In November I left Atlassian after 15 incredible years. I watched the company grow from 12 to 4,000 employees, working with amazing people along the way. I had to work out what I want “life after Atlassian” to look like.
Take some extended time off? I want to do that when our twin boys are a little bit older and we can create memories together.
Work for another tech giant? Potentially again in the future. Building teams and helping people grow has been one of the most rewarding parts of my job.
Start my own company? That is still on my list of life goals. I don’t want to look back when I’m older and say “I should have done that”.
But what really drives me is building great products. Solving real and interesting problems is what gets me out of bed in the morning.
The information accessibility problem
The amount of information we create on a daily basis is growing exponentially. But how we enable people to find that information has hardly changed.
Google has set the bar high. Finding anything amongst the 1.5+ billion sites on the internet seems effortless. But why do most other search experiences not live up to the same expectations?
Sure, most companies don’t have the resources that Google has. That’s certainly part of the reason. But ultimately it comes down to how we think about the problem. Search is a big data problem, yet most companies attempt to manually optimize their search algorithms. Often not even understanding their own data well.
The pace of innovation in the space hasn't helped. At its core Search technology remains static, unable to adapt to different use-cases. Let me give you a concrete example. Atlassian’s Confluence is an incredibly versatile tool. People use it for many different use-cases. Documentation, Intranet and many more. Each of those use-cases has different requirements in the context of Search.
While recency of content is important for an intranet, it’s less relevant for documentation. Yet, the developers had to make a choice to favor one or the other. And because Search is static and unable to adapt, the end result is that the search is optimized for a specific use-case. That’s not a problem unique to Confluence, that’s a problem with the state of Search technology today.
Smarter search experiences
Search.io is looking to solve these problems with the ambitious mission to connect people and information, starting with creating smarter search experiences.
Initially, the team explored to build on top of Apache Lucene. The community has done an amazing job building a technology that’s at the core of thousands of products. Unfortunately, Lucene’s architecture made it hard to add machine learning capabilities to the core of the search engine, while maintaining great performance.
To solve these problems, Search.io developed a new type of Search engine from the ground up. It allows for dynamically optimizing relevance based on reinforced learning. It basically understands which records are most relevant to users’ queries.
Aside from keywords entered by the user, Search.io can take into account a plethora of business data all at once for more nuanced relevance scoring. For example, in e-commerce search, Sajari enables the use of different combinations of stock levels, profit margins, and conversion rates when ranking search results, boosting product results accordingly. Most e-commerce Search engines only allow the use of these factors one by one to order results.
I’ve been following Search.io's journey for a few years and have been actively involved as an advisor for the past year. A fast-growing list of customers and amazing results when Sajari is compared with other solutions has always impressed me. When Hamish approached me about joining to lead product at Search.io, I got excited about the opportunity to take Search to the next level. I’ll be helping scale the great team and fantastic product we have in place to bring smarter search experiences to organizations of all sizes.
That’s why I joined Search.io.