The Search for Comprehensive Search

Back in the days before cloud computing became ubiquitous, finding the files you needed was pretty simple. But today, as organizations move some or all applications to the cloud, the files you need could be literally anywhere—on premise, in a remote data center or in the cloud. But that’s only part of the problem. Companies today are much more likely to use cloud-based apps like Slack for collaboration, along with cloud-only apps like Salesforce. These apps are rich sources of a company’s data. In fact, there are files everywhere—in emails, email attachments, repositories like Dropbox and Box, and in applications.

So how do you find the data you need today? Not easily, experts say. You would have to search each repository you use, both in the cloud and on premise, one by one until you find the file you need. It’s a slow and often frustrating process.

“It’s a very manual process. You try different search terms, and you still might not find what you’re looking for,” says Marc Staimer, founder of Dragon Slayer Consulting.

One solution is Enterprise File Search, but it only goes so far. These solutionsElastic, for example--have their place, but they don’t have the breadth and depth required by most companies today. Typically, they search they can search a company’s website and infrastructure, but can’t search everywhere files may exist, such as collaboration apps, email, Box or Dropbox.

And then there is Box, with its new AI partnership. It could be useful, especially for companies that rely on Box as a storage repository, but it only searches Box. For data workflows that involve more than Box, it’s only a partial solution.

That’s where Global File Search comes in. GFS is designed to search every file on every platform, regardless of where it is located. It should be able to quickly locate files by name, sender, date, file type, keyword, content and other attributes, regardless of where they are stored.

“It searches a lot of different repositories you might not even expect to be relevant to your search,” Staimer explains. You wouldn’t expect Atlassian or Salesforce or OneDrive to be a storage repository, but they are, because our definition of where we store files has changed.”

Global File Search is a tough nut to crack, but some vendors are making progress. One is Cloudtenna, which has developed a solution that searches everything—on-premise repositories, email apps, cloud file services and most hosted collaboration suites that store documents. It does this by setting up connectors in advance with all systems a company uses.

“The idea is that this will be the first thing you go to when you switch on your work terminal at the beginning of the workday. You would simply type in what you need and the file will come up,” says Aaron Ganek, Cloudtenna CEO.

A Deep Dive into GFS

So what makes a good Global File Search system? It comes down to speed, accuracy, security and intelligence.

Searching for the right file often takes more time than it should, and GFS technology aims to change that. In most cases, it should take no more than one second to find the file you’re searching for. Cloudtenna’s Direct Search, for example, takes no more than 600 milliseconds. It achieves this speed by using something called real-time binding, where much of the real work--such as scanning and indexing connected repositories, building indices and performing consistency checks--is done with the help of machine learning before queries are even issued. This approach means that users are actually searching a much smaller group of files, which takes less time.

A good GFS system also should be accurate and comprehensive. If even one potential source is left out of the mix, results could be skewed. Ideally, the system would find not only the file you’re looking for (on the first try), but also the most recent version of that file.

When it comes to any type of search, it’s critical that users be able to see results only for files they have permission to view. Cloudtenna’s DirectSearch accomplishes this by applying access permissions to its index of connected repositories. Real-time binding helps with this process by performing consistency checks to the index, so that any deltas are captured at the time a security change is made, Ganek explains.

An effective GFS also is intelligent enough to understand patterns and use them to better categorize data and return results. Search results can be ranked by contextual insights modeled on individual file activity, history, teams and relationships. With this insight, GFS system can rank search result by relevance to the user performing the search. Cloudtenna’s DirectSearch also makes recommendations as queries are typed, based on what the user is typing, the person making the query, and that person’s search history. And as it learns more about the users and the data it gets smarter over time, Ganek adds.

Spending time searching for files that may be anywhere is frustrating, and a poor use of company time, and it’s only getting worse. According to Forrester Research, file sprawl is the top challenge to finding files. Forrester calls Global File Search “the next generation of enterprise search designed to solve the file sprawl problem.” For this reason, intelligent content management services, including GFS, will hit the mainstream by 2022.