Set up once. Watch your data grow.
Everything you need to install the tracking script, confirm your ICPs, map your developer journey, and start getting weekly agent evaluations.
How plans work
Every plan includes the full platform. Plans differ on two dimensions only: how many ICPs you evaluate and how many tracked sessions per month get AI summaries. Build ($299/mo) covers 1 ICP and 1,000 sessions, Growth ($799/mo) covers 5 ICPs and 10,000 sessions, and Scale ($1,999/mo) covers 15 ICPs and 50,000 sessions.
New Build subscriptions start with a 14-day free trial — card upfront, no charge until the trial ends, cancel anytime before then.
ICPs and weekly agent evaluations
An ICP is a specific developer audience: stack, role, and experience level. During onboarding, Built for Devs detects your likely ICPs from your product and docs, and you confirm or correct them. The moment you confirm an ICP, its first agent evaluation runs — an AI agent built to match that profile works through your product from docs to first value and produces a scored report — and it repeats weekly after that. Reports arrive by email and live in your dashboard.
Agent evaluation sessions are labeled in your session data and excluded from your real-developer metrics, so agent traffic never skews your funnel or TTV numbers.
Session caps
Tracking never stops. If you exceed your plan's monthly session volume, every session keeps recording and raw timelines stay available. AI session summaries pause for sessions above the cap until your next billing cycle or an upgrade. You lose narration, not data.
Install the tracking script
Every Built for Devs account comes with a unique tracking script. Paste it into the <head> of every property developers touch: your marketing site, your docs, your product app. If any of those live in separate codebases or subdomains, the script needs to go in each one. The more of the journey you cover, the more complete your data will be.
It loads asynchronously and won’t affect page performance. Once installed, it automatically tracks developer sessions (page views, time on page, and navigation patterns) without any additional configuration.
Requirements & compatibility
bfd.js is a web-only browser script. It works on web-based interfaces — your marketing site, docs, and web app — but it cannot be embedded in native mobile apps (React Native, Flutter, Swift, Kotlin). If your developer-facing product is a native app with no web surface, there is nowhere to install the script and verification will never pass. You need a web-based interface for tracking to work.
Your script tag
<script src="https://cdn.builtfor.dev/bfd.js" data-api-key="YOUR_API_KEY" defer></script>
Retrieve your actual script tag with your API key already filled in from your settings page.
Tracking across multiple domains
Your primary domain is verified automatically when the first tracking event fires from it. If developers also visit separate properties, like a docs site on a different domain, a subdomain such as docs.yourproduct.com, or a staging environment you want to include, you need to register those domains in Settings.
Go to Settings → Tracking Script → Additional Domains and add each domain you want to track. Events from unregistered domains are accepted but not attributed to your account, so add every domain where your script is installed.
Get your script →Verify your script is working
After installing the script, visit your site in a browser, then return to your Built for Devs dashboard. The “Last seen” indicator on your project will update within a few minutes of the first event firing. You can also check Settings → Tracking Script → Domain Verification to confirm your domain shows as verified.
If the indicator isn’t updating after 5 minutes, check these common causes:
-
The script tag is not inside the
<head>element, or it was placed after a blocking script that errors -
A Content Security Policy (CSP) header is blocking the script from loading. You’ll need to add
cdn.builtfor.devto yourscript-srcallowlist - A browser extension is blocking third-party scripts. Test in a clean browser profile or incognito mode
- The domain where you installed the script hasn’t been added to Settings → Additional Domains. Add it so events from that domain are attributed to your account
Filter out internal traffic
Your team visits your own product constantly: testing features, reviewing docs, debugging issues. Without filtering, that activity contaminates your developer funnel data and skews your TTV measurements. Set up exclusions before your data starts accumulating.
IP Exclusions
Go to Settings → IP Exclusions and add your office IP address or CIDR range (e.g. 192.168.1.0/24). Any traffic originating from an excluded IP is flagged as internal and excluded from funnel metrics. If your team works remotely, each team member can add their home IP, or you can exclude by email domain instead.
Email Domain Exclusions
Go to Settings → Email Domain Exclusions and add your company’s email domain (e.g. yourcompany.com). When a developer authenticates with an email matching an excluded domain, their sessions are flagged as internal and excluded from your metrics. This catches team members regardless of what IP they’re on.
Define your Time to Value
Time to value (TTV) is the moment a developer goes from curious to having gotten real value from your product. In the dashboard, it’s measured as the time from a developer’s first session to the TTV event you define.
