As medical market researchers, we are tasked with providing crucial data to pharmaceutical companies. They are counting on us to deliver accurate research results that will have a critical impact on future business and marketing decisions, trickling all the way down to how medications are prescribed and patients are treated.
Yet many medical market research companies tend to overlook one simple issue, resulting in erroneous data that can lead to costly, even multi-million-dollar mistakes.
Over half of our research projects are built using target lists provided by our clients. But as a general rule, when we receive a list, no one has yet taken a thorough look at it to confirm its quality. As a result, client lists are often riddled with errors. My team and I have found that many lists contain thousands of duplicates, physicians identified as the wrong specialty, or even office workers mis-identified as physicians, to name just a few issues.
Never assume from the get-go that the client list you’ve been provided is accurate.
In many cases, a client list that looks as if it can yield 100 completes will in fact yield only 70—or fewer. While one frequently offered solution is to obtain the remaining 30+ completes from a panel, often the panel is not representative of the audience our client wishes to target. Without putting in the extra work to effectively compare the two sets of data, we end up with results that misrepresent the facts.
The key to avoiding major errors? Never assume from the get-go that the client list you’ve been provided is accurate. As researchers, it is our responsibility to ensure that our client list will allow us to reach our target audience and fill our quotas. If we start with a faulty list, we will inevitably end up with faulty data. You cannot build a quality research project on a foundation of sand.
Fortunately, identifying problems with a client list doesn’t have to be an arduous process. We’ve adopted a quality-assurance process as a matter of routine, and we highly recommend that all medical market researchers take the same measures.
About one in three client lists will contain duplicates—and a lot of them.
Without giving away our secret recipe, here are some simple steps you can take to identify any issues—before the project goes to field and panic alarms start going off.
Step 1: De-dupe, de-dupe, de-dupe. About one in three client lists will contain duplicates—and a lot of them. There are more than enough cases where out of a list of 10,000 physicians, 3,000 will be duplicates. Just the other day, we received two lists from a client who wanted to target both lists; but on closer inspection, we found that there was 98% overlap between the two lists. Fortunately, identifying duplicates in a client list is often as simple as running a five-minute process in Excel. Ideally, we want to look for a unique identifier, such as an IMS ID or NPI number. In other cases, a simple sort by first name, last name, and specialty can be revealing.
Step 2: Check to ensure that we can fill the quotas our clients are requesting. For example, if a client needs 30 completes within a certain segment of a list—say, endocrinologists—in order to run an accurate statistical analysis, we need to be sure that we have enough endocrinologists to meet that quota. Based on my 15 years of experience, I’ve found that a good rule of thumb is that for every 100 people on a list, you’ll get about three completes. If you’re getting more than that, there may well be a problem.
One effective strategy is to take the NPI numbers of physicians and compare the specialty on the client list with that on the publicly available government CMS database. You might be surprised to find that some physicians—say, internists—have identified themselves as another specialty. If our client is targeting high-prescribers of a drug, we need to focus on specialists who prescribe the drug on a regular basis, and remove internists from our list.
Step 3: Confirm that there are no missing data fields needed for the study, such as first name, last name, or address.
A Note On Sales Detail Lists: The above steps are especially important when working with lists put together by a sales team—and this applies to market research across the board. Sales representatives want to be in front of people, not entering data into spreadsheets. Despite their best intentions, they often put together lists that need to be cleaned up a great deal. Errors may be as simple as misspellings or empty fields, but often include quite a few erroneous titles for people our client does not want to target. In many cases, a sales representative may have spoken not to the prescribing physician, but to a finance person or office manager. It is not uncommon to find that seven out of ten targeted physicians have not personally spoken to a sales representative at all.
We’ve Found Problems with Our Client List—Now What?
If you’ve found significant problems with a client list, don’t be afraid to tackle the issue. Instead, use it as a positive. This is an opportunity to help our clients, not to hinder them. We’ve found that our clients are often grateful for our integrity and transparency, and we can work together productively to find solutions that will get them the accurate information they’re seeking.
It’s not always easy to tell a client that obtaining accurate research results may be more costly or challenging than expected—but it’s certainly our obligation.
Several years ago, a new client came to us with a longitudinal research project they had been conducting over a period of many years. As part of our due diligence, we ran the names on their target list against the government CMS database and quickly discovered that almost none of the names on the list were healthcare professionals. Stunningly, the list our client had been working with for years was about 98% inaccurate. Although the client was located several hundred miles away, I immediately hopped on a plane to meet with them. I explained that the completes they had been getting for years had been erroneous and proposed a detailed strategy to help them achieve their goals moving forward. While we certainly didn’t want to be the bearer of bad news, in the end our client was so grateful for our honesty and dedication, they not only reimbursed us for the flight, but put extra money into the project. Everyone ended up looking good, further costly errors were averted, and we’ve been managing the project for them ever since.
It’s not always easy to tell a client that obtaining accurate research results may be more costly or challenging than expected—but it’s certainly our obligation. In the end, the small cost of gaining more robust results is a drop in the bucket compared to the millions of dollars lost when business decisions are made based on flawed data.
Ultimately, our job is not simply to deliver completes. Our job is to empower companies to make decisions with confidence—and to do so, we need to inform them about pitfalls today and potential challenges down the road, as well as to offer creative and effective solutions.
This article was originally posted in the GreenBook Blog.