The Problem of Legacy Search Systems
Most AEC teams have experienced the frustration of poor search tools. The data exists, but legacy systems make it hard to retrieve. Typically, you’re forced to search using only high-level project info such as project name or folder rather than the actual contents of files. As a result, professionals waste time opening file after file to verify if it contains the needed information. There are usually two outdated search methods in play:
- Filename-only search: Using tools like shared drives or basic cloud storage, you can search by file name, but not much else. If the file naming isn’t perfectly descriptive (and let’s face it, it rarely is), you’re left guessing. One AEC technology expert summed it up: “You can only search by file name or by who took it and when.” This limitation means people resort to manually renaming files or remembering exact dates just to retrieve information which is hardly efficient.
- Attribute-based search: Some firms invest in in-house systems or project databases where you define preset attributes (project number, phase, building type, etc.) for each project folder. These allow filtered searches, but they only work if every piece of data was meticulously organised upfront. If a particular detail wasn’t entered as a database field, you won’t find it later. In many cases, the drawings or documents themselves aren’t text-searchable within these systems – the structured database might list a drawing by title and date, but you’d still have to download or open the PDF to see its contents.
The result of these legacy limitations is a huge amount of time wasted and knowledge overlooked. Users end up doing multiple searches and trial-and-error, often with slow, unreliable results. Non-text content (like scanned drawings or photos) is especially difficult to locate with old tools. It’s no surprise that industry surveys have found about 13% of construction professionals’ working hours are spent just searching for project information. As one analysis on connected data put it,
Teams spend excessive time searching for information, leading to delays, missed deadlines, and decreased productivity.
In an industry where every hour counts, this is a serious productivity drain and it also means valuable lessons from past projects often go unused.
Legacy search methods not only waste time, they also impede decision-making. If you can’t quickly find that past design detail or issue report, you risk “reinventing the wheel” or repeating mistakes simply because the knowledge stayed hidden in the archives.
Clearly, a better approach is needed to unlock all that latent know-how.
The Power of Tektome KnowledgeBuilder’s Unified Search
Tektome KnowledgeBuilder tackles this problem head-on by providing a unified search across your organisation’s project knowledge base. It’s only possible thanks to the way KnowledgeBuilder first structures your data. The platform uses AI to extract key information from unstructured files – drawings, documents, images, you name it – and turns it into a structured database of project knowledge.
In other words, it builds an organised index of everything (locations, dimensions, materials, issues, etc. from past projects) that can be fully customised by the user at any time. This structured data doesn’t need to be pre-defined, meaning the user can modify everything by themselves directly in the KnowledgeBuilder app. With this rich,structured and flexible data in place, KnowledgeBuilder offers powerful search features that legacy systems simply couldn’t support.
Below are a couple of the most powerful unified search features of KnowledgeBuilder.
Combined Attribute & Keyword Search
You are not limited to only one type of query or keyword anymore. You can combine multiple keywords and search across file and project attributes in one go until you narrow down to exactly what you’re looking for.
Problem / Task
Let’s suppose an architect is starting a new project. In the early design stage, the architect wants to review a list of similar past projects for inspiration. Later, when the design develops, the need becomes more specific – for example, finding detailed drawings of similar building components. Previously, this required knowing exactly which past project contained the relevant information, which was often time-consuming and limiting.
How advanced filtering solves this
With advanced filtering, this architect can run a cross-functional search across all projects. For instance, they can filter by project attributes such as location, usage, or size to quickly generate a shortlist of relevant projects. From there, the architect can apply file-level filters – such as drawing name or detail keywords – to pinpoint and retrieve the exact drawings needed, even if they are spread across multiple projects.
Benefits / Results
The architect no longer needs to know in advance which projects to explore. Since it’s impossible to “know what you don’t know,” advanced filtering helps uncover the right information at the right level of detail. This makes the search process faster, smarter, and better aligned with the purpose of each design phase.
Semantic Keyword Search with Highlights
A common pain in searching old records is that terminology changes – what one project called “seminar room” another might have logged as “meeting room”.
KnowledgeBuilder addresses this with semantic AI search. You can perform a similar keyword search and still get useful results. The AI understands synonyms and context, so you don’t have to guess the exact wording. Even better, when you open a result, the system will highlight the relevant text or annotation in the document and even snap to the precise section of a drawing where the term appears.
Problem / Task
Let’s suppose the architect is searching for something very specific that isn’t set up as a project attribute – for example, details of a handrail or a particular piece of MEP equipment. The challenge is that naming conventions vary across projects, making it difficult to locate exactly what’s needed.
How Similar Keyword Search solved this
With Similar Keyword Search, the system looks directly into the data itself and retrieves all information the AI identifies as relevant. It considers synonyms and related terms, suggesting documents that may not be an exact keyword match but still align with the architect’s intent.
Benefits / Results
This allows the architect to search effectively without having to be precise about the wording. Even if different projects use inconsistent naming, Similar Keyword Search ensures the right information can still be found quickly.
Summary
Legacy search systems in construction have long been a source of frustration – they’re slow, shallow, and require too much effort from the user. Important knowledge remained locked up in files because searching across projects, file types, and content was too difficult.
Tektome’s KnowledgeBuilder offers a much-needed alternative: a unified search that combines structured project data with powerful AI-driven keyword search. By structuring historical project information, it enables faster retrieval (spend minutes instead of hours finding the right detail), greater accuracy in results (search is deep and context-aware, so you get what you’re actually looking for), and vastly improved knowledge reuse across the organisation.
Schedule a demo with us today to find out how KnowledgeBuilder can help you power your project efficiency.
