Optimizing Business Operations: The Steve Dahl Network

The Steve Dahl Network (SDN) is a subscription-based WP-built podcast service with satirical talks about life. Steve Dahl, the owner and creator, has been running the whole project since its inception in 2011. With a growing number of subscribers, however, the numerous tasks required to service his listeners and the technical backbone of the system generated a lot of responsibilities for the owner and his team.

To be able to have more time to deliver best comedy podcast content and worry less about the technicalities, Steve integrated the SDN with Libsyn — a podcast-hosting platform for subscription-based content delivery. The integration freed Steve from roughly 90% of technical work, which included:

• Managing hosting and upkeep of servers
• Uploading content
• Implementing security
• Administering payments
• Controlling access and subscriptions

Integration with Libsyn decreases the technical and servicing burden on podcast hosts, but in case of Steve it also required an update of his website, including appropriate redirections for stray links in public search engines.

That’s when Steve reached out to us.

The migration was a multi-step process that required cooperation and coordination, with many challenges spread across the whole project.

What We Did

1. Rebuilt the entire website with multiple pages into a one-page landing site with references to podcast station.
2. Transferred the website to a fast and cost-effective server.
3. Fixed SEO Google indexing from the old website to the new Libsyn location and/or to the new www.dahl.com site. The process required complete SEO remapping: multiple 301 redirects/RSS feed redirects.
4. Performed a timely and undisrupted launch of the new website.

Outcome

• Optimized costs—dedicated and expensive server replaced with a fast and cheaper alternative hosting service.
• Updated podcast content management experience in the Libsyn’s environment.
• Decreased amount of technical workload—increased ability to focus on creating content.
• Faster, easier, and better service for subscribers.
• Fast, intuitive, and user-friendly website.

The Big Data Migration—Legia

Legia Warsaw is one of the most famous Polish football clubs. In 2018, to stimulate fans’ brand engagement and continue building brand loyalty, the club decided to build three new web applications.

The new apps called for migration of data from the old system to a new architecture.

There were several challenges along the way:

• Converting the data so that it fitted the new system—the data is often structured differently between old and new systems.
• Ensuring none of the information on 14,000 users, such as loyalty points, personal and profile data, or access privileges, got lost.
• Importing years-long data on stadium entries.

The third task was particularly tricky because the import, without special mapping, would have taken us very long to complete. But, with millions of data entries, saving one millisecond per entry would save us days of processing and be more cost-effective for the client. So to speed up the migration, creativity and know-how of our migration engineers allowed to find those millisecond-per-operation savings to map the data so that the transition took the least amount of time possible while data accuracy was retained.

Outcome

• Fast and efficient system that quickly retrieves and feeds data.
• Better fan interactivity features.
• Innovative approach to fanbase cultivation.

Bypassing Limits—Reworking the Backend to Speed Up the Frontend

Another company we helped migrate provides a variety of IT consulting services to a broad range of organizations.

Because of the way the company’s website was built, it had inherent performance issues that significantly affected user experience, stunting growth and disrupting the stability of business operations. To fix it, we had to resolve to unconventional and inventive development approaches.

The website was based largely on the integration of external services implemented on the backend. This is a fairly common mistake when web applications are built without due consideration given to a possible future increase in traffic. The problem this setup generates is that the site won’t be displayed until all requests to the backend have been made.

What We Did

To speed up load time, our team rewrote the application so that the app downloaded the necessary information asynchronously directly from the frontend instead of waiting for the backend.

On top of that, we created a simple interface for the distributed database structure so that it could operate on the data generated by the client. We have also developed a new database schema, along with a new quick API tailored to the needs of the site.

Outcome

The new version of the website brought many business improvements:

• Enhanced presentation of promotions and discounts.
• Upgraded offer views.
• Boost in performance.
• Better reach with added English version of the website.

When There Are Thousands of Products: The Nitty-Gritty Hassle with Attributes

Migrating to a newer technology is especially important in the life of a big web store. However, for some online shop owners, the decision to migrate is hard to make. The reason why is quite simple: with hundreds of thousands of products in the database (sometimes multiple databases) and related sources of information, ensuring all i’s are dotted and all t’s are crossed during large migrations is the key to undisturbed business continuity.

One of our clients owned an online store selling over half a million products. The complexity of this migration was compounded by special categories of sold goods, which had many product attributes expressed in different values (numerical, descriptive). There was also business logic embedded in product categorization, for instance, replacement parts that may fit multiple makes, models, years, and trims of vehicles, mixed with engine attributes and specifics of the powertrain.

The situation got even more complicated when those goods and information about them were supplied by multiple manufacturers—the more manufacturers, the wider disparity between values and descriptions provided.

Thus, the successful migration of such incongruous datasets from multiple sources required meticulous data mapping: preparing the outline of relationships between the source data and target data model so that the converted database retains the original information but is best optimized for latest usage and expressed in a new format.

Data migration in databases with thousands of products for e-commerce had numerous challenges:

• Ensuring SKUs matched and were unique for each product.
• Unifying data from different sources to meet the requirements of the target data model.
• Linkage to web-optimized photo galleries of product images.
• Appropriate management of stock and price information.
• Deletion of missing or erroneous data.

What We Did

Because we’re talking about a large number of products, the migration would have taken days or even weeks to finish when performed without any changes in standard data mappings. To complete the migration in a timely manner and without loss of information, we split the process into multiple phases (some happening in parallel).

We used separate servers tackling different concerns:

• Prices and stock availability
• Image galleries of products
• Attributes data and uniqueness of SKU

Outcome

This strategy allowed for a dynamic implementation of data into the system. The client could see incremental results of the migration within days after its start: the products were gradually accumulating in the new database, being readily available for browsing in staging site’s search engine.

Migrating Toward Optimized Business

Migrations come in many flavors and levels of complexity. From small technology migrations that require simple upgrades and compatibility fixes, to monolithic data migrations that have millions of entries, to system migrations from dedicated legacy technologies to innovative cloud-based solutions.

Regardless of the flavor, the aim and outcome of a migration should always be growth, increase in performance, better operational efficiency, and the introduction of features that further improve customer and user experience.

Polcode is an international full-cycle software house with over 1,300 completed projects. Propelled by passion and ambition, we’ve coded for over 800 businesses across the globe. If you need to transition to a new system, but you’re not sure where to start and how to go about getting the most out of your existing data, contact us. We’ll make sure the migration is painless and fast.