The problem of information search in business is huge.
Documents are hard copy and digital. Which is the updated version? Where is that Purchase Order? The revision changed, but I don’t know where the print is? What are the lots for that part number? How many parts where shipped?
And the list goes on and on and on.
All day workers are receiving requests, searching for information, and passing it on. The fact that most of this is non-value work is a topic for another post (and software which we built over at Procurem).
This search time can account for 30% - 50% of workers time by our measurements.
How often do you receive an email, which causes you to go look for some archived email, or saved document, or worse, hike to someone else’s office to get the information.
Information search, retrieval, and communication is a massive, non-value adding time suck in most manufacturing and supply chain firms! It might be in other’s as well, but I’ll stick to what I know.
Systems can help. Software can help. Controls can help.
However, without knowing the exact situation, I am not sure how.
Here is a something we did to cut down search time for everyone.
Replace “Location” with “Index”
If you are 30 or up, you most likely came to know computers in the age of Windows (Windows 95 anyone?) pre internet. Every file was organized in a set of cascading folders. Organization was implemented through location.
Folders and files had names and were placed in locations that made sense and could be easily ascertained by anyone else.
Where are the Purchase Orders? In the Purchase Order folder, which has sub folders for each month. Anyone can find a purchase order.
This is almost impossible to maintain now for a number of reasons.
Increase in the three V’s
Information has much greater variety now than it did 25 years ago. If you created a folder for every unique type of information, you would have 1000’s of folders with only 1 or 2 files in it. This slows down computers, search, and retrieval time and is no longer practical. You would spend hours creating locations that would rarely be used.
Information comes at us with greater velocity. The speed at which we are getting pdfs, order requests, verbal commands, new prints, excel files with open orders, supplier lists, customer lists, and tracking numbers is astounding. Gather and correctly saving all of this in the correct location is impossible.
Information volume is growing. As the number of files we are saving increases, we need an increased number of subdivided folders in order to keep it organized enough for someone else to ascertain its location without a direct request to you.
Firms are being composed of closer working units within a global world. This means what might be an intuitive location for you, might not be intuitive for others in your department across the globe.
In 1998 Google was founded in Menlo Park, and 2 years later, in 2000, it became the premier search engine. Not only did the revolutionize how web pages were found, but how search was thought about.
Webpages are files. You don’t know where the webpage you are looking at is stored. You don’t know the name of the actual file either. Truth is, you don’t care.
You are looking for the information from the file, not the file itself. This is the best way to handle data in the current environment. Forget about locations, and instead focus on indexing with multiple indices.
I’ll get more into this in a minute.
Understand the most sought after information
To understand what indices you might need to implement this new type of information storage, search, and retrieval strategy, start with what you are looking for most.
For us, the most sought after pieces of information include:
- Engineering Prints
- Purchase Orders
- Lot Documents
- Quote Documents
We created the lowest number of folders needed to roughly parse the information that we received. For us it was 15 folders, the four above, as well some other basics like
These were then all indexed using OS software (Mac uses Spotlight)
Create or modify tools to locate those fast!
In our case, we built a custom tool, but here is one that will work just as good and has some amazing superpowers!
We connected Alfred to a shortcut (CMD + Space Bar). We can type this at any point, and the search bar will appear.
From here we can instantly search for a few things. Alfred searches everything including:
- webpages visited
- documents inside of apps like Quip or Trello
It also does many functions like calculators, or unit conversions (useful for dealing with GD&T or Metric Conversions for Shipping). It can control applications, convert timezone calculations, and run any custom script you give it. You can even build in complex custom workflows.
Alfred also comes with a number of features and customizations. Here are a few we did.
When you search for a file, it searches for the name. By typing “in” first, and then your search, Alfred will search inside of all your documents, spreadsheets, apps, emails, and other files for your search term.
When typing “find” before your search term, it will open up the folder containing the file you searched for. Sometimes I want to open up the entire lot folder, so I search “find” and then something related to the lots. The result will open up the containing folder rather than directly opening the file.
I search a lot of prints. When search for prints, I get a number of other files with similar names or information like demand plans, or purchase orders for that part. By creating a condition on “p”, I made it so it only searches the prints index. So by typing “p” and then some part, the results only show me prints.
This alone is worth the price of the software given the reduction in search time. We did this for lots and quotes as well.
You can also filter results by file type. I am often looking for a folder, but get all types of files. By typing “f” first, I can see only folder results that match my search term.
Line up systems and processes to complement
Once these tools are in place, you can adjust nomenclature and storage to complement the tools.
For example, we name all prints with a prefix consisting of a three letter acronym for each customer. Now when we type “p xxx- “ It automatically shows us prints only for that customer. So we used the “p” customization along with nomenclature for quick results.
There are endless ideas and scenarios with combining tools and process changes to reduce search time. If you are looking for any ideas or solution help, feel free to reach out.