. 04UnlimitedTesting of algos and bots.
  5. 05Same codeTo switch to live trading.
  6. 06Dec 2018The launch post.
What the Paper Trading API launch post promised, 17 Dec 2018 (read 19 Nov 2026).
AI, agent, MCP, LLM or GPT titles per year
2018 1
2020 3
2021 1
2022 1
2023 1
2024 1
2025 7
2026 26
Pages with AI, agent, MCP, LLM or GPT in the title, per year (read 19 Nov 2026).
Example prompts at launch
→ 01 Portfolio How has my portfolio performed daily this week?
→ 02 Crypto Compare BTC and ETH and their correlation
→ 03 Screen Healthcare large caps and their daily change
→ 04 Visibility Inputs, parameters and results shown
→ 05 Keys Check live or paper first
Example prompts in the MCP server launch post, 20 Nov 2025 (read 19 Nov 2026).
MCP or CLI
  1. 01MCPKeeps data tools close to a conversation.
  2. 02CLIMakes a request explicit in a command.
  3. 03MCP whenThe question is still evolving.
  4. 04CLI whenThe request needs replaying or scheduling.
  5. 05EitherAn agent or application can use both.
  6. 06RuleResponsibility stays with the application.
MCP or CLI, from the Learn guide of 5 Nov 2026 (read 19 Nov 2026).
The five roles in trace before trade
→ 01 Principal On whose behalf it runs
→ 02 Operator Prepares the candidate request
→ 03 Authorizer Records whether policy is met
→ 04 Executor Submits the approved request
→ 05 Verifier Reconciles the outcome
The five roles in trace before trade, from the Learn guide of 9 Nov 2026 (read 19 Nov 2026).
SPX, XSP and SPY options
  1. 01SPXSettles in cash.
  2. 02XSPSettles in cash.
  3. 03SPYSettles in shares.
  4. 04EuropeanSPX and XSP, no early assignment.
  5. 05EarlyA short SPY option can be assigned.
  6. 0610 to 1One SPX is about 10 XSP or 10 SPY.
SPX, XSP and SPY options, from the Learn guide of 19 Nov 2026 (read 19 Nov 2026).
Pages per weekday, 2025 and 2026
Thursday 52
Wednesday 48
Monday 42
Tuesday 37
Friday 7
Saturday 2
Pages per weekday in 2025 and 2026, blog and Learn together (read 19 Nov 2026).

What any trading platform can copy

What any trading platform can copy
  1. 01CodeTeach with working examples.
  2. 02QuestionsOne search question per heading.
  3. 03TakeawaysOpen long guides with the short version.
  4. 04PartnersA case study for every launch.
  5. 05PromptsShow AI features with real prompts.
  6. 06ClipsCut each tutorial for short form.

Alpaca shows how a trading brand can reach developers through content. It teaches with working code, answers one search question per heading, lets students and users write, turns each partner launch into a case study, and shows each AI feature with prompts and plain limits. Even its guides on cost stay practical: Inference optimization begins with deciding how much work to send through the model. Count the whole decision cycle.

The gap is video. Every tutorial, prompt and partner launch in this library is a short clip waiting to be cut, posted on accounts the brand owns and tracked by clicks. That is where Convrt comes in. For an investing platform that explains each step of its product, see our M1 Finance breakdown. For a trading platform blog that began as a founder diary in 2012, see our QuantConnect breakdown.

How we would run it for you Trading platform marketing, run for you

We run creator content and posting for trading platforms on accounts you own, built for your compliance sign-off, with every post tracked for views, link clicks and click-through rate.

Frequently asked questions

Alpaca is a brokerage that offers access to financial markets through APIs for developers and fintech companies. Its About page says it has helped partners open over 7 million brokerage accounts across 40+ countries and is supported by over 61,000 monthly active developers.

We read 696 pages: 279 dated blog posts from 26 January 2018 to 18 November 2026 and 417 Learn articles from 8 June 2018 to 19 November 2026. The busiest year for both together was 2022 with 146 pages, and 2026 has 122 so far.

Learn is Alpaca’s tutorial library. Trading API is a topic on 279 of its 417 articles, How-to Guide on 122 and Algorithmic Trading on 96. The median article runs 1,292 words and 173 articles include a code block.

Through product posts and careful guides. Its November 2025 post launched an official MCP server, 2026 posts added Claude and ChatGPT connectors, and Learn guides compare MCP with a CLI and show how to keep an evidence trail for orders an AI helped prepare. Titles naming AI, agents, MCP, LLMs or GPT rose to 26 in 2026.

Teach with working code, give each guide a short set of key takeaways, let users and students write, publish one case study for each partner launch, use example prompts to show an AI feature, state the limits plainly, and cut every tutorial into short video.

More breakdowns

All breakdowns →

Want distribution done properly?

Apply and we'll map your existing content to real reach on accounts you own.

Pinch to zoom · swipe down to close

Apply to work with us.

Five quick questions, then pick a time.