The rise of composable technologies, post-pandemic ecommerce growth, and an uncertain economic climate have significantly shifted digital commerce technology investments. Companies are turning away from long-term back-end investments and toward front-end facing technologies that can prove faster time-to-value and measurable ROI.
Ecommerce product search and discovery solutions do just that. Elevating both the shopping experience and bottom line metrics, it’s no wonder that they’re at the top of the list for savvy leaders in digital commerce.
Offering solutions to improve the digital shopping experience, the vendor landscape has evolved rapidly since the hype of the pandemic years.
Digital commerce leaders often voice frustration about the confusion this causes in the evaluation process. It’s hard to keep up with the never-ending merry-go-round of mergers, acquisitions, and portfolio extensions for point solutions like ecommerce product search engines as the market coalesces around product discovery.
They’re the experts in digital commerce, after all. But the complexity of these turbulent times just doesn’t leave time to keep track of vendor technology market movements. And very understandably so.
Choosing the right product discovery solution for your team depends on your unique business needs, timelines, technology principles, and (of course) budget. Here are some critical criteria you should consider:
Ecommerce product search is not a new discipline, and many offerings are decades old. That means they may not be built on modern, cloud-native technology principles. What may appear like ‘stability’ in the marketing brochure quickly turns into an anchor dragging down your customer experience, agility, performance, and availability.
Look to vendors who offer a fresh technology foundation, using Microservices, API-First, Cloud-Native SaaS, and Headless (MACH) principles. That ensures speed, reliability, and a superior customer experience. MACH-certified vendors who are members of the MACH Alliance are a great place to start.
For AI that powers a product discovery portfolio, it is critical to pay attention to what drives the underlying learnings.
This is where vendor offerings differ dramatically. Some use outdated search cores like SOLR, whereas others are built on Elasticsearch cores. In both cases, the roots are in full-text search functions and focus on keywords, meaning that search results are not optimized for ecommerce use cases.
More modern, commerce-centric cores employ state-of-the-art machine learning models — like transformers, large language models (LLMs), and natural language processing (NLP) — in tandem with clickstream data. This allows them to create an advanced shopping experience that accurately displays the most attractive products to every individual while simultaneously optimizing for a range of ecommerce-specific metrics, rather than content conversion only.
Ask vendors about the foundation of the search and discovery core, and invest in technologies that are purpose-built on advanced AI for ecommerce use cases. Also, ensure the technology is proven to work for possible future initiatives, like AI Shopping Assistants supported by Generative AI (GenAI).
The success of the product discovery category has led to more M&As in recent years, in which large legacy vendors supplement their offerings with small niche solutions. Often, these technologies are not yet fully integrated, meaning that they introduce silos, breaking points, and missed opportunities for learnings across different moments in the product discovery journey. (We refer to them as Frankensearch solutions.)
Before you know it, you play software vendor by trying to plug those holes with your own development team.
Carefully investigate if your ecommerce product search and discovery vendor has integrated algorithms and products beyond the brochure. Be on the lookout for AI washing and run a proof of concept to validate cross-portfolio learnings leveraged in a common core.
Search algorithms, semantic models, NLP, and ML are complex fields of data science. Many vendors will throw hyped terms at you and expect you to trust that their offering will work for your needs. Flamboyant case studies with impressive stats follow. Your carefully crafted RFP checklist quickly becomes a full bingo sheet.
Ask how a dedicated Customer Success team will support tuning the core engine on your site or app and verify your live clickstream during onboarding. Every product catalog is different, as is every use case and every traffic pattern. Getting the most out of your product discovery partner means tuning the underlying core in your ecosystem and against your primary goals. If a vendor tries to convince you otherwise, treat it as a big red flag.
The most important aspect of making sure that your next ecommerce product search and discovery solution fits your needs is to run a live assessment as part of the evaluation process.
Over the last decade, search technology has changed fundamentally. AI-powered cores have gained mainstream adoption and cloud-native and API-first principles are becoming the de facto standard for scalability and data connectivity. Tools that were great 10 years ago may just not be up to standard anymore.
On one hand, marketing materials tout all the hyped buzzwords — but data science has evolved so far that evaluating the core search engines meaningfully requires a post-grad degree in Artificial Intelligence.
Understandably, technology buyers have time for neither.
Commerce search and product discovery is a technology choice that depends on real-life circumstances, and every vendor has their own sweet spot. Only a live assessment can tell if it will work for your unique circumstances.
Factors that might come into play are the volume and nature of your traffic, the differences across regions and channels, the nature of your product catalog (including the quality of attributes and metadata), and the breadth and depth of data integrations with adjacent customer intelligence.
And importantly, what does success look like for your ecommerce goals? Beyond conversion rates, do you care about revenue per visitor (RPV), average order value (AOV), margins, or maybe cost reduction in delivery or inventory velocity? Different tools optimize for different metrics, and the best tools are flexible to meet multiple goals and retrain to changing ones. But to know if they’re the best, you have to test drive them in your conditions.
And your commerce search and product discovery partner should be able to offer that opportunity.
Want to learn more about criteria you should consider when evaluating search and product discovery vendors? Download the official RFP template below.