Wealth gives a person capacity to give more money to their passions. Prospect researchers rely upon dozens of data sources to assess wealth, from airplane registrations to real estate ownership to SEC filings. Employment data is often critical for understanding someone’s career path and earnings.
Several data sources offer current career and employment data, but only one motivates more than 300 million professionals to keep their information spotless and up to date. LinkedIn’s professional network is the world’s most valuable trove of professional data.
Advancement and fundraising teams would all love to append LinkedIn data directly to their constituent records. Some have taken steps to procure LinkedIn data directly. Who can blame them?
Is It Ethical?
Wise and wily advancement professionals have grappled deeply with this question. Sarah Bernstein wrote an excellent post on the ethics of gathering social data for prospect research, and Jennifer Filla followed up with the perfect headline to get APRA hearts racing.
I’ll play the devil’s advocate. Let’s pretend that acquiring LinkedIn data without the consent of LinkedIn or your constituents is ethical. This raises a new question, which has not been considered as deeply.
Is Scraping Effective?
Scraping is certainly possible. In the past few months, CASE Prospect Research listserv participants mentioned browser extensions, programming libraries, and other tools. However, scraping consistently and accurately is very hard, particularly against the modern Javascript-driven, rapidly changing web interfaces of our era. The difficulty increases when against a well resourced, large social network who has nothing to gain, and plenty to lose, by letting you get away with it.
Let’s consider how many records you want to scrape. LinkedIn employs very clever engineers, at least some of whom work with LinkedIn’s general counsel to identify, prevent, and litigate against scrapers. How many of your constituents will you get through until your IP address (which may implicate your institution, by the way) is flagged, rate limited, or blocked?
Is It Legal?
Quick note: I am not a lawyer, nor do I play one on television, and this article does not constitute legal advice.
I believe the only legal approaches rely on public information indexed in search engines. LinkedIn’s terms of use—that thing you skipped through when creating your LinkedIn account —forbids copying while acting as a LinkedIn user.
The LinkedIn User Agreement (as of May 2015), says a user must not “Scrape or copy profiles and information of others through any means (including crawlers, browser plugins and add-ons, and any other technology or manual work).” LinkedIn used this agreement to litigate victoriously.
Need Some LinkedIn Data? I Know a Guy…
Advancement teams buy data appends all the time. If scraping is difficult (or illegal), let’s offload the task to a vendor. How is buying LinkedIn data any different from NCOA or wealth screening data?
Public Data: What’s In a Name?
If a vendor never agrees to the LinkedIn terms of use, they can gather data from public search engine results.
Microsoft and Bing provide a fabulous API that allows efficient name search. Problem solved! Let’s let the computers do the hard work and check out a couple names. Go ahead and click through to these LinkedIn profiles:
While it is probably legal to acquire public LinkedIn data in this way, how do you know if it’s right? Name matching consistently and correctly is hard too. Have you ever had to split a column of names in Excel into first and last names? Rev. Richard Wayne Gary Wayne will tell you that’s no picnic.
Good news: Vendors that offer you LinkedIn data and that have not been sued by LinkedIn are using this tactic, and are hard at work normalizing the Tim/Timothy