Tech Transfer Central
University-Industry Engagement Advisor
ARTICLE REPRINT

Consistency, standardization are keys to solidifying your TTO’s data integrity

This article appeared in the January 2020 issue of Technology Transfer Tactics. Click here for a free sample issue or click here to subscribe.

Don't forget to sign up for Tech Transfer eNews, our free email newsletter filled with helpful tips, industry news, special reports, and key legal and regulatory updates, sent to your inbox every Wednesday!

You'll also receive info on upcoming webinars and other related products.

It takes a good deal of time to plan properly when implementing a new database (or cleaning up an existing one) and instituting processes that will lead to greater data integrity. However, failure to do so will take even more time and will ultimately cost more money, warned a pair of experts from Fuentek, LLC, the presenters in a webinar sponsored by TTT entitled “Improving Data Integrity: Boost TTO Operations and Improve IP Management Database Reliability.”

“Take time at the beginning of the process,” advised Fuentek Vice President Becky Stoughton. “If you think about [possible pitfalls] ahead of time there is a greater potential to minimize them.”

So, what exactly is data integrity? “Data integrity essentially means you have accurate and complete information within your database system so can get to it quickly and easily and can provide it in reasonable amount of time,” said Laura Schoppe, president of Fuentek. “How you present it — and to whom — takes some thought. The strategic [approach] is also important.”

What causes poor data integrity? One of the main causes, said Schoppe, is missing and incomplete information. Another, she said, is inconsistent data entry. Some very common reasons behind those problems, she noted, include lack of adequate training, lack of clarity in assigning database responsibility, and a lack of standardization in how data should be entered and what should be included. Other causes of poor data integrity, she added, include staff turnover or changes in jobs or responsibilities within the office. “Another is lack of time — people say there are a lot of steps and they do not have the time,” said Schoppe.

Here, she echoed one of the main themes of the presentation. “I’m here to tell you it will cost you a lot more on the back end,” she warned. “Entering the data completely and accurately will pay off manifold on the back end.”

To put it simply, she said, it all depends on having good standard operating procedures (SOPs), training people properly, clearly defining responsibilities, and establishing standards for what data to include and how it should be formatted.

Paying the price

Many of the impacts of “bad data,” Schoppe noted, are “fairly obvious.” Errors and omissions, for example can lead to significant inefficiency, decreased productivity, and as a result, increased costs. “We had a government client,” she recalled. “While we had meticulous records [of their disclosures], they had poor data management and would get requests for information from external parties.”

Fuentek, she said, would get “fire drill” calls from the client, asking what technologies they had in specific vectors, and which were the top-ranked assets. “We were able to do it, but the fact that they couldn’t was very concerning,” said Schoppe. “They have fixed things since then, but it’s a good example of how [bad data] can have a negative impact on how you get to good opportunities or cause serious problems internally and externally.”

People “get unhappy with you,” she continued, if you can’t get them data on the fly. “You may also have the potential of missing out on revenue,” Schoppe observed. For example, a university client was running their database essentially off an Excel spreadsheet; it had become their primary way of managing data. “It was not complete; it was not connected to additional things like licensing activity,” she explained. “They had not asked to see licensing revenue for years and were not collecting everything. The office did not have good data and was perceived as inept and disorganized. These clients did miss out on opportunities when they were not able to get to data as quickly as possible.”

In short, she summarized, the old adage “garbage in/garbage out” really applies. “No matter what system you use, how you enter data has a huge impact on output,” Schoppe asserted.

Data impacts multiple stakeholders

So, who is impacted by your poor data integrity? “The TTO itself cares,” said Stoughton. “You need to make sure your data is reliable; what projects have the highest priority? Who is the manager on what project? You need accurate due dates, so no balls get dropped.” In short, she said, you must understand the data you’re working with and keep it on hand — especially if there are staff transitions.

Beyond the TTO itself, she continued, are many stakeholders — for example, inventors who want to know the status of their inventions. “They will want more data from the TTO, not less,” she asserted.

Another stakeholder is institutional leadership — whoever the TTO reports to. “They want reports; they want the data sliced and diced,” said Stoughton. “Department heads may want annual reports, or data for promotions or tenure.” Administration, she added, wants to understand the overall picture; it can help with alumni solicitation, reports to the legislature, with annual reports, and so forth.

Partners or potential partners, start-ups, institutions, and research sponsors, she continued, are all interested in your data. “It not only helps identify areas for good matches, but often they have the obligation to report to invention sponsors — and often within specific time frames,” Stoughton noted.

Finally, she said, agencies and organizations, such as NIH or AUTM, are interested in seeing your data.

