At AutogenAI we are making our clients' bid writing teams up to 8X more efficient and boosting their win rates by up to 30%. One of the ways we do this is by unleashing the power of the written content that already exists across their companies. Now they can search & repurpose text from previous bids, marketing materials, annual reports, sales collateral, policies and any other document in their knowledge library, with the click of the button.
This feature is powered by AutogenAI’s proprietary Library Search Technology. In this article, we explain how it works.
Understanding AutogenAI’s Library Search Technology
To understand AutogenAI’s organisational library search technology, we must first familiarise ourselves with the foundational elements of modern search algorithms. This article aims to guide you through this learning journey. We will begin by first exploring traditional keyword-based search, a technique widely leveraged by internet search engines. Following this, we will shift our focus to semantic search, explaining its unique ability to interpret nuanced human language and context. Finally, we will explain how hybrid search works and how that applies to AutogenAI’s Adaptive Optimised Search Environment which is so loved by our users.
Keyword Search (The Traditional Librarian)
Imagine a traditional librarian who assists you in finding books by searching for the exact words or phrases you mention. If you inquire about books on “baking bread,” this librarian would search the library’s index or database for titles, authors, or descriptions containing the exact phrase “baking bread.” This method is straightforward and effective if you are certain about the specific terms you’re searching for. However, they might miss relevant books that discuss the topic without using the exact phrase “baking bread,” such as those that could cover “artisanal breads” or “sourdough starters” without directly mentioning “baking bread.”
Semantic Search (The Insightful Librarian)
Now imagine a more insightful librarian who considers not just the exact words you use, but also the meaning and context behind your request. When you ask for books on “baking bread,” this librarian contemplates the broader concept of baking bread, including related terms and topics like “bread recipes,” “yeast fermentation,” or “gluten-free baking.” This approach allows you to discover a wider range of books relevant to your interests, even if they don’t specifically mention your initial phrase. However, this method can yield results that feel off-target if the librarian’s interpretation of your needs doesn’t align perfectly with your actual intent.
Hybrid Search (The Versatile Librarian)
Finally, imagine a versatile librarian who combines the best aspects of both approaches. Hybrid search merges the precision of keyword search with the depth of semantic understanding, listening to your exact words while also comprehending their broader implications. This method ensures you receive both the specific results you asked for and a broader array of resources related to your topic.
Hybrid search alleviates the potential shortcomings of semantic search by grounding its broader interpretations in the specifics of your query, offering a more balanced and pertinent set of results.
AutogenAI’s Adaptive Optimised Search Environment
At AutogenAI we go a step further and use a proprietary algorithm that takes a weighted combination of semantic and keyword search that adapts to the user query to create an optimal search environment for each user query to pull in the most relevant information for their bids.
How It Works
Key word search (like Google) is very simple, it highly ranks the pieces of text that contain many of the key words matching the user query.
Semantic search is more complex as it uses AI models to cluster (group together) pieces of text that have similar meanings (“bread recipes,” “yeast fermentation,” or “gluten-free baking.”), regardless if any of the words match at all. Therefore, it can find other pieces of text that have a similar meaning, known as semantic similarity.
Semantic Similarity for Topics
AutogenAI’s adaptive optimised search environment uses a proprietary algorithm to search for the pieces of text in an organisation’s knowledge base that score the highest on both keyword and semantic search – based on the user’s query. We won’t bore you with the underlying mathematics but in simple terms:
AutogenAI Search Environment = Highest Scoring Results (Key Word Search + Semantic Search)
Conclusion
In the high-stakes world of bid writing, the ability for bid writers to quickly and effectively sift through their company knowledge base can be the determining factor in whether a proposal succeeds or fails. Bid writers are tasked with the immense responsibility of composing responses that are not only accurate and compelling but also fully tailored to the unique requirements of each RFP and which maintain a consistent tone of voice throughout and incorporate as much of the organisation’s existing knowledge and case studies as possible. At the moment, they rely on memory to do this.
With AutogenAI’s adaptive optimised search environment you can say goodbye to the days of struggling to remember where important information is stored, meticulously searching through documents, and tediously redrafting text. Now you can effortlessly retrieve, locate, and adapt all the information in your knowledge library with the click of a button and maximise your chances of writing a winning bid.
Want to learn more? Contact us today.