The Diagnosis: Stuff Blind

My wife and I comment on “stuff blindness” – the ability to “not see” things we leave lying around the house. She doesn’t notice elastic hair bands that she takes out wherever she happens to be sitting. I don’t notice when I leave a pair of pliers or a screwdriver on a counter. Perhaps each of us has just become used to the ubiquity of that thing, or maybe some inner voice is telling us that we are going to need it again soon, so why put it away? Whatever the reason, each is unaware that there is a problem with that thing being there.

Similarly, many of our clients come to us with their own version of stuff blindness, only it’s all about their data. They’ve cobbled together a collection of names, addresses, notes, invoices, etc. that suffers from many problems, but they don’t see what’s right under their noses. It’s not that our clients are inherently messy people, or that they don’t care about the quality of their data; it’s just that they have stopped seeing the pitfalls of doing things in a disorganized or incomplete way.
As consultants, it’s our job to straighten all of this out. This begins with understanding how things got this way in the first place. Below are some of the things we look for.
Getting Old Sucks – Data goes out of date almost as soon as it’s entered. People move, change offices, get new phone numbers, and change their last names. And exactly zero percent of them feel obligated to let you know about it. Worse yet, they can’t remember who they have told and who they have not, so Elizabeth Cassiano, with whom you did business three months ago, is now Elizabeth Cassiano-Blightly. Or, possibly, Elizabeth Blightly. Tell you what, you figure out how to find her while she’s waiting on the phone. Or, maybe it’s just easier to create a new contact record and lose all the history you had on the old, single version of Elizabeth. Tempting, but highly confusing the next time she calls. 

Consistency is Not Key – Most apps are pretty specific about search terms. It’s an extremely involved task for programmers to build in “fuzzy logic.” Which means “McKinley Ave.” is not the same thing as “McKinley Avenue”. Type either of these as a whole, and you will not find the other. Another common problem data type is phone numbers — some people like to type (xxx) xxx-xxxx, while others prefer xxx-xxx-xxxx, or even xxx.xxx.xxxx if they are the artistic type. In my experience, it’s difficult to teach and enforce a set of rules for people to remember for this sort of thing, although it can be done. More on this later. 

Just Put It Anywhere – What do you do when you have two phone numbers and one phone number field? Why, put the second one wherever you can, of course. Notes field? Sure. Email address field? Why not, I don’t have an email address for him anyway. People will be able to find it, right? Right? Hello? It’s all fun and games until you need to search by that phone number, then you have a real problem. 
Knowing Right From Wrong – If you use many apps to accomplish your tasks — Excel, Airtable, and QuickBooks, for example — you have to enter information many times. Name, address, Information about work status, parts lists – whatever is essential to your business. So let’s say you have two different addresses for a business, and you know it only has one location. Which one’s right? You may have clues or processes overlaid on top of your data that help with this. You may not.  Modification dates are always helpful, if all of the places you store the same data automatically track that information. If even one of these does not, you don’t know which version is the freshest. This is a potentially costly problem that can lead to incorrect order fulfillment, missed deadlines, etc. Worse yet, it’s very common. Another way this problem manifests is when users enter more than one record for a thing — person, location, order, inventory — and again, it can be difficult to know which one is the correct or most up-to-date version. 

The Prognosis: 20-20 Vision

The news is good. Veteran developers such as myself are experienced at not only healing such ailing data as that described above, but we are even better at building clean, integrated environments for housing it once it’s healthy. Here are the keys to healthy data storage.

In a properly constructed business system, there is only one place where each piece of data is stored. The motto is, “enter once, use many.” In other words, there should be only one Customer record per customer. The customer’s name, address(es), phone number(s), etc. should each be stored in a way that allows them to appear wherever they are needed, but only as a reference, not as potentially conflicting information in two different parts of the system.
In a properly constructed business system, there is only one place where each piece of data is stored. The motto is, “enter once, use many.” In other words, there should be only one Customer record per customer. The customer’s name, address(es), phone number(s), etc. should each be stored in a way that allows them to appear wherever they are needed, but only as a reference, not as potentially conflicting information in two different parts of the system.
Making sure data can easily be found, using robust search tools that allow for variations in the results. You shouldn’t have any doubt in your mind that a search will give you a definitive answer on whether the record exists, and if it’s the right one. 
Once you have found it, you should be able to edit the data easily, empowering users to keep everything up-to-date and accurate. 

Data should allow for the possibility that there are more than one of some things. A classic example is people who have more than one phone number, or several email addresses. We design every system to allow unlimited entry of this type of data, to avoid the “just put it anywhere” mentality. 

Data entry should make it easy to follow the rules. If you always want 10-digit US phone numbers to follow the (xxx) xxx-xxxx format, we will format them for you, regardless of how you enter them. No one has to remember how it’s supposed to be done — it just works. 

Careful tracking of modifications. A good business system should have an audit log that tracks changes to the records (what was changed, who changed it, and when). This helps immensely when trying to figure out if a record is up-to-date or not. 

Eliminate the need for a cluster of independent apps. You should be able to accomplish everything you need with one well-constructed business system that is custom-designed to meet all of your needs. You may have a favorite or required-use accounting system to also work with, but that can be integrated to share information with the business system, so that the two are tightly integrated and always in sync.

Looking Ahead

Migrating data between older systems like those described above usually comes later in the development process. When we build you a new app, we plan it very carefully to accommodate the unique needs of your company or organization. We have the vision to see the endpoint for your new business system and can help smooth the way to a better, more functional, revenue-saving way of doing things.