“More subtly, it really behooves you to think about the data you provide and how you provide it; it can get people in your favor or not,” said Stoughton. “Think about how the data is sorted, grouped, and filtered; it can impact funding or perception of performance.”

Preventing pitfalls

Stoughton shared her recommendations for ensuring data integrity, whether that means cleaning up a mess or preventing one in the first place.

“In either case, the basics are the same,” she noted. “Just like the key to real estate is location, location, location, the most important factors for data integrity are reporting, reporting, reporting.” It’s important, she noted, to think about what reports you will need and then build a structure and process to ensure you get them.

Then, she continued, consider who wants what; think about your audience. This can include, she noted, all the stakeholders mentioned earlier. “How do they need the data? How often do they need it?” she posed. “Make sure to collect it in a way that supports their timeframe.” This, she said, requires that you interact with those stakeholders. “You will not know all the answers without talking to them,” she asserted. Equally important, Stoughton advised, is to be sure to document what you hear. “Documenting how the reports should look is very helpful,” she emphasized. “Do they want lists? Graphical charts? Reports often involve grouping, or filtering, so understanding that from the beginning is also helpful; we call that ‘slicing and dicing’ the data.”

It’s also helpful to think of the pros and cons of various formats “right from the beginning,” said Stoughton. Once you understand the types of data, she advised, identify the key fields in the system where you’ll be capturing them and define consistent standards to use so you can sort and count the data. For example, fields might include organizations and contacts, patents, agreements, and invention status.

“It’s really important to define standard formats for the relevant field,” she continued. “If they are centers, or departments, or colleges, you do not want a person entering a center one way today and another person using a different name a year from now.”

If you are working with a new data system, said Stoughton, “begin all of this ASAP. Think about what IT management system makes the most sense for your needs. Consider things like price versus budget, and the size of the office over the next few years — expected number of disclosures, staff size, and so on.”

Excel, said Stoughton, is probably the most basic and inexpensive system. “But even if you use that it’s helpful to explore powerful capabilities like standardizing columns or dropdowns; they can be very helpful,” she shared.

Establish SOPs

This is also a golden opportunity to document SOPs from the start, she added. “They needn’t necessarily be complicated,” said Stoughton. “Minimal outlines with appropriate screenshots go a long way. Start with frequently conducted procedures and add as time permits. It’s easier to get people to follow the rules once data integrity is already there.”

If you’re in the process of cleaning up an existing portfolio, the first thing to do is characterize the situation — how the data gets there, who enters it, and where they get it, said Stoughton. “Explore what reports you can pull and which you use and rely on, and which ones you couldn’t pull and why you can’t — what data do you believe, and where the holes are,” she suggested. “It’s important to recognize that this is going to take time, and require interviews, meetings, and digging through the database.”

Only then, she continued, can you start clarifying goals for going forward. “That’s important, because it’s impossible to do everything at once,” she pointed out. “Are you trying to improve technology marketing? Then focus on getting clarity on the IP itself, and its categories. But if your emphasis is more on increasing revenues, then focusing first on agreements in place and status, and understanding what’s due to you might be one of your early priorities.” Once you have your goals and plans you can start chipping away at things, Stoughton continued. “Once people see that, you will get support and people will be motivated,” she asserted.

Stoughton also encouraged TTOs to remember this old adage in terms of prioritization: ‘Don’t fix what ain’t broke.’ “Figure out what the worst pain point is and start from there,” she advised.

Pain points, anyone?

Schoppe drilled down on Stoughton’s advice with some “nuts & bolts” examples — one of which came from a real-life experience with a university client.

“They could not come up with a good number of how many inventions and patents there were associated with different departments and colleges,” she recalled. “Various people entered data whichever way they wanted to and did not necessarily put departments in the right colleges.”

The database clean-up effort started off with over 90 departments in 11 different colleges, she shared — but there were really only four colleges at this institution. “Once we did the clean-up and re-mapping, we ended up with the four colleges that were actually true and only 40 departments — so less than 50% of what they had in there truly should have been,” noted Schoppe. After that, she continued, they were able to run accurate reports.

The school had over 900 inventors, and since this was not an automated process, they had to look at each individual record. “This was super time-consuming and therefore expensive,” said Schoppe. “If they did it right the first time, they would not have had to spend that money.”

Another common pain point, she continued, is often focused on patent data. “We see commas and slashes and spaces used in how patent numbers are put in, but worse yet we will see inconsistency in entering all data,” said Schoppe. “Initial filing and publication may be put in — but as new records rather than as continuation, which gives you a miscount. A small institution with about 200 inventions had over 20 published patents — but the system had only four. At the same time, they were put in a lot of files, so the data was in the system but not entered specifically in the fields — just attachments. It took a lot of time to hunt a patent down and find it.”

Most of the systems in use today, she pointed out, are pretty sophisticated when it comes to helping you track financial data, “but if you do not enter the terms correctly, they can’t help you,” she emphasized. Or, she added, you may not be entering the data on a regular basis. “Systems can help you by setting up alerts; use them to full power,” Schoppe urged.

Inventions, she noted, tend to be the “least messed up” aspect of a database because there is usually only one person — or only a few — involved in the reporting. “If you use electronic information there are fewer errors, and important areas tend to be clean,” she said. “But the farther you are in your process with patents and agreements, you see less consistency. That’s not to your advantage, because patents and agreements tend to be the data the administrators and inventors are more interested in.”

If you leverage dropdowns, you can come up with good consistent lists on types and status (i.e., disclosure status, invention status, agreements), said Schoppe. “This helps in managing activities, but also in monitoring on when you need to do things. If you know the date of a provisional patent you can send an alert, or when you need to nationalize. These are very important to maintain.”

E-mails, noted Schoppe, “are the bane of my existence.” When e-mails arrive, she pointed out, everything within them may be attached. “That may also mean logos, which are junk and you do not want,” she explained. “You may see correspondence with an attorney included in a patent application; that’s great, but there may be three additional image files. That’s just cluttering up your system.”

One large institution Fuentek worked with had 8,000 files, 3,200 of which were e-mails, she recounted. “About 10% were logo images; we couldn’t differentiate between them and the images of inventions,” said Schoppe. “So, we set up procedures for how to manage e-mails and delete things that really shouldn’t be in there.”

Focus on standardization

And how do you avoid most of these pain points? “Just get standardized,” Schoppe said. Take, for example, patent numbers. (See Figure 1.) “We always do it the same,” she said. “You can cut & paste and get out quickly without a lot of ambiguity.” You can choose your own version, Schoppe noted, but “just be consistent. Or, you could use ours and leverage it.”

You should also standardize how to set up files and folder names. “One large organization had over 1,000 disclosures but only 140 attachments; [the others] were not catalogued as invention disclosures because so many were not identified as type,” said Schoppe. A smaller institution with 200 inventions had only seven typed as disclosures, she added. “A lot of them had the file name ‘tech assessment;’ you’d have to download, then go in and read them.”

Tags and categories are also very helpful., Schoppe continued. “Very often tags are used on the web now,” she pointed out. For example, if you use categories like “nanotechnology,” and if rankings are noted, you can pull them out and show stakeholders the five best in a given category.

“Standardize how you set up people in an organization,” Schoppe added. “If somebody leaves, how will you flag that? Will you track their new contact information? How will you manage that in the system? Just decide and be consistent.”

Schoppe offered an example of setting up a process that can go a long way toward ensuring that consistency — in this case, IP status flow. (See Figure 2.)

“This shows what to do when, types, and statuses,” she noted. “This can be done for any process — like agreements. It will help you clean up a process and stay on top of the data you enter; that translates into what an SOP looks like — this happens, that happens, what the triggers are, how to enter in the field, and what the format of the information should be.”

Finally, she summarized, make sure you talk to your database providers to minimize time wasted, and possibly create more efficient ways of cleaning up your data. “Make sure to be reasonable about what you choose [to clean]; you will not be able to get everything 100%. Accept that early on in the process,” she emphasized. “Identify what things are really not worth it under the ‘80/20’ rule, change the data entry process, and know going forward that the bad data will expire.”

Contact Schoppe at 919-249-0327 or laschoppe@fuentek.com; contact Stoughton at rstoughton@fuentek.com.

Editor’s notes:

The entire recorded webinar “Improving Data Integrity: Boost TTO Operations and Improve IP Management Database Reliability” — including all handouts — is available on DVD, on-demand video, and print transcript. For details, click here, or call 239-263-0605.

The entire recorded webinar “Standard Operating Procedures for TTOs: How Consistency Can Improve TTO Performance and Productivity” — including all handouts — is available on DVD, on-demand video, and print transcript. For details, click here, or call 239-263-0605.


About Technology Transfer Tactics monthly newsletter...

Find more articles like this one when you subscribe to Technology Transfer Tactics monthly newsletter. Sign up today and get immediate access to our Subscriber-Only Online Resource Center, which includes the entire archive of TTT back issues (since 2007), as well as our treasury of industry research reports, legal opinions, sample forms and contracts, government documents and more.