Buy vs. Make: Recommendation Engines
October 9, 2019

Delivering a truly remarkable digital experience requires you to personalize your properties based on your users’ persona and behavior. Classifying your visitors into different segments, identifying their unique needs and patterns across your digital properties, and finally, personalizing your brand’s online experience to match these preferences is, to say the least, not easy.

Why Recommendation Engines?

Aside from providing a great digital experience, Recommendation Engines have several other benefits for the publisher such as:

  • Increase Revenue: personalized recommendations drive high click-through-rate and therefore increase audience engagement. They increase PageViews, Time Spent on Site, Pages/Session, and decrease Bounce Rate. Increase in PageViews is directly linked to advertising revenues.
  • Content Discovery: the engines allow users to find out more content they are interested in than what was displayed to them otherwise.
  • Increase in content re-circulation, which is key to understanding your readers’ loyalty.
  • Saving your users time while increasing overall session time.

Should you buy one or develop your own?

In our opinion, publishers building a recommendation engine is surely an unwise decision for the following reasons:

High Cost

The price varies based on factors such as:

  • how much data you have and how complex it is
  • your business goals
  • cost of hiring talent
  • the existing technology you are currently using

In most cases, it is definitely more cost-efficient to purchase recommendation engines, due to the fact that the companies selling these solutions have been developing the latter for mass use and not on a single user.

Resources

There are many Recommendation Engines out there that offer poor performance, due to the complexity of building such an engine. A true recommendation system is based on vigorous data science – not something you can build by simply installing a plugin.

The evolution and development of the science behind recommendation engines goes back at least 20 years. The compounded experience of those dedicated to these solutions and their talent cannot be replicated overnight into a plug-in or a feature. A company that provides these tools has a more devoted team of engineers, data scientists etc. that have much more time on a daily basis to grow and improve the solutions, as compared to a brand new department you establish in-house.

Time Constraint

As mentioned previously, hiring in-house also means looking for the right candidate and going through the onboarding process and finally hoping your investment pays off – which is all very time-consuming with no guarantee. Additionally, companies do not have the luxury to take their time to test and enhance their solutions. The benefit of outsourcing is time and cost-efficiency because you have access to cheaper talent with the same high-quality proficiency.

Maintenance

Since it would just be a project and not your company’s main product, you won’t have time to keep up with the latest technology since you probably have other more pressing priorities. Whereas, for the company you hire, it’s their bread and butter.

Finally, recommendation engines are constantly improving. By subscribing to a good service, you get improvements over time without having to invest a ton of money upfront and doing your own research to keep up.

Recently, our team has announced the official launch of a new service that allows Publishers to affect what is returned to their visitors when using Magnet Personalized Recommendations: Magnet Rules Engine. Get in touch to see what Magnet by Klangoo can do for your news website.

Klangoo NLP
Subscribe to our Investors' newsletter to download the Investor Relations Brief & receive exclusive content!