In the dynamic world of decentralized applications (dapps), swift identification and resolution of issues are critical to maintaining a seamless user experience and fostering customer trust.

The key lies in transforming traditional reactive troubleshooting into proactive health management of your dapps through data-driven engineering.

What is unique about data-driven engineering in web3?

Data-driven engineering involves making decisions based on hard data instead of hunches or gut feelings. Data-driven engineering empowers engineering teams to predict, understand, and solve problems before they affect end-users.

What are the challenges of data-driven engineering in web3?

Despite the immense potential that data-driven engineering offers, at present many web3 developers find themselves stuck in a reactive mode because of challenges such as a lack of proper tooling, the complexity of web3, and 24/7 nature of the blockchain industry.

1. Lack of Web3 Observability Tools

The lack of effective web3-specific monitoring tools means that issues often go undetected until they start affecting end-users, leading to frantic scrambles to deploy fixes. Web3 developers frequently end up sacrificing their peace of mind and nights of sleep, as they resolve active issues, which lead to downstream effects:

1. Users have a poorer experience
2. Developers are consistently working in high-stress, high-stakes scenarios
3. Issue fatigue threatens the overall success of the dapp

2. Traditional Tooling Does Not Cover the Complexity of Web3

Moreover, the situation is exacerbated by traditional analytics tools that, despite their strengths in other domains, often fail to serve the unique needs of web3. 

These tools don't capture the depth and breadth of data that developers require to anticipate and mitigate issues in a decentralized environment. 

From user interactions to smart contract executions to token transfers, the world of web3 requires monitoring and analytics that can handle its multifaceted complexity.

What are the advantages of data-driven engineering tools?

The three main advantages of data-driven engineering tools are: proactive issue management, performance optimization, and enabling engineers to make more informed decisions.

1. Proactive Issue Management

Real-time data helps in detecting potential problems early and mitigating them before they impact users. Potential issues that can be flagged early on range from performance bottlenecks to abnormal user behavior.

Some examples include:

• API consumption limits being hit during peak times
• Utilized nodes being out of sync
• Third party providers being unavailable
• Blockchain congestion
• Bugs in the frontend being deployed to production causing transactions to fail
• Detecting bad actors trying to interact with your smart contracts

A real world example was a dapp deploying a feature update into production leading to a significant increase in failed transactions, costing gas to the users without transactions successfully executing. 

After being alerted about this spike, the engineering team could quickly find the root cause, which was a bug in the dapp’s frontend trying to call a smart contract function by passing a hardcoded parameter that consistently violated a revert statement in this function. 

As the revert statement did not include an explanatory revert reason, finding the root cause without an observability tool would have impacted a larger number of their users. 

2. Performance Optimization

Analyzing data can help identify bottlenecks and inefficiencies, leading to optimized performance and a better user experience. Such performance issues include slow smart contract execution, oracle latency issues, inefficient resource usage, and issues with interoperability across the web3 stack. 

In a concrete instance, a development team using Blocktorch identified a recurring inefficiency within their dapp where a specific smart contract was causing higher-than-normal gas fees due to its complex nature. 

With this information, the dapp developers were able to refactor the smart contract to streamline its execution and significantly reduce gas fees. This efficiency gain not only enhanced the performance of their dapp but also resulted in cost savings and UX improvements for their users. This scenario illustrates how data-driven insights can lead to tangible benefits and a competitive edge in the Web3 landscape.

3. Informed Decision Making

Teams can make better decisions about feature development, resource allocation, and system enhancements by using insights derived from data. 

Observability provides visibility into how users interact with different features of the dapp and trends in usage over time. Such information can for example be the specific chains that are used more than others by your users (for best user experience, ensure the most used chain in your dApp is the default chain) as well as comparing this information to competing dApps. 

Another example is optimizing support for specific browsers or wallets that an increasing number of users are using. 

3 Tools to Help Monitor Engineering Issues in Web3

To become more data-driven as a web3 engineering team, it's crucial to incorporate tools that provide observability and in-depth data analysis. Here are a few tools that can help web3 engineers have great observability and monitoring into their applications.

1. Alchemy Monitor

Alchemy Monitor provides a comprehensive suite of dashboards for monitoring and analytics to understand dapp user behavior and Alchemy API health and performance. 

Monitor allows dapp developers to evaluate their API usage analytics by call type in order to better understand how to optimzie and debug their programs. It also provides a platform for creating alerts and digests about your application’s activity.

2. Blocktorch

Blocktorch is an Alchemy Ventures portfolio company that provides blockchain developers with end-to-end insights into dapps. Positioned as a full stack monitoring platform for web3 applications, blocktorch delivers an unrivaled breadth and depth of insights including frontends, blockchain transactions, smart contract performance, and each layer of the web3 stack.

Blocktorch enables engineering teams to not just react to real-time data, but also anticipate potential issues. They take proactive management a step further by identifying patterns, spotting anomalies, and visualizing these insights in a format that's simple to interpret. 

This empowers teams to not only comprehend the complete health of their dapp but also make informed decisions and deploy rapid fixes.

3. Tenderly

Tenderly is a smart contract developer tool to build, test, monitor and operate smart contracts from development to adoption. The suite of tools support engineers to test their smart contracts thoroughly with a powerful transaction simulator before deploying to mainnet and to identify issues in production by finding the specific line of code in smart contracts that lead to issues. 

5 Factors to Consider Before Choosing a Web3 Monitoring Tool

When choosing a web3 observability and monitoring tool for your decentralized application, consider the following factors: coverage, real-time alerting, data interpretation, scalability, and integrations with other providers.

1. Comprehensive Coverage

The tool should provide insights into every layer of your web3 stack, from user interactions to smart contract execution. Blocktorch gives engineers visibility into the frontend of their dApps, the smart contracts developed by the engineering team and also other protocols integrated with, and on demand the used oracles and decentralized file storage. 

In addition Blocktorch is interoperable with the web2 data standard Open Telemetry, to be usable in conjunction with any web2 observability tools in case the application is only partly decentralized but still relies on cloud services. 

2. Real-Time Alerting

The tool should offer real-time alerts for potential issues, enabling the team to rectify them before they escalate. Some of the most frequently used alerts set up by web3 engineers are:

• Average time of transactions pending in the mempool
• Percentage of dropped and failed transactions
• Sudden drops in usage which indicates potential faulty functionality
• Spikes in gas fees especially in relation to overall network gas cost 

All the listed examples directly impact the users as either their cost of using the dapp increases and they look for cheaper alternatives, or they face issues using the dapp which leads to churn. 

3. Easy Data Interpretation

Look for tools that provide easy-to-understand data visualizations and actionable insights. 

Out-of-the box dashboards provide insights in very short time, custom dashboards give engineers the flexibility to monitor exactly according to the team's needs. The optimal tool should support both. Dashboards offer a bird’s eye view on the systems to understand trends, spikes and liveness to then dive deeper into. 

4. Scalability 

As your dapp grows, the monitoring tool should be able to scale up seamlessly. For example, if your dapp is adding support for a new blockchain network, your monitoring tool should similarly be able to extend to the new chain. 

The tool and its pricing tiers should also be scalable itself, so the real-time functionalities work also when your dapp experiences an increase in usage. 

5. Integration

The tool should integrate smoothly with other tools and services your team uses. For example, integrations with IDEs for smart contract deployments as well as API providers for nodes and other services can be helpful for a smooth observability experience and the richest data. 

Conclusion

Adopting a data-driven approach is imperative for web3 engineering teams to deliver a seamless user experience. Tools like Alchemy and blocktorch, with their advanced observability features, play a crucial role in enabling teams to identify and rectify issues swiftly.