To build or to buy?
So many strategists and implementation teams face the question of building their own solution with open-source technology or purchasing an out-of-the-box solution when looking at search options for digital projects. We sat down with Chris Brady, Partner Alliance Manager at Velir, to find out how they answer this question for their clients based on their 18+ years of experience with digital projects.
Are you struggling with this decision with replacing your Google Search Appliance on your website? Find out more about making this decision during our next webinar on May 29.
Q: Thanks for chatting with us, Chris. To start, is it cheaper to build a site search from scratch rather than buy it?
Chris: Thanks for the opportunity—happy to be part of the discussion. So, that’s a great question, and I’d like to address it from a slightly different angle, because to me, the question is more about the value that you can deliver to the audiences and customers that are interacting with your brand online versus the up-front costs. For simple keyword search, taxonomy-based faceting, etc. building a search solution from scratch and buying an out-of-the-box search platform can be comparable. However, with an increased investment and a platform like Coveo, for example, you can do a tremendous amount more.
Q: I am glad that you suggest focusing more on the opportunity cost rather than upfront investment cost. Do you often get requests for building a site search from scratch?
At Velir, every full-service digital project that we undertake has a search component. Many of these involve robust faceting, federation, authenticated search, exclusion logic and more to deliver on the client’s goals and audience needs. On occasion, the client prescribes a preferred search platform as part of the Request for Proposals (RFP), but more often, it is left to us to identify the best path forward based on what we believe will best fit the provided requirements. More than ever, RFPs contain a ‘dream list’ of features which, if built from scratch, would push the project out of the client’s budget range.
Q: If a customer has custom needs for a search solution, which platform do you leverage?
Today, we build search solutions with either Solr or Coveo. In the past, we have implemented Google Search Appliance for our clients as well, but with Google sunsetting that program, it is no longer a possibility.
We chose to focus on these two solutions so that we can address both the typical, more straightforward, search requests as well as the more ambitious. Solr provides excellent indexing capabilities, and we have built libraries to augment Solr with some features around stemming, PDF indexing, and weighting. That said, clients often request other features that are more involved to implement on Solr, which can take budget away from other client requests.
For our clients who are seeking an efficient way to provide their audiences with better digital experiences and more complex search capabilities, we highly recommend Coveo.
Q: What are the pros and cons of building with Solr rather than buying an out-of-the-box search solution like Coveo?
The challenge with building any digital search solution is that client expectations are set by the powerful search experiences they have on external search engines like Google or consumer sites like Amazon. These experiences instill preconceived notions of what a search platform should do “out of the box” and usually, the expected “standard” is actually quite advanced.
Our clients are often aware of the shortcomings of Solr when they reach out for help. Robust customization of Solr’s drop-in search solution to return more relevant results and provide more varied search features involves a significant and complex amount of development effort. Most of our prospective clients request a custom-built, federated solution with many facets, and may end up reconsidering once they realize the power of solutions like Coveo.
Q: What are the popular myths around building a site search solution?
We all experience highly sophisticated search mechanisms during our daily online interactions, so they can feel deceptively simple to pull off. Concepts like federated search, restricted content aligned to corporate governance rules, multilingual search, unique results layouts, and searching that includes content from 3rd party websites like a client’s YouTube feed are not possible out of the box in Solr.
One myth that we see often is that the search platform “should just group things the way I expect.” Our client wanted to show related assets in the search results by nesting them within the parent content item on the results page. With Coveo, we were able to highlight the event details page as the main result, and group “child” assets that were not actually on the page with that item. Related assets like transcripts, webinar recordings, collateral, notes, attendees, etc. were grouped correctly in Coveo. Consequently, a search for the parent item returned both the event page and its associated content, providing a much richer and more helpful search experience to the end user. This would have been far more difficult to accomplish without Coveo.
Q: What sets Coveo apart?
There are three areas where Coveo provides differentiated features, which resonate with our clients. First, the built-in connectors. All our clients have diverse and complex digital marketing environments. The ability to quickly add in content sources for search is a game changer. Clients love that we can deliver added functionality without the expense of building custom connectors.
Second, Coveo is available in the cloud, which is in line with our client expectations. Increasingly, we find clients are seeking solutions that do not introduce the overhead associated with running on-premise infrastructure.
Finally, for our Sitecore clients specifically, the ability to take advantage of Sitecore’s Experience Platform in conjunction with Coveo’s search capabilities is fantastic. The goal of any solution, search included, is to deliver desired content to the customer at the right time and in the right context. By combining the power of the Sitecore XP with Coveo’s machine learning and insights, end users benefit with the best possible results.
Q: You mentioned that customers want to reduce IT overhead costs while still ensuring they have the right tools available for their business users. How important is the post-sales support from vendors?
Support from any technical or implementation partner is vital to ensure that clients can get the most value out of their investment in these search solutions. As we start to index our clients’ websites, we inevitably run into scenarios where tuning and optimization guidance is required.
We have also seen that having the Coveo professional services team available to assist during implementations allows us to build highly customized integrations, which account for complex user access control scenarios. One of our clients has business rules that if a user was a member of one committee at a given access tier, that user should only see a certain selection of content in the search results and should never be shown the alternative information. This sort of business logic is made possible with Coveo.
Q: Thank you, Chris, for chatting with us about this common question. Our readers will appreciate it. Do you have anything else to add?
Working with Coveo over the past five years, we’ve been privileged to witness the growth of the platform and push new boundaries together, delivering cutting-edge search solutions for our clients. We are excited about Coveo’s next phase of evolution—the movement to the cloud and its new machine learning capabilities—as this aligns with how the industry, at large, is evolving and what our clients need to meet their changing audience and customer expectations.
Velir is a full-service digital agency that designs and develops enterprise web solutions for experience-driven websites. We specialize in Sitecore, Drupal and Adobe AEM web content management implementations.
Find out more about Chris and Velir by reading up on their recent case studies of successful digital projects for clients like Main Line Health, Kauffman Family Foundation, Brookings Institution, and more.