ProgrammableMarketing/mongodb.com (opens in a new tab)
PDF

Views

  • Plan
  • Roadmap
  • Research

On this page

  • The brief
  • Start here
  • Idea portfolio
  • Effort vs impact
  • This week · 6
  • This quarter · 10
  • Big bets · 3
  • Wildcards · 2
Generated Oct 22, 2026
  1. Programmable
  2. Playbooks
  3. MongoDB
MongoDB logo

growth playbook · Oct 22, 2026

mongodb.com (opens in a new tab)

MongoDB growth playbook

Diagnosis

MongoDB is hiring to win AI builders: a growth marketer for Voyage AI adoption and cross-sell, a growth product leader for the self-serve journey from Community and Atlas to sales, a VP of AI Marketing to own how MongoDB appears in AI answers and agent frameworks, marketing automation, and new field hubs in Mexico City, Paris and India.

Its agent-readable docs are already ahead of most. The gaps are in the joins: voyageai.com still signs new developers into a separate Voyage account, two official Voyage doc sets give different setup steps, the homepage fails Core Web Vitals for real visitors, and there are no comparisons for the vector and embedding choices builders make today.

This playbook fixes the front door first, then builds cross-sell, agent attribution, product-qualified handoffs and partner defaults.

Growth Score

8/10

Biggest Opportunity

Turn Voyage AI into an Atlas on-ramp: one front door, one set of docs and cross-sell into Vector Search, measured from first embedding to production cluster.

Biggest Risk

Builders default to dedicated vector databases or Postgres with vector extensions (Pinecone, Supabase) before they ever compare, and agents copy whichever setup instructions they find first.

01

Start here

The three moves to make first. Each one is written up in full below.

  1. 1

    Give Voyage AI one front door: voyageai.com still signs developers up somewhere other than Atlas

    Count last month's new Voyage dashboard signups versus Atlas signups with the Voyage onboarding scenario.

    High impact First result in 2 weeks
  2. 2

    Pick one canonical home for Voyage docs and pricing, so AI assistants cite one answer

    Ask three coding assistants "How do I get a Voyage AI API key?" and record which instructions they give.

    High impact First result in 1 quarter
  3. 3

    Cut the homepage from 13 MB: real visitors fail Core Web Vitals on the page that starts every signup

    List every font file and weight the homepage loads.

    High impact First result in 1 month

02

Your idea portfolio

21 ideas, grouped by when to run them. Open any idea for how to run it: the steps, who to approach, copy you can use, and how to measure it.

  • 6 ideas are marked Unexpected: plays you are unlikely to have tried.

Every idea placed by the effort it takes against the impact it could have. Quick wins sit top left; select a number to open that idea. The filters below narrow the list by status, timing, category and what Programmable can build.

Effort vs impactEach number is an idea. Select one to open it.
Impact →
Quick wins · 1Big bets · 11Fill-ins · 5Reconsider · 4
Effort →

Showing 21 of 21 ideas

This week 6

Quick wins with a clear first action.

  • 01

    MongoDB's own Voyage AI quick start says to create the model API key in the Atlas UI, and mongodb.com's Voyage page sends "Get Started" to Atlas with a Voyage onboarding scenario. But voyageai.com, titled "Voyage AI by MongoDB", sends both "Get Started" and "Sign in" to a separate Voyage dashboard login, and docs.voyageai.com tells developers to create a key there. A developer who starts at the older brand site lands in an account that is not Atlas, so the new Growth Marketing Manager cannot run the cross-sell into Atlas and Vector Search that the role is hired for.

    High impactMedium effort First result in 2 weeks Growth Marketing Manager

    First action

    Count last month's new Voyage dashboard signups versus Atlas signups with the Voyage onboarding scenario.

    How to measure it

    New Voyage developers starting in Atlas

    Now
    –
    Target
    90%
    By
    90 days

    Steps

    1. 1

      Size the split

      New Voyage dashboard accounts versus Atlas Voyage signups over the last 90 days, and how many of each later created a cluster.

      Marketing Analytics 3 days

    2. 2

      Agree the rule

      New developers go to Atlas; existing dashboard users keep signing in where they are.

      Growth Marketing Manager, Voyage AI 2 days

    3. 3

      Change the buttons

      "Get Started" on voyageai.com to Atlas with the Voyage onboarding scenario and a source tag.

      Web Engineering Lead 2 days

    4. 4

      Match the docs

      docs.voyageai.com key page points new users to the Atlas key path first.

      Documentation Lead 1 week

    5. 5

      Offer the move

      Guided migration for dashboard accounts that keeps their free tokens and keys.

      Senior Director Product Management, Growth 1 month

    6. 6

      Start the lifecycle

      Atlas-native onboarding emails from first key to first embedding to first vector index.

      Growth Marketing Manager, Voyage AI 2 weeks

    7. 7

      Read the result

      Activation and Atlas cluster creation by front door, weekly.

      Marketing Analytics weekly

    First result
    2 weeks
    Cost
    Engineering and docs time
    Owner
    Growth Marketing Manager, Voyage AI

    Smallest test

    Change the voyageai.com "Get Started" button for half of visitors for two weeks and compare first-embedding rates.

    Why it works

    Point voyageai.com's "Get Started" at Atlas registration with the Voyage onboarding scenario (keep "Sign in" for existing dashboard users), add a short "Already on Atlas? Create a model API key" path, and tag every entry with its source. For existing dashboard accounts, offer a guided move that keeps their free-token balance and key, then measure activation (first embedding call) and Atlas cluster creation by front door.

    Every new Voyage developer then starts inside Atlas, where in-product prompts, lifecycle email and Vector Search are one click away, and the funnel is measured in one place.

    Ready to use

    ChecklistVoyage front door checklist

    [ ] voyageai.com Get Started -> Atlas register (onboardingScenario=voyage, source tag)
    [ ] "Sign in" kept for existing dashboard users
    [ ] docs.voyageai.com key page: Atlas path first
    [ ] Pricing pages say the same free-token allowance
    [ ] Migration offer for dashboard accounts
    [ ] One activation event (first embedding) tracked for both paths
    [ ] Weekly report by front door

    Risks and how to handle them

    • Existing Voyage customers feel pushed

      Keep their login and keys working; make the move optional and keep their balance.

    • Standalone API users do not want a database account

      Say on the page that no cluster is needed to use the models.

    Evidence

    • Voyage AI websitevoyageai.com (checked October 21, 2026) is titled "Voyage AI by MongoDB" and is its own canonical site. Its "Get Started" and "Sign in" buttons go to dash.voyageai.com, a separate Voyage account login, and its docs links go to docs.voyageai.com. It loads its own email marketing script.Source
    • Your websitemongodb.com's Voyage AI page sends "Get Started" to Atlas registration with onboardingScenario=voyage, and the MongoDB docs quick start says: "To access Voyage AI models, create a model API key in the MongoDB Atlas UI", under "AI Model APIs" in an Atlas project.Source
    • Voyage AI docsdocs.voyageai.com is a second, self-canonical documentation set with its own llms.txt, its own pricing page and its own key instructions: "sign in with your Voyage AI account and click the 'Create new secret key' button" in the Voyage dashboard. Its pricing gives "the first 200 million tokens" free for voyage-4 models "for every account".Source
    • HiringGrowth Marketing Manager, Voyage AI (two openings, United States and Toronto, posted October 8, 2026): drives "growth and adoption of Voyage AI by MongoDB" across "discovery and onboarding through activation, adoption, and expansion", with lifecycle work "across email, in-product, content, and emerging channels", and asks for experience "designing and executing cross-sell/up-sell strategies to expand customer adoption across product offerings".Source
  • 02

    There are two Voyage documentation sets, each canonical to itself, each with its own llms.txt and its own setup steps: one says to create keys in Atlas, the other in the Voyage dashboard. The VP of AI Marketing is hired to own how MongoDB appears in AI answers and treats documentation quality as a distribution lever. Two answers to "how do I get a Voyage API key" is exactly what coding assistants get wrong.

    High impactMedium effort First result in 1 quarter VP

    First action

    Ask three coding assistants "How do I get a Voyage AI API key?" and record which instructions they give.

    How to measure it

    Correct assistant answers for ten setup questions

    Now
    –
    Target
    9 of 10
    By
    2 quarters

    Steps

    1. 1

      Baseline

      Ten setup questions across the main assistants.

      VP, AI Marketing Strategy 2 days

    2. 2

      Map pages

      Each docs.voyageai.com page to its mongodb.com equivalent.

      Documentation Lead 1 week

    3. 3

      Add canonicals

      Or redirects where the content is the same.

      Documentation Lead 1 week

    4. 4

      Merge llms.txt

      One Voyage index linked from the docs index.

      Documentation Lead 2 days

    5. 5

      One pricing table

      Same free tokens and prices in one place.

      Product Marketing Manager, AI 2 days

    6. 6

      Retest

      Monthly assistant check.

      VP, AI Marketing Strategy monthly

    First result
    1 quarter
    Cost
    Docs time
    Owner
    VP, AI Marketing Strategy

    Smallest test

    Add a canonical to the API key page and retest in a month.

    Why it works

    Choose mongodb.com/docs/voyageai as the canonical set, point docs.voyageai.com pages at their equivalents with canonical tags (or redirects page by page), merge the llms.txt files, and keep one pricing table. Test monthly what the main coding assistants answer for ten Voyage setup questions.

    Assistants and search engines converge on the instructions that lead into Atlas, and docs teams maintain one set instead of two.

    Risks and how to handle them

    • Losing search traffic from the older domain

      Move page by page and watch Search Console.

    Evidence

    • Voyage AI docsdocs.voyageai.com is a second, self-canonical documentation set with its own llms.txt, its own pricing page and its own key instructions: "sign in with your Voyage AI account and click the 'Create new secret key' button" in the Voyage dashboard. Its pricing gives "the first 200 million tokens" free for voyage-4 models "for every account".Source
    • Your websitemongodb.com's Voyage AI page sends "Get Started" to Atlas registration with onboardingScenario=voyage, and the MongoDB docs quick start says: "To access Voyage AI models, create a model API key in the MongoDB Atlas UI", under "AI Model APIs" in an Atlas project.Source
    • HiringVP, AI Marketing Strategy (San Francisco or Palo Alto, reporting to the CMO): exists "to secure the next generation of builders before they default to another platform"; owns "how MongoDB appears in AI-generated responses, agent framework recommendations, and developer tool suggestions", "ecosystem integration presence", co-marketing with "agent frameworks, deployment platforms, and frontier model providers" and "agentic builder acquisition"; treats "documentation quality, code example density, and dataset presence" as "distribution levers".Source
    • Your websiteMongoDB already publishes an llms.txt for the site and a separate one for the docs ("How agents should use this file"), plus a Markdown pricing page (updated September 29, 2026): Atlas Free with 512 MB, Flex "Up to $30/month", Dedicated from $56.94 a month, and Atlas Infinite "for unpredictable AI demand".Source
  • 03

    Google's field data shows real Chrome users wait 5.3 s for the homepage's main content and 318 ms for it to respond, failing Core Web Vitals. The page downloads 13 MB, including 4.2 MB of fonts, and 37 third parties make up 7.7 MB. The growth team runs a conversion-optimised pricing page; the homepage that feeds it is the slowest step.

    High impactMedium effort First result in 1 month Web Engineering Lead

    First action

    List every font file and weight the homepage loads.

    How to measure it

    Field LCP

    Now
    –
    Target
    under 2.5 s
    By
    90 days

    Steps

    1. 1

      Inventory

      Fonts, images and third parties by size.

      Web Engineering Lead 2 days

    2. 2

      Fonts

      Subset, two weights, WOFF2, preload one.

      Web Engineering Lead 1 week

    3. 3

      Images

      Responsive sizes from the CMS.

      Web Engineering Lead 1 week

    4. 4

      Tags

      Defer marketing tags until after interaction.

      Marketing Automation Manager 1 week

    5. 5

      Remove

      Third parties with no owner.

      Marketing Automation Manager 1 week

    6. 6

      Watch

      Field data and signup rate from the homepage.

      Marketing Analytics weekly

    First result
    1 month
    Cost
    Engineering time
    Owner
    Web Engineering Lead

    Smallest test

    Ship the font change alone and compare lab and field numbers after 28 days.

    Why it works

    Subset and self-host the web fonts (two weights, WOFF2), serve hero images at display size, load tag manager and ad pixels after interaction or consent, and remove third parties no team uses. Track field data in Search Console until both metrics pass.

    Faster pages convert more of the developers who arrive from search, ads and AI answers, and the fix is engineering time, not budget.

    Risks and how to handle them

    • Attribution gaps from deferred tags

      Keep server-side events for signups.

    Evidence

    • Google PageSpeedHomepage, October 21, 2026: real Chrome users fail Core Web Vitals (largest content at 5.3 s and interaction delay of 318 ms at the 75th percentile, both over Google's limits). Lab mobile score 31/100. The page downloads 13 MB in 296 requests: 4.2 MB of fonts, 3.0 MB of images and 2.7 MB of JavaScript; 37 third parties account for 7.7 MB, and Google Tag Manager alone blocks the main thread for about 1 s.Source
    • HiringSoftware Engineer 3, Atlas Growth 2 (Dublin): Atlas Growth uses "an in-house experiments platform"; squads focus "on a single metric (e.g. retention)"; recent projects include "an AI Chatbot for cluster creation, a recommendation system that offers tips for better database performance, and a pricing page designed to optimize conversion rates".Source
    • HiringSenior Director Product Management, Growth (United States, posted August 27, 2026): owns "the end-to-end product-led growth journey, from developer acquisition and onboarding through activation, adoption, monetization, retention, expansion, and sales handoffs" for "both MongoDB Community Edition and MongoDB Atlas", including "packaging, pricing, and monetization that accelerate self-service adoption".Source
  • 04

    MongoDB has comparison pages against eleven databases, from MySQL to Elasticsearch, but none for the choice AI builders make today: a dedicated vector database, Postgres with vector extensions, or another embedding provider. Pinecone, Supabase and Cohere all market directly to the builders the VP of AI Marketing must win "before they default to another platform".

    High impactLow effort First result in 1 month Product Marketing Manager

    First action

    Pick the three rivals that come up most in lost deals and support tickets.

    How to measure it

    Atlas signups from AI comparison pages

    Now
    –
    Target
    2,000 a quarter
    By
    2 quarters

    Steps

    1. 1

      Pick rivals

      From deal notes and search data.

      Product Marketing Manager, AI 1 week

    2. 2

      Run tests

      Reproducible cost and accuracy runs.

      Developer Advocate 2 weeks

    3. 3

      Write pages

      Tables, code and migration steps.

      Product Marketing Manager, AI 2 weeks

    4. 4

      Legal review

      Claims and trademarks.

      Legal 1 week

    5. 5

      Publish

      With llms.txt entries.

      Documentation Lead 1 day

    6. 6

      Refresh

      Quarterly with new versions.

      Product Marketing Manager, AI quarterly

    First result
    1 month
    Cost
    Staff time
    Owner
    Product Marketing Manager, AI

    Smallest test

    Publish the Pinecone comparison first.

    Why it works

    Publish fair comparisons: Atlas Vector Search vs Pinecone, vs Postgres with pgvector (Supabase), and Voyage vs other embedding models on cost, accuracy, dimensions and setup. Include runnable code, a migration guide and a free-token offer.

    Comparison searches come right before a decision, and AI assistants quote well-structured comparisons when asked which tool to choose.

    Risks and how to handle them

    • Benchmarks disputed

      Publish code and data so anyone can rerun them.

    Evidence

    • Your websiteSitemap (checked October 21, 2026): comparison pages against MySQL, PostgreSQL, Oracle, DynamoDB, Redis, Cosmos DB, DocumentDB, Couchbase, Cassandra, HBase and Elasticsearch, but none against dedicated vector databases or other embedding model providers. The English page sitemap lists 1,224 URLs; the Spanish one lists 127.Source
    • Competitor researchRivals' own homepages: Pinecone, "the trusted AI knowledge company" with "10,000+ customers and 1M developers"; Supabase, "The Postgres Development Platform" with "Vector embeddings" built in; Couchbase, "the Operational Data Platform for AI"; Cohere Embed for "semantic retrieval" over "multilingual, and multimodal data".Source
    • HiringVP, AI Marketing Strategy (San Francisco or Palo Alto, reporting to the CMO): exists "to secure the next generation of builders before they default to another platform"; owns "how MongoDB appears in AI-generated responses, agent framework recommendations, and developer tool suggestions", "ecosystem integration presence", co-marketing with "agent frameworks, deployment platforms, and frontier model providers" and "agentic builder acquisition"; treats "documentation quality, code example density, and dataset presence" as "distribution levers".Source
    • Voyage AI docsdocs.voyageai.com is a second, self-canonical documentation set with its own llms.txt, its own pricing page and its own key instructions: "sign in with your Voyage AI account and click the 'Create new secret key' button" in the Voyage dashboard. Its pricing gives "the first 200 million tokens" free for voyage-4 models "for every account".Source
  • 05

    MongoDB for Startups credits expire 12 months after activation, and the page asks startups to get in touch with three months remaining. That relies on founders remembering. The Marketing Automation Manager is hired to build automations across Iterable, Salesforce and Workato.

    Medium impactLow effort First result in 1 month Marketing Automation Manager

    First action

    Count startups with credits expiring in the next 90 days and their unused balances.

    How to measure it

    Credits used before expiry

    Now
    –
    Target
    +20%
    By
    2 quarters

    Steps

    1. 1

      Pull the data

      Expiry dates and balances.

      Marketing Analytics 2 days

    2. 2

      Write the emails

      Three, with a next step each.

      Startup Program Manager 3 days

    3. 3

      Build the flow

      Workato to Iterable.

      Marketing Automation Manager 1 week

    4. 4

      Route extensions

      Salesforce task for the startup team.

      Marketing Automation Manager 2 days

    5. 5

      Launch

      All active startups.

      Marketing Automation Manager 1 day

    6. 6

      Measure

      Utilisation and conversion to paid.

      Marketing Analytics monthly

    First result
    1 month
    Cost
    $0
    Owner
    Marketing Automation Manager

    Smallest test

    Send the three-month email by hand to 50 startups.

    Why it works

    Trigger automatic emails at six, three and one month before expiry with the remaining balance, a suggested next step (move to a dedicated cluster, add Vector Search, try Voyage tokens) and a one-click extension request routed to the startup team.

    Startups that use their credits build on MongoDB; credits that lapse unused are a lost chance to convert them.

    Risks and how to handle them

    • Too many emails

      Merge with existing billing notices.

    Evidence

    • Your websiteMongoDB for Startups offers "Atlas credits and Voyage AI tokens" in four tiers (Inspire, Grow, Innovate, Scale). "Once activated, startups have 12 months to use your Atlas credits"; extensions are case by case, and startups are asked to get in touch "when you have three months remaining".Source
    • HiringMarketing Automation Manager (Dublin, posted October 5, 2026): builds "AI driven automations and integrations" with "Workato, Claude", working with the owners of "Marketo, Iterable, Splash, Intercom, Workfront, and Salesforce".Source
    • HiringIndia: a Manager, Startup Marketing (Gurugram, posted September 22, 2026) to "engage and convert high-potential startups in the region into the MongoDB for Startups program" through VCs, accelerators, hackathons and founder dinners; and a Senior Manager, Marketing (posted September 16) for "Global accounts with developer arms in India".Source
  • 06

    The homepage loads 37 third parties and Google Tag Manager blocks the main thread for about 1 s. Ad exchanges, data vendors and several overlapping tools are among them. Each added tag costs speed and adds data-sharing exposure.

    Medium impactLow effort First result in 2 weeks Marketing Automation Manager

    First action

    Export the tag manager container and mark each tag's owner.

    How to measure it

    Third-party count

    Now
    –
    Target
    under 15
    By
    1 quarter

    Steps

    1. 1

      Export

      All tags and triggers.

      Marketing Automation Manager 1 day

    2. 2

      Assign owners

      Each tag to a team.

      Marketing Automation Manager 1 week

    3. 3

      Remove

      Tags no team uses.

      Marketing Automation Manager 1 week

    4. 4

      Check overlap

      Tools that do the same job.

      Marketing Automation Manager 1 week

    5. 5

      Gate

      Consent and interaction triggers.

      Marketing Automation Manager 1 week

    6. 6

      Review

      Quarterly.

      Marketing Automation Manager quarterly

    First result
    2 weeks
    Cost
    Staff time
    Owner
    Marketing Automation Manager

    Smallest test

    Remove one tag no team uses this week.

    Why it works

    List every tag with an owner and purpose, remove those without one, move ad and enrichment tags behind consent and after interaction, and review the list quarterly with marketing operations.

    Fewer, owned tags make the page faster and keep the data flows in line with the consent banner.

    Risks and how to handle them

    • Breaking campaign tracking

      Keep server-side conversions in place first.

    Evidence

    • Google PageSpeedHomepage, October 21, 2026: real Chrome users fail Core Web Vitals (largest content at 5.3 s and interaction delay of 318 ms at the 75th percentile, both over Google's limits). Lab mobile score 31/100. The page downloads 13 MB in 296 requests: 4.2 MB of fonts, 3.0 MB of images and 2.7 MB of JavaScript; 37 third parties account for 7.7 MB, and Google Tag Manager alone blocks the main thread for about 1 s.Source
    • HiringMarketing Automation Manager (Dublin, posted October 5, 2026): builds "AI driven automations and integrations" with "Workato, Claude", working with the owners of "Marketo, Iterable, Splash, Intercom, Workfront, and Salesforce".Source

This quarter 10

The core plays for the next three months.

  • 07

    The Voyage growth role is asked for cross-sell and up-sell across products. Atlas users who create vector indexes already store embeddings from some model; Voyage users already have embeddings to store.

    High impactMedium effort First result in 1 quarter Growth Marketing Manager

    First action

    Count Atlas projects with vector indexes and no Voyage key.

    How to measure it

    Projects using Voyage and Vector Search

    Now
    –
    Target
    +30%
    By
    2 quarters

    Steps

    1. 1

      Size both groups

      Vector users without Voyage and the reverse.

      Marketing Analytics 1 week

    2. 2

      Design prompts

      One in Atlas, one in the Voyage key page.

      Growth Marketing Manager, Voyage AI 2 weeks

    3. 3

      Build the recall test

      Sample of the user's own data.

      Software Engineer 3, Atlas Growth 1 month

    4. 4

      Lifecycle email

      Follow-up for those who click.

      Growth Marketing Manager, Voyage AI 2 weeks

    5. 5

      Experiment

      On the Atlas experiments platform.

      Software Engineer 3, Atlas Growth 1 month

    6. 6

      Report

      Adoption in both directions.

      Marketing Analytics monthly

    First result
    1 quarter
    Cost
    Engineering time
    Owner
    Growth Marketing Manager, Voyage AI

    Smallest test

    Email 1,000 Vector Search users the recall test.

    Why it works

    In Atlas, show Vector Search users a one-line switch to Voyage with their free-token allowance and a side-by-side recall test on their own sample. Show Voyage key holders a "store these in Atlas Vector Search" quick start. Track both directions.

    The prompt appears at the moment the developer is doing the related task, which beats any campaign.

    Risks and how to handle them

    • Prompts feel pushy

      Show once per project and only to active users.

    Evidence

    • HiringGrowth Marketing Manager, Voyage AI (two openings, United States and Toronto, posted October 8, 2026): drives "growth and adoption of Voyage AI by MongoDB" across "discovery and onboarding through activation, adoption, and expansion", with lifecycle work "across email, in-product, content, and emerging channels", and asks for experience "designing and executing cross-sell/up-sell strategies to expand customer adoption across product offerings".Source
    • Your websitemongodb.com's Voyage AI page sends "Get Started" to Atlas registration with onboardingScenario=voyage, and the MongoDB docs quick start says: "To access Voyage AI models, create a model API key in the MongoDB Atlas UI", under "AI Model APIs" in an Atlas project.Source
    • Voyage AI docsdocs.voyageai.com is a second, self-canonical documentation set with its own llms.txt, its own pricing page and its own key instructions: "sign in with your Voyage AI account and click the 'Create new secret key' button" in the Voyage dashboard. Its pricing gives "the first 200 million tokens" free for voyage-4 models "for every account".Source
    • HiringSenior Director Product Management, Growth (United States, posted August 27, 2026): owns "the end-to-end product-led growth journey, from developer acquisition and onboarding through activation, adoption, monetization, retention, expansion, and sales handoffs" for "both MongoDB Community Edition and MongoDB Atlas", including "packaging, pricing, and monetization that accelerate self-service adoption".Source
  • 08

    MongoDB already publishes llms.txt files and Markdown pages for agents, and the VP of AI Marketing owns "agentic builder acquisition". What is missing is a measure: how many Atlas projects are created because an agent recommended or set up MongoDB.

    High impactMedium effort First result in 2 quarters VP

    First action

    Add a source tag to the registration link in llms.txt.

    How to measure it

    Agent-sourced projects per month

    Now
    –
    Target
    baseline then +50%
    By
    2 quarters

    Steps

    1. 1

      Tag links

      llms.txt, docs Markdown, MCP server.

      Documentation Lead 1 week

    2. 2

      Record source

      On CLI and MCP project creation.

      Software Engineer 3, Atlas Growth 1 month

    3. 3

      Report

      Agent-started projects weekly.

      Marketing Analytics 2 weeks

    4. 4

      Compare

      Activation versus other channels.

      Marketing Analytics 1 month

    5. 5

      Invest

      Improve the docs with the best agent outcomes.

      VP, AI Marketing Strategy ongoing

    6. 6

      Share

      In the quarterly business review.

      VP, AI Marketing Strategy quarterly

    First result
    2 quarters
    Cost
    Engineering time
    Owner
    VP, AI Marketing Strategy

    Smallest test

    Tag the registration link in llms.txt and count for a month.

    Why it works

    Add a source parameter to every registration and CLI link in the llms.txt files, MCP server and Build with AI docs; record the agent or tool on project creation where the CLI or MCP server is used; report agent-started projects alongside paid and organic.

    A channel that is measured gets budget and attention; one that is not stays a hunch.

    Risks and how to handle them

    • Undercounting

      Combine tags with a signup question.

    Evidence

    • Your websiteMongoDB already publishes an llms.txt for the site and a separate one for the docs ("How agents should use this file"), plus a Markdown pricing page (updated September 29, 2026): Atlas Free with 512 MB, Flex "Up to $30/month", Dedicated from $56.94 a month, and Atlas Infinite "for unpredictable AI demand".Source
    • HiringVP, AI Marketing Strategy (San Francisco or Palo Alto, reporting to the CMO): exists "to secure the next generation of builders before they default to another platform"; owns "how MongoDB appears in AI-generated responses, agent framework recommendations, and developer tool suggestions", "ecosystem integration presence", co-marketing with "agent frameworks, deployment platforms, and frontier model providers" and "agentic builder acquisition"; treats "documentation quality, code example density, and dataset presence" as "distribution levers".Source
    • HiringSenior Director Product Management, Growth (United States, posted August 27, 2026): owns "the end-to-end product-led growth journey, from developer acquisition and onboarding through activation, adoption, monetization, retention, expansion, and sales handoffs" for "both MongoDB Community Edition and MongoDB Atlas", including "packaging, pricing, and monetization that accelerate self-service adoption".Source
  • 09

    The new AMER hub in Mexico City runs Skill Badge workshops and reports "developers enabled and certified"; the site promotes "3 new free, focused MongoDB gen AI credentials". A badge is a learning outcome; a cluster with data is a product outcome.

    High impactMedium effort First result in 1 quarter Field Marketing Manager

    First action

    Add a "launch this on a free cluster" link to one badge.

    How to measure it

    Active clusters from badges

    Now
    –
    Target
    30% of badge holders
    By
    2 quarters

    Steps

    1. 1

      Pick badges

      The three gen AI badges.

      Field Marketing Manager (Mexico City) 1 week

    2. 2

      Build templates

      Preloaded datasets.

      Developer Advocate 1 month

    3. 3

      Add links

      Badge completion page.

      Web Engineering Lead 1 week

    4. 4

      Follow up

      Project idea email.

      Marketing Automation Manager 2 weeks

    5. 5

      Run workshops

      With the cluster step live.

      Field Marketing Manager (Mexico City) monthly

    6. 6

      Report

      Clusters active by region.

      Marketing Analytics monthly

    First result
    1 quarter
    Cost
    Staff time
    Owner
    Field Marketing Manager (Mexico City)

    Smallest test

    Run one workshop ending in cluster launch.

    Why it works

    End each badge with a one-click free cluster preloaded with the course dataset and a Voyage key, and follow up with a project idea. Report developers enabled and clusters still active after 30 days by region.

    It moves the field team's metric from attendance to product use.

    Risks and how to handle them

    • Free clusters left idle

      Pause after inactivity and remind.

    Evidence

    • HiringField marketing: a Field Marketing Manager in Mexico City (posted October 5, 2026) for the "newly created AMER Marketing Hub", running "Developer Days, AI Builder Days" and "Skill Badge workshops and webinars" and "tracking and reporting on developers enabled and certified"; a Field Marketing Manager in Paris (posted September 11) for account-based marketing to "our largest and most strategic French accounts" and campaigns for "digital-native organizations and AI companies".Source
    • Your websiteEvents and learning: MongoDB.local events in Toronto, New York, Dallas, Mumbai and Tel Aviv, a Build Fest, and a site-wide banner: "Level up your AI skills with 3 new free, focused MongoDB gen AI credentials. Earn a skill badge".Source
    • Your websitemongodb.com's Voyage AI page sends "Get Started" to Atlas registration with onboardingScenario=voyage, and the MongoDB docs quick start says: "To access Voyage AI models, create a model API key in the MongoDB Atlas UI", under "AI Model APIs" in an Atlas project.Source
  • 10

    The growth product leader owns "sales handoffs", and MongoDB has 32 open account executive roles with "Growth" in the title. Self-serve signals such as spend growth, a second project or a Vector Search index are the best timing for a sales call.

    High impactHigh effort First result in 2 quarters Senior Director Product Manag…

    First action

    Pick three triggers from accounts that upgraded last year.

    How to measure it

    PQA to opportunity rate

    Now
    –
    Target
    25%
    By
    2 quarters

    Steps

    1. 1

      Find triggers

      From past upgrades.

      Marketing Analytics 2 weeks

    2. 2

      Build scores

      In the data warehouse.

      Marketing Analytics 1 month

    3. 3

      Sync

      To Salesforce with the reason.

      Marketing Automation Manager 2 weeks

    4. 4

      Scripts

      One per trigger.

      Sales Enablement 1 week

    5. 5

      Holdout test

      Half the accounts.

      Senior Director Product Management, Growth 1 quarter

    6. 6

      Scale

      Across regions.

      Senior Director Product Management, Growth 1 quarter

    First result
    2 quarters
    Cost
    Analytics time
    Owner
    Senior Director Product Management, Growth

    Smallest test

    Send one trigger to one AE team for a month.

    Why it works

    Define three product-qualified triggers, send them to Salesforce with the account and the trigger, and give AEs a short script per trigger. Test against a holdout.

    Sales calls timed to usage convert better and feel helpful rather than cold.

    Risks and how to handle them

    • Developers dislike sales calls

      Offer technical help first.

    Evidence

    • HiringSenior Director Product Management, Growth (United States, posted August 27, 2026): owns "the end-to-end product-led growth journey, from developer acquisition and onboarding through activation, adoption, monetization, retention, expansion, and sales handoffs" for "both MongoDB Community Edition and MongoDB Atlas", including "packaging, pricing, and monetization that accelerate self-service adoption".Source
    • HiringRevenue and partner marketing: a Regional Revenue Marketing Manager (Chicago or Minneapolis, posted September 4, 2026) measured on "pipeline, account expansion, and net annual recurring revenue (NARR)" and "developers enabled"; a Senior Partner Marketing Manager for "ISV/Tech & AI partners" to "establish MongoDB the default data layer across these partners"; and 32 open account executive roles with "Growth" in the title.Source
    • HiringSoftware Engineer 3, Atlas Growth 2 (Dublin): Atlas Growth uses "an in-house experiments platform"; squads focus "on a single metric (e.g. retention)"; recent projects include "an AI Chatbot for cluster creation, a recommendation system that offers tips for better database performance, and a pricing page designed to optimize conversion rates".Source
  • 11

    MongoDB's Spanish page sitemap lists 127 pages against 1,224 in English. The new AMER hub in Mexico City reports to the LATAM marketing lead, and the AI pages that win builders today are mostly English.

    Medium impactMedium effort First result in 1 quarter Field Marketing Manager

    First action

    List the top ten AI pages by signups.

    How to measure it

    LATAM signups from localized AI pages

    Now
    –
    Target
    +40%
    By
    2 quarters

    Steps

    1. 1

      Pick pages

      Top ten AI pages.

      Field Marketing Manager (Mexico City) 1 week

    2. 2

      Translate

      Spanish and Portuguese.

      Localization Manager 1 month

    3. 3

      Local proof

      Two LATAM customer examples.

      Field Marketing Manager (Mexico City) 1 month

    4. 4

      Link

      From workshops and events.

      Field Marketing Manager (Mexico City) 2 weeks

    5. 5

      hreflang

      Check tags.

      Web Engineering Lead 1 day

    6. 6

      Measure

      Signups by locale.

      Marketing Analytics monthly

    First result
    1 quarter
    Cost
    Translation budget
    Owner
    Field Marketing Manager (Mexico City)

    Smallest test

    Translate the Voyage page.

    Why it works

    Translate the ten AI pages that bring most signups (Vector Search, Voyage, Agent Engine, the gen AI badges) into Spanish and Brazilian Portuguese, with local examples and events, and link them from LATAM workshops.

    Developers convert faster in their first language, and few AI data vendors serve Spanish or Portuguese well.

    Risks and how to handle them

    • Docs stay English

      Link to translated quick starts first.

    Evidence

    • Your websiteSitemap (checked October 21, 2026): comparison pages against MySQL, PostgreSQL, Oracle, DynamoDB, Redis, Cosmos DB, DocumentDB, Couchbase, Cassandra, HBase and Elasticsearch, but none against dedicated vector databases or other embedding model providers. The English page sitemap lists 1,224 URLs; the Spanish one lists 127.Source
    • HiringField marketing: a Field Marketing Manager in Mexico City (posted October 5, 2026) for the "newly created AMER Marketing Hub", running "Developer Days, AI Builder Days" and "Skill Badge workshops and webinars" and "tracking and reporting on developers enabled and certified"; a Field Marketing Manager in Paris (posted September 11) for account-based marketing to "our largest and most strategic French accounts" and campaigns for "digital-native organizations and AI companies".Source
    • Your websiteEvents and learning: MongoDB.local events in Toronto, New York, Dallas, Mumbai and Tel Aviv, a Build Fest, and a site-wide banner: "Level up your AI skills with 3 new free, focused MongoDB gen AI credentials. Earn a skill badge".Source
  • 12

    MongoDB runs MongoDB.local events across many cities and uses Splash for events, Iterable for email and Salesforce for sales. The new automation role works with Workato and Claude. Generic "thanks for coming" emails waste the moment.

    Medium impactMedium effort First result in 1 quarter Marketing Automation Manager

    First action

    Draft follow-ups for one event's workshop attendees.

    How to measure it

    Meetings per 100 attendees

    Now
    –
    Target
    +30%
    By
    2 quarters

    Steps

    1. 1

      Map data

      Splash, scans, Salesforce.

      Marketing Automation Manager 1 week

    2. 2

      Build the flow

      Workato with a model draft.

      Marketing Automation Manager 2 weeks

    3. 3

      Review step

      Field marketer approves.

      Field Marketing Manager (Mexico City) 1 week

    4. 4

      Send

      Within 48 hours.

      Marketing Automation Manager per event

    5. 5

      Route

      AE summary for hot accounts.

      Marketing Automation Manager 1 week

    6. 6

      Measure

      Meetings per event.

      Marketing Analytics per event

    First result
    1 quarter
    Cost
    Staff time
    Owner
    Marketing Automation Manager

    Smallest test

    Pilot on the next MongoDB.local.

    Why it works

    Use session check-ins, booth scans and workshop completions to draft each attendee's follow-up: the sessions they attended, the matching docs and a next step, reviewed by the field marketer. Route hot accounts to the AE with the same summary.

    Relevant follow-up lifts post-event pipeline without more events.

    Risks and how to handle them

    • Wrong details

      Human review before sending.

    Evidence

    • HiringMarketing Automation Manager (Dublin, posted October 5, 2026): builds "AI driven automations and integrations" with "Workato, Claude", working with the owners of "Marketo, Iterable, Splash, Intercom, Workfront, and Salesforce".Source
    • Your websiteEvents and learning: MongoDB.local events in Toronto, New York, Dallas, Mumbai and Tel Aviv, a Build Fest, and a site-wide banner: "Level up your AI skills with 3 new free, focused MongoDB gen AI credentials. Earn a skill badge".Source
    • HiringRevenue and partner marketing: a Regional Revenue Marketing Manager (Chicago or Minneapolis, posted September 4, 2026) measured on "pipeline, account expansion, and net annual recurring revenue (NARR)" and "developers enabled"; a Senior Partner Marketing Manager for "ISV/Tech & AI partners" to "establish MongoDB the default data layer across these partners"; and 32 open account executive roles with "Growth" in the title.Source
  • 13

    The partner marketer is hired to make MongoDB "the default data layer" across ISV, tech and AI partners, and the VP of AI Marketing owns presence inside agent frameworks. Builders copy the first quick start they find in a framework's docs.

    High impactMedium effort First result in 2 quarters Senior Partner Marketing Mana…

    First action

    Audit the quick starts of ten agent frameworks.

    How to measure it

    Default or listed in top-20 quick starts

    Now
    –
    Target
    12 of 20
    By
    3 quarters

    Steps

    1. 1

      Audit

      Top 20 frameworks and platforms.

      Senior Partner Marketing Manager 2 weeks

    2. 2

      Prioritise

      By developer reach.

      Senior Partner Marketing Manager 1 week

    3. 3

      Build templates

      Maintained code.

      Developer Advocate 1 quarter

    4. 4

      Pitch

      Co-marketing offers.

      Senior Partner Marketing Manager 1 quarter

    5. 5

      Track

      Tagged links.

      Marketing Analytics monthly

    6. 6

      Maintain

      Update with releases.

      Developer Advocate monthly

    First result
    2 quarters
    Cost
    Developer advocate time
    Owner
    Senior Partner Marketing Manager

    Smallest test

    Get one framework to list a MongoDB template.

    Why it works

    List the top 20 agent frameworks and deployment platforms, check which data store each quick start uses, and offer each a maintained MongoDB template with a tracked link. Co-market each new template with the partner.

    Being in the default template wins developers before they compare options.

    Risks and how to handle them

    • Partners want neutrality

      Offer templates alongside others.

    Evidence

    • HiringRevenue and partner marketing: a Regional Revenue Marketing Manager (Chicago or Minneapolis, posted September 4, 2026) measured on "pipeline, account expansion, and net annual recurring revenue (NARR)" and "developers enabled"; a Senior Partner Marketing Manager for "ISV/Tech & AI partners" to "establish MongoDB the default data layer across these partners"; and 32 open account executive roles with "Growth" in the title.Source
    • HiringVP, AI Marketing Strategy (San Francisco or Palo Alto, reporting to the CMO): exists "to secure the next generation of builders before they default to another platform"; owns "how MongoDB appears in AI-generated responses, agent framework recommendations, and developer tool suggestions", "ecosystem integration presence", co-marketing with "agent frameworks, deployment platforms, and frontier model providers" and "agentic builder acquisition"; treats "documentation quality, code example density, and dataset presence" as "distribution levers".Source
    • News coverageMongoDB press releases: MongoDB 9.0 and Atlas Infinite (September 29, 2026), a partnership with Cognition to modernize enterprise infrastructure (September 29), and Atlas Agent Engine "to put AI agents in production without a new stack" (September 30). MongoDB describes itself as having "over 67,000 customers, including AI-native startups and approximately 75% of the Fortune 100".Source
  • 14

    The India startup marketer is hired to convert startups into MongoDB for Startups through VCs and accelerators. The program has four tiers of credits and Voyage tokens.

    Medium impactLow effort First result in 1 quarter Manager

    First action

    Sign three Indian funds to partner links.

    How to measure it

    Partner-sourced startups

    Now
    –
    Target
    150 a quarter
    By
    2 quarters

    Steps

    1. 1

      List funds

      Top 20 in India.

      Manager, Startup Marketing (India) 1 week

    2. 2

      Create links

      Per partner with tier.

      Marketing Automation Manager 1 week

    3. 3

      Sign partners

      Three to start.

      Manager, Startup Marketing (India) 1 month

    4. 4

      Run events

      Portfolio dinners.

      Manager, Startup Marketing (India) monthly

    5. 5

      Report

      To each fund.

      Manager, Startup Marketing (India) quarterly

    6. 6

      Expand

      To more funds.

      Manager, Startup Marketing (India) quarterly

    First result
    1 quarter
    Cost
    Event budget
    Owner
    Manager, Startup Marketing (India)

    Smallest test

    Run one portfolio session.

    Why it works

    Give each partner fund and accelerator its own application link and tier, run founder dinners with their portfolios, and report signups and activation per partner to the funds.

    Funds promote what helps their portfolio, and per-partner tracking shows which ones to invest in.

    Risks and how to handle them

    • Low activation

      Pair credits with a technical session.

    Evidence

    • HiringIndia: a Manager, Startup Marketing (Gurugram, posted September 22, 2026) to "engage and convert high-potential startups in the region into the MongoDB for Startups program" through VCs, accelerators, hackathons and founder dinners; and a Senior Manager, Marketing (posted September 16) for "Global accounts with developer arms in India".Source
    • Your websiteMongoDB for Startups offers "Atlas credits and Voyage AI tokens" in four tiers (Inspire, Grow, Innovate, Scale). "Once activated, startups have 12 months to use your Atlas credits"; extensions are case by case, and startups are asked to get in touch "when you have three months remaining".Source
  • 15

    The pricing page now asks builders to choose between Atlas Core and Atlas Infinite, plus Free, Flex and Dedicated tiers. Atlas Growth has already built a pricing page tuned for conversion and an AI chatbot for cluster creation.

    Medium impactMedium effort First result in 1 quarter Software Engineer 3

    First action

    Write the decision rules with solutions architects.

    How to measure it

    Pricing-to-cluster rate

    Now
    –
    Target
    +10%
    By
    1 quarter

    Steps

    1. 1

      Rules

      With solutions architects.

      Senior Director Product Management, Growth 1 week

    2. 2

      Design

      Three questions.

      Product Designer, Growth 1 week

    3. 3

      Build

      On the pricing page.

      Software Engineer 3, Atlas Growth 1 month

    4. 4

      Prefill

      Cluster creation.

      Software Engineer 3, Atlas Growth 2 weeks

    5. 5

      Experiment

      A/B test.

      Software Engineer 3, Atlas Growth 1 month

    6. 6

      Review

      Recommendation accuracy.

      Senior Director Product Management, Growth monthly

    First result
    1 quarter
    Cost
    Engineering time
    Owner
    Software Engineer 3, Atlas Growth

    Smallest test

    Test the questions as a static decision table.

    Why it works

    Add three questions (steady or spiky load, data size, AI workload) that recommend a deployment and tier with an estimated monthly cost, then open cluster creation with those settings.

    A recommendation removes the hardest choice on the page and starts the build.

    Risks and how to handle them

    • Wrong recommendations

      Show the reasoning and allow changes.

    Evidence

    • Your websiteMongoDB already publishes an llms.txt for the site and a separate one for the docs ("How agents should use this file"), plus a Markdown pricing page (updated September 29, 2026): Atlas Free with 512 MB, Flex "Up to $30/month", Dedicated from $56.94 a month, and Atlas Infinite "for unpredictable AI demand".Source
    • HiringSoftware Engineer 3, Atlas Growth 2 (Dublin): Atlas Growth uses "an in-house experiments platform"; squads focus "on a single metric (e.g. retention)"; recent projects include "an AI Chatbot for cluster creation, a recommendation system that offers tips for better database performance, and a pricing page designed to optimize conversion rates".Source
    • News coverageMongoDB press releases: MongoDB 9.0 and Atlas Infinite (September 29, 2026), a partnership with Cognition to modernize enterprise infrastructure (September 29), and Atlas Agent Engine "to put AI agents in production without a new stack" (September 30). MongoDB describes itself as having "over 67,000 customers, including AI-native startups and approximately 75% of the Fortune 100".Source
  • 16

    The Paris field marketer runs account-based programs for the largest French accounts; the India marketer targets global accounts with developer teams there. MongoDB has huge self-serve and community usage (8.8 million weekly driver downloads) that shows where developers already use it.

    Medium impactMedium effort First result in 1 quarter Field Marketing Manager

    First action

    Rank the French target list by Atlas signups from company domains.

    How to measure it

    Accounts with 5+ engaged developers

    Now
    –
    Target
    +50%
    By
    2 quarters

    Steps

    1. 1

      Match data

      Signups and activity to accounts.

      Marketing Analytics 2 weeks

    2. 2

      Rank

      Target lists.

      Field Marketing Manager (Paris) 1 week

    3. 3

      Invite

      Developer workshops.

      Field Marketing Manager (Paris) 1 month

    4. 4

      Brief AEs

      Champions per account.

      Field Marketing Manager (Paris) 2 weeks

    5. 5

      Repeat

      India accounts.

      Senior Manager, Marketing (India) 1 month

    6. 6

      Measure

      Pipeline from ranked accounts.

      Marketing Analytics quarterly

    First result
    1 quarter
    Cost
    Analytics time
    Owner
    Field Marketing Manager (Paris)

    Smallest test

    Rank one region's accounts.

    Why it works

    Match Atlas signups, Community downloads, forum and course activity to target accounts, rank accounts by developer activity, and invite those developers to account-specific workshops before executive outreach.

    Starting from existing developer champions shortens enterprise deals.

    Risks and how to handle them

    • Privacy

      Use company-level counts and consented contacts only.

    Evidence

    • HiringField marketing: a Field Marketing Manager in Mexico City (posted October 5, 2026) for the "newly created AMER Marketing Hub", running "Developer Days, AI Builder Days" and "Skill Badge workshops and webinars" and "tracking and reporting on developers enabled and certified"; a Field Marketing Manager in Paris (posted September 11) for account-based marketing to "our largest and most strategic French accounts" and campaigns for "digital-native organizations and AI companies".Source
    • HiringIndia: a Manager, Startup Marketing (Gurugram, posted September 22, 2026) to "engage and convert high-potential startups in the region into the MongoDB for Startups program" through VCs, accelerators, hackathons and founder dinners; and a Senior Manager, Marketing (posted September 16) for "Global accounts with developer arms in India".Source
    • Developer footprintThe mongodb npm driver has about 8.8 million weekly downloads; MongoDB's Docker images have about 71 million pulls; the mongodb/mongo GitHub repository has 28,656 stars, across 310 public repositories.Source

Big bets 3

Larger plays that can change the trajectory.

  • 17

    Voyage leads its homepage with benchmark rankings, and Pinecone, Cohere and others make similar claims. Developers trust results on their own data more than public leaderboards.

    High impactHigh effort First result in 3 quarters Growth Marketing Manager

    First action

    Prototype with a notebook on three public datasets.

    How to measure it

    Test-to-deploy rate

    Now
    –
    Target
    25%
    By
    3 quarters

    Steps

    1. 1

      Prototype

      Notebook version.

      Developer Advocate 1 month

    2. 2

      Build

      Hosted tool in Atlas.

      Software Engineer 3, Atlas Growth 1 quarter

    3. 3

      Privacy

      Data handling and deletion.

      Legal 1 month

    4. 4

      Launch

      With comparison pages.

      Growth Marketing Manager, Voyage AI 1 month

    5. 5

      Lifecycle

      Follow-up from results.

      Growth Marketing Manager, Voyage AI ongoing

    6. 6

      Measure

      Deploys from tests.

      Marketing Analytics monthly

    First result
    3 quarters
    Cost
    Engineering and inference cost
    Owner
    Growth Marketing Manager, Voyage AI

    Smallest test

    Run the notebook with five design partners.

    Why it works

    Build a hosted tool in Atlas: upload a sample of documents and queries, run Voyage and the developer's current model, and see recall, cost and vector size side by side. Save the result as a project ready to deploy.

    It turns benchmark claims into a personal proof and ends inside Atlas.

    Risks and how to handle them

    • Users upload sensitive data

      Small sample limits and automatic deletion.

    Evidence

    • Voyage AI websitevoyageai.com (checked October 21, 2026) is titled "Voyage AI by MongoDB" and is its own canonical site. Its "Get Started" and "Sign in" buttons go to dash.voyageai.com, a separate Voyage account login, and its docs links go to docs.voyageai.com. It loads its own email marketing script.Source
    • Competitor researchRivals' own homepages: Pinecone, "the trusted AI knowledge company" with "10,000+ customers and 1M developers"; Supabase, "The Postgres Development Platform" with "Vector embeddings" built in; Couchbase, "the Operational Data Platform for AI"; Cohere Embed for "semantic retrieval" over "multilingual, and multimodal data".Source
    • HiringGrowth Marketing Manager, Voyage AI (two openings, United States and Toronto, posted October 8, 2026): drives "growth and adoption of Voyage AI by MongoDB" across "discovery and onboarding through activation, adoption, and expansion", with lifecycle work "across email, in-product, content, and emerging channels", and asks for experience "designing and executing cross-sell/up-sell strategies to expand customer adoption across product offerings".Source
    • Voyage AI docsdocs.voyageai.com is a second, self-canonical documentation set with its own llms.txt, its own pricing page and its own key instructions: "sign in with your Voyage AI account and click the 'Create new secret key' button" in the Voyage dashboard. Its pricing gives "the first 200 million tokens" free for voyage-4 models "for every account".Source
  • 18

    The growth product leader owns the journey for both Community Edition and Atlas, and MongoDB's Docker images have about 71 million pulls. Local users who outgrow a laptop face a migration step that loses many of them.

    High impactHigh effort First result in 4 quarters Senior Director Product Manag…

    First action

    Measure how many Atlas signups come from existing Community users.

    How to measure it

    Monthly Community-to-Atlas moves

    Now
    –
    Target
    +50%
    By
    4 quarters

    Steps

    1. 1

      Research

      Interview recent movers.

      Senior Director Product Management, Growth 1 month

    2. 2

      Build

      CLI command.

      Software Engineer 3, Atlas Growth 1 quarter

    3. 3

      Promote

      Docker, Compass, docs.

      Growth Marketing Manager, Voyage AI 1 month

    4. 4

      Track

      Source on project creation.

      Marketing Analytics 1 month

    5. 5

      Iterate

      Fix drop-off.

      Senior Director Product Management, Growth 1 quarter

    6. 6

      Expand

      Larger databases.

      Software Engineer 3, Atlas Growth 2 quarters

    First result
    4 quarters
    Cost
    Engineering time
    Owner
    Senior Director Product Management, Growth

    Smallest test

    Release it as a beta flag in the CLI.

    Why it works

    Ship a CLI command that creates a free or Flex cluster, copies a local database and switches the connection string, with a dry run. Promote it in the Docker image, Compass and Community docs.

    Removing the move step converts local users at the moment they need more.

    Risks and how to handle them

    • Large data fails

      Start with small databases and resumable copies.

    Evidence

    • HiringSenior Director Product Management, Growth (United States, posted August 27, 2026): owns "the end-to-end product-led growth journey, from developer acquisition and onboarding through activation, adoption, monetization, retention, expansion, and sales handoffs" for "both MongoDB Community Edition and MongoDB Atlas", including "packaging, pricing, and monetization that accelerate self-service adoption".Source
    • Developer footprintThe mongodb npm driver has about 8.8 million weekly downloads; MongoDB's Docker images have about 71 million pulls; the mongodb/mongo GitHub repository has 28,656 stars, across 310 public repositories.Source
    • Your websiteMongoDB already publishes an llms.txt for the site and a separate one for the docs ("How agents should use this file"), plus a Markdown pricing page (updated September 29, 2026): Atlas Free with 512 MB, Flex "Up to $30/month", Dedicated from $56.94 a month, and Atlas Infinite "for unpredictable AI demand".Source
  • 19

    Atlas Agent Engine launched to put agents in production without a new stack. Builders choose platforms by what others have shipped.

    High impactHigh effort First result in 3 quarters VP

    First action

    Publish five templates with deploy buttons.

    How to measure it

    Template deploys

    Now
    –
    Target
    5,000
    By
    3 quarters

    Steps

    1. 1

      Pick five

      Common agent use cases.

      VP, AI Marketing Strategy 2 weeks

    2. 2

      Build

      Templates.

      Developer Advocate 1 month

    3. 3

      Gallery page

      With counts.

      Web Engineering Lead 1 month

    4. 4

      Submissions

      From partners.

      Senior Partner Marketing Manager 1 quarter

    5. 5

      Feature

      At events.

      Field Marketing Manager (Mexico City) ongoing

    6. 6

      Measure

      Deploys.

      Marketing Analytics monthly

    First result
    3 quarters
    Cost
    Developer time
    Owner
    VP, AI Marketing Strategy

    Smallest test

    Launch with five templates.

    Why it works

    Publish a gallery of open templates (support agent, research, retrieval agent) with deploy buttons, usage counts and partner badges; let partners and customers submit their own; feature the best at MongoDB.local.

    A gallery is a growth loop: each template brings deploys, and each deploy brings new templates.

    Risks and how to handle them

    • Low-quality submissions

      Review before listing.

    Evidence

    • News coverageMongoDB press releases: MongoDB 9.0 and Atlas Infinite (September 29, 2026), a partnership with Cognition to modernize enterprise infrastructure (September 29), and Atlas Agent Engine "to put AI agents in production without a new stack" (September 30). MongoDB describes itself as having "over 67,000 customers, including AI-native startups and approximately 75% of the Fortune 100".Source
    • HiringVP, AI Marketing Strategy (San Francisco or Palo Alto, reporting to the CMO): exists "to secure the next generation of builders before they default to another platform"; owns "how MongoDB appears in AI-generated responses, agent framework recommendations, and developer tool suggestions", "ecosystem integration presence", co-marketing with "agent frameworks, deployment platforms, and frontier model providers" and "agentic builder acquisition"; treats "documentation quality, code example density, and dataset presence" as "distribution levers".Source
    • Your websiteEvents and learning: MongoDB.local events in Toronto, New York, Dallas, Mumbai and Tel Aviv, a Build Fest, and a site-wide banner: "Level up your AI skills with 3 new free, focused MongoDB gen AI credentials. Earn a skill badge".Source

Wildcards 2

Unexpected experiments, each with a cheap way to test it.

  • 20

    The Node.js driver alone has about 8.8 million weekly npm downloads. The README and package pages are read by developers choosing where to run their database.

    Medium impactLow effort First result in 1 month Senior Director Product Manag…

    First action

    Add the link to the Node.js driver README.

    How to measure it

    Monthly signups from driver links

    Now
    –
    Target
    baseline then +25%
    By
    2 quarters

    Steps

    1. 1

      Write the line

      One sentence.

      Growth Marketing Manager, Voyage AI 1 day

    2. 2

      Add links

      Node, Python, Java, Go.

      Driver Team Leads 1 week

    3. 3

      Docker

      Image description.

      Developer Advocate 1 day

    4. 4

      Track

      By driver.

      Marketing Analytics 1 week

    5. 5

      Iterate

      Wording tests.

      Growth Marketing Manager, Voyage AI monthly

    6. 6

      Review

      Community feedback.

      Developer Advocate monthly

    First result
    1 month
    Cost
    $0
    Owner
    Senior Director Product Management, Growth

    Smallest test

    Add one line to one README.

    Why it works

    Add one short "Run it free on Atlas" line with a tracked link to each official driver README and package page, and to the Docker image description. Report signups by driver.

    It reaches developers already using MongoDB at no media cost.

    Risks and how to handle them

    • Open-source users see it as advertising

      Keep it short and useful.

    Evidence

    • Developer footprintThe mongodb npm driver has about 8.8 million weekly downloads; MongoDB's Docker images have about 71 million pulls; the mongodb/mongo GitHub repository has 28,656 stars, across 310 public repositories.Source
    • HiringSenior Director Product Management, Growth (United States, posted August 27, 2026): owns "the end-to-end product-led growth journey, from developer acquisition and onboarding through activation, adoption, monetization, retention, expansion, and sales handoffs" for "both MongoDB Community Edition and MongoDB Atlas", including "packaging, pricing, and monetization that accelerate self-service adoption".Source
  • 21

    MongoDB already publishes a Markdown pricing page for agents. Agents that plan a build also estimate cost, and a reliable quote favours the vendor that provides one.

    Medium impactLow effort First result in 1 quarter VP

    First action

    Publish the Flex and Dedicated prices as JSON.

    How to measure it

    Monthly pricing-file requests

    Now
    –
    Target
    baseline then +100%
    By
    2 quarters

    Steps

    1. 1

      Define format

      Tiers and regions.

      Senior Director Product Management, Growth 1 week

    2. 2

      Publish

      JSON file.

      Web Engineering Lead 1 week

    3. 3

      Reference

      llms.txt and MCP server.

      Documentation Lead 1 day

    4. 4

      Sync

      From pricing source.

      Web Engineering Lead 2 weeks

    5. 5

      Monitor

      Requests.

      Marketing Analytics monthly

    6. 6

      Extend

      Estimate endpoint.

      Software Engineer 3, Atlas Growth 1 quarter

    First result
    1 quarter
    Cost
    Engineering time
    Owner
    VP, AI Marketing Strategy

    Smallest test

    Publish one JSON file.

    Why it works

    Publish a small JSON pricing file and estimate endpoint (tier, region, storage, Voyage tokens) referenced in llms.txt and the MCP server, kept in sync with the pricing page.

    Agents that can price MongoDB accurately are more likely to recommend it.

    Risks and how to handle them

    • Out-of-date prices

      Generate from the same source as the page.

    Evidence

    • Your websiteMongoDB already publishes an llms.txt for the site and a separate one for the docs ("How agents should use this file"), plus a Markdown pricing page (updated September 29, 2026): Atlas Free with 512 MB, Flex "Up to $30/month", Dedicated from $56.94 a month, and Atlas Infinite "for unpredictable AI demand".Source
    • HiringVP, AI Marketing Strategy (San Francisco or Palo Alto, reporting to the CMO): exists "to secure the next generation of builders before they default to another platform"; owns "how MongoDB appears in AI-generated responses, agent framework recommendations, and developer tool suggestions", "ecosystem integration presence", co-marketing with "agent frameworks, deployment platforms, and frontier model providers" and "agentic builder acquisition"; treats "documentation quality, code example density, and dataset presence" as "distribution levers".Source
More example playbooks and reviews
ProgrammableDigital

Show MongoDB’s product working

Turn screenshots or the website into an animated product demo in its own brand, for the homepage, launch posts and outreach in this playbook.

Make a demo (opens programmable.digital in a new tab)

More growth playbooks

Other companies we have researched, each read from what they are hiring for.

  • Gale growth strategyA business-immigration platform (HRR Technologies, Inc., d/b/a Gale) that pairs structured intake, document collection, drafting and case…22 growth ideas
  • Pediatrics Plus growth strategyA specialized pediatric healthcare provider offering physical, occupational and speech therapy, applied behavior analysis, early…22 growth ideas
  • Outlook Amusements (California Psychics) growth strategyRuns California Psychics, an online marketplace for psychic, tarot and astrology readings by phone and chat with vetted advisors, plus…20 growth ideas
  • Haskell growth strategyAn integrated architecture, engineering, construction and consulting firm delivering food and beverage plants, healthcare, water and…21 growth ideas
  • RedTape growth strategyAn Indian footwear and fashion brand (listed as Redtape Ltd) selling shoes, sneakers, sports shoes, apparel and accessories for men and…22 growth ideas
  • Motive growth strategyAn AI-powered operations platform for physical operations: driver safety and AI dashcams, ELD compliance, fleet and equipment tracking…21 growth ideas
  • Plaud growth strategyAI note-taking hardware (Plaud Note, Note Pro, NotePin, NotePin S and the coming Plaud One) and the Plaud Intelligence software…23 growth ideas
  • AKQA growth strategyA design, innovation and AI agency (part of WPP) creating brands, digital products, AI-powered systems, campaigns and experiences, from…21 growth ideas

See every playbook and website review

Independent research. Programmable wrote this playbook from public sources, as they stood on October 22, 2026, and cites them throughout. MongoDB did not commission, review or endorse it, and Programmable is not affiliated with MongoDB. Company and product names are trademarks of their owners, used only to identify them. The ideas are suggestions and opinion, not professional, legal or financial advice, and not statements about how MongoDB runs its business.

Something wrong, or do you represent MongoDB? Email [email protected] and we will correct it or take it down.

30/60/90 roadmap

What to focus on in each month, and the ideas that carry it.

  1. Days 1 to 307 ideas

    Voyage front door and canonical docs, homepage weight, AI comparisons, credit-expiry emails and the tag audit; tag driver links.

  2. Days 31 to 6011 ideas

    Cross-sell, agent attribution, badges to clusters, product-qualified handoffs, LATAM pages, event follow-up, partner templates, India funds, the deployment chooser and account ranking; the pricing file.

  3. Days 61 to 903 ideas

    Commit to the big bets: the bring-your-own-data retrieval test, one command from Community to Atlas, and the Agent Engine gallery.

The research behind the plan

3 research areas, each summarised in a few lines. Open any of them for the full detail.

Customers

What customers say

Customers value flexible document model. The most cited problem: choosing between deployment options and tiers.

  • Fully managed Atlas on all three major clouds
  • Stitching separate vector, embedding and database tools
  • In their words: “RAG”

Market

Competitive landscape

5 competitors compared: Pinecone, Supabase, Couchbase, Amazon DynamoDB / DocumentDB and 1 more.

  • Gap to exploit: One platform from embeddings to production database
  • Pinecone: Separate from the operational database
  • Supabase: Relational model for document-shaped data

Hiring signals

MongoDB's open roles point at AI builders and self-serve growth.

  • Growth Marketing Manager, Voyage AI (US and Toronto): Voyage adoption, lifecycle and cross-sell. Ideas 1, 7, 17.
  • Senior Director Product Management, Growth: Self-serve journey and sales handoffs for Community and Atlas. Ideas 10, 18.
  • VP, AI Marketing Strategy: How MongoDB appears in AI answers and agent frameworks. Ideas 2, 8, 21.

From the team that wrote this playbook

All products
  • ProgrammableAgencyWorking AI systems in days, not quarters (opens programmable.agency in a new tab)
  • ProgrammableInfoAn agent that finds, researches and asks first (opens programmable.info in a new tab)
  • ProgrammableDigitalScreenshots in, an animated product demo out (opens programmable.digital in a new tab)
  • ProgrammableWatchKnow the day a supplier falls out of compliance (opens programmable.watch in a new tab)
  • ProgrammableTalkGet registrants for your events, and know who’s coming (opens programmable.talk in a new tab)
  • ProgrammablePageText what you need, your customer gets a page that works (opens programmable.page in a new tab)
  • ProgrammableCommunityForums with labelled AI members, never empty (opens programmable.community in a new tab)
ProgrammableMarketing
ExamplesLeaderboardsProductsTermsPrivacyCookiesDPALegalContact

© 2026 AI-Powered Growth Intelligence