Limited budget and timeline
GunLib
Project overview
GunLib is an interactive knowledge base about small arms that combines a structured technical encyclopedia with an ecosystem for military personnel, instructors, and enthusiasts. Starting from a founder-instructor's basic idea, we went from express concept validation to creating a scalable MVP.
Industry Defense Tech • EdTech • Digital Library
Role Product designer and business analyst
Date 2026
Platform Responsive web (Mobile-First focus)
Project context
The project founder is an experienced military instructor at a Ukrainian Armed Forces training centre who specializes in firearms training. In his practice, he encountered every day that information about small arms was scattered chaotically across PDF files, closed forums, and outdated documents.
GunLib’s main goal is to centralize and carefully structure verified, well-organized, and accessible information about weapons and their sighting-in, so that the necessary instruction can be applied quickly and accurately in field or training conditions.
Constraints
Use in field conditions
Complex information architecture
Donation-based monetization model
Data accuracy and verification
Analytical stage:From research to MVP
Since time and resources at launch were limited, at the first stage we deliberately opted out of a lengthy classic Discovery. Instead, we applied a Lean MVP approach.
Competitor Analysis
Competitor analysis showed that the firearms website market was stuck in outdated 2000s formats or locked behind paid subscriptions.
- Filling a market gap: GunLib closes the gap between free but chaotic sources (forums, Reddit) and paid private catalogues (Firearms Guide).
- Transparent open-data model: Rejecting a hard paywall builds strong community trust and removes barriers to quick access to verified information and primary sources.
- Modern UX and specialized tooling: Unlike outdated competitor websites, GunLib offers a responsive Mobile-First interface, deep filtering, sighting-in modules, and cross-links between weapons and optics.
JTBD Framework
The JTBD framework transformed the chaotic needs of three segments (shooter, instructor, enthusiast) into precise UI mechanics and development priorities.
- Segmentation by usage context: Instead of presenting dry text, we created a branching system for user scenarios: from quick sighting-in adjustments to complete instructor manuals.
- Minimizing time to find the right data: Progressive Disclosure and quick anchors help users find the required instruction in less than five seconds.
- Deep navigation and caching: Deep Linking and autonomous caching ensure uninterrupted access to critical data blocks even with a weak internet connection.
Lean Canvas
The completed Lean Canvas defined a sustainable, low-cost business model by connecting the value proposition to the real market context.
- Expert content model: Engaging subject-matter authors and instructors makes it possible to systematically expand the knowledge base with verified materials while preserving high quality and data accuracy.
- Focus on background productivity: The key product metric (Minimizing Time-to-Find) became the basis for smart filtering and a minimalist UI.
- Entering the market through free access: Replacing a paywall with donations, Patreon, and future B2B/API offerings lowers the entry barrier and builds user trust.
AARRR Funnel
The pirate metrics funnel focuses on organic product growth and natural retention conditions without excessive spending.
- Immediate value in the first session: Users get value immediately after searching for technical specifications, without mandatory registration, reducing the bounce rate.
- Organic audience acquisition loop: Direct links to specific page sections help attract potential users through sharing useful information.
- Ecosystem monetization: Converting a free user into a patron is built on a sense of belonging, community access, and an exclusive backlog.
Product Roadmap
The GunLib product roadmap creates a predictable path for development, avoiding premature complexity and helping allocate MVP resources.
- Reducing development effort: Complex tools are deferred to future updates so the value of the core version can be validated first.
- Evolution of the workspace: A modular page template with basic content management can later become a full platform for collaborative authoring.
- Transition to a partner ecosystem: A deliberate path from initial community feedback to automated support through Patreon and partnerships with manufacturers.
1 / 5
Behavioural analytics: Validating UX hypotheses in the real environment
After launching the MVP, we analysed behavioural data in GA4 and Microsoft Clarity: devices, interactions with key elements, navigation, and early signals of support for the project.
Mobile traffic dominates — supporting the Mobile-First priority:
During the research stage, we hypothesized that GunLib’s key usage context was the quick search for technical specifications, instructions, and sighting-in diagrams directly in the field or at training locations.
In the Google Analytics 4 report, covering approximately 13 July — 11 August, mobile traffic accounted for (78.05%), supporting the Mobile-First priority. At the same time, the data shows differences in interaction depth:
Higher share of engaged sessions (51.0% vs 44.24%): The mobile engagement rate is slightly higher than desktop. Average engagement time is 1:27 versus 2:04. This signal helps define the Mobile-First direction; the next step is to explore how different mobile scenarios affect interaction depth.
Mobile is the primary device for new users (79.03%): Around 80% of new users first opened GunLib on a smartphone. This supports providing instant access without mandatory registration. The next step is to check the effectiveness of this approach through a separate funnel.
A working hypothesis about different device scenarios: The mobile version may support quick, focused tasks in the field, while desktop may serve extended research and preparation of materials. In the next iteration, we plan to compare these scenarios with path, content, and task-completion data.
Interaction analysisacross key platform sections
Interaction with practical instructions:
In the documentation block, the highest density of clicks was recorded on the view and download icons for sighting-in instructions, quick-reference notes, and official manuals. This shows that GunLib is used not only as a reference, but also as a practical tool for preparing and working with firearms.
Users interact with the interactive diagrams:
Within the “Parts and mechanisms” section, users clicked the numbered markers and component labels. This shows that the diagram functions not only as an illustration, but also as a navigation tool.
An early signal of interest in supporting the project:
Users clicked “Support the project”, and a few donations came in after launch.
Use of anchor navigation on desktop:
The heatmap showed grouped clicks on the side menu of the desktop version: “Specifications”, “Sighting-in”, and “Main parts”. This indicates that users move directly to the relevant topic block instead of always reading the page sequentially.
Yet some hypotheses were disproved
Analytics helped us identify successful decisions alongside areas where our initial hypotheses need to be revisited. For each signal, we defined a next decision or direction for further validation.
Low engagement with text search:
Only 12% of mobile visitors used the search field, preferring direct links from messengers or navigating through the catalogue. Possible causes include low need due to the small weapon catalogue, low visibility, or a preference for direct links and the catalogue.
We plan to move search into a compact header icon. The next step is to track search opens, successful queries, and result clicks.
Minimal interaction with mobile tabs:
Users rarely switched between the section tabs fixed in the footer. This may be related to low element visibility or a preference for continuous vertical scrolling.
The next step is to test alternative adaptation formats and compare navigation and completion of target actions through an A/B test.
A dead-end scenario after reviewing specifications:
After reviewing specifications, users had to return to the Home page to choose the next model. This added unnecessary steps to the scenario of exploring several models in sequence.
We planned to add “Other weapon models” blocks at the bottom of the page and one-click switching between cards.
Not the end, but the beginning
Data from real sessions helped us reject ineffective hypotheses in time, refine and reduce unnecessary user actions, and adjust our improvement plans.
Work on GunLib continues.