How to identify your TTV moment
Ask yourself: what’s the first moment a developer has done the thing your product actually does? Not signed up, not viewed the dashboard, but genuinely used it. That’s your TTV. It’s the “aha” moment where the value becomes undeniable. For a monitoring tool, it might be the first alert firing. For an API product, it might be the first successful call returning real data. For a deployment tool, it might be the first successful deploy.
If you’re unsure, ask your team: “What’s the one thing a developer does where we know they’re going to stick around?” That’s your TTV event.
Two ways to define it
Go to Settings → TTV Definition and choose one of:
-
URL pattern: use this if reaching value lands the developer on a specific page. Enter the URL or pattern (e.g.
/quickstart/completeor/dashboard/first-project). The tracking script detects the page visit automatically, no code changes needed. -
Custom milestone event: use this if reaching value happens in the product itself rather than on a specific page. Fire a milestone from your frontend when the moment occurs, then enter that event name in Settings.
Firing a milestone event
Call BFD.milestone() in your frontend JavaScript at the moment the developer reaches value. The tracking script must already be installed on that page.
Example
// Fire this when the developer reaches value
BFD.milestone('first_api_call_success');
Use a name that’s specific to the moment, then enter that exact name in Settings → TTV Definition. From that point on, any developer who fires that milestone will have their TTV recorded.
Set your TTV →Map your developer journey
Built for Devs tracks developers through seven stages: Discovery, First Impression, Onboarding Setup, First Value, Adoption, Expansion, Advocacy. Journey mapping tells the system which URLs and events on your site correspond to each stage.
After installing the script, the platform auto-detects likely journey stages based on your URL structure. Go to Settings → Journey Mapping to review the suggestions, confirm the ones that are right, and adjust any that need changes.
The more accurately your journey is mapped, the more precise your TTV measurements and funnel drop-off data will be. If a stage isn’t mapped, time spent there won’t be attributed correctly.
Configure journey →Get human evaluation sessions
Your weekly agent evaluations catch what breaks mechanically. Human evaluations tell you what the experience feels like to a real developer, in their own words. Each human evaluation is $499, available on any plan, and a 3-pack is $1,299. Once your script is installed and your TTV is defined, you have enough context to make evaluation findings actionable.
How to get started
Go to the Evaluations section in your dashboard and purchase the number of sessions you want. During checkout, you’ll specify the ICP you want evaluated against. Be specific: the tighter the ICP, the more useful the findings.
What to expect
Once your purchase is complete:
- We match developers from the network to your ICP and invite them to run a session
- Each developer works through your product the way any new developer would, following your docs and onboarding naturally with the goal of reaching working implementation, while recording their screen and audio
- Each session is reviewed, and approved sessions are immediately available for you to watch in your dashboard
- Once all sessions are approved, your findings report is automatically generated for the project
Your findings report
The findings report synthesizes all sessions into a single project-level view. It includes an executive summary, a scorecard, critical issues, quick wins, what developers loved, and specific recommendations tied to your developer journey stages. Read it alongside your funnel data. The combination tells you both where developers are dropping off and exactly why.
Purchase evaluations →Connect your AI agent
The MCP integration gives your AI agent direct access to your developer funnel, friction signals, TTV trends, session data, and evaluation reports, without any manual exports.
Works with Claude Code, Claude Desktop, Cursor, and any MCP-compatible client. Your MCP key is available in Settings.
MCP integration docs →Understanding the dashboard
Adoption Score
A score (0–100) of how ready your product is for developers, synthesized from the evidence the platform collects: agent evaluation results, human evaluation findings, and live tracking signal. It moves as new evidence arrives, and the dashboard shows which layers it currently stands on. Scores are classified as Exceptional (85–100), Good (70–84), Needs Work (55–69), or Poor (below 55).
Agent evaluations
The weekly reports from your ICP-matched agent runs: an overall score, where the agent got stuck, what it misread, and how long it took to reach value. Each run includes a step-by-step path drilldown. Agent sessions appear in your session list with an “Agent evaluation” badge and are excluded from aggregate metrics.
TTV funnel
Shows the average time developers spend in each stage of your journey, Discovery through Advocacy, and where they’re dropping off before reaching your TTV event. Identifies the highest-leverage stages to improve.
Recommendations
AI-generated, specific improvement suggestions based on your funnel data, adoption score, and evaluation findings. Surfaced when patterns indicate a clear opportunity for improvement. Each recommendation carries a deliverable type (content pr, strategy doc, or advisory) and, when it's serviceable, a quote you can purchase as a result to have it written and shipped for you.
Sessions
Individual developer session recordings and timeline views. See how real developers move through your product: which pages they visit, how long they stay, and where they exit.
Evaluations
Human evaluation sessions, their status, findings reports, and full session recordings. Each completed evaluation links to a structured findings report with timestamped observations and recommendations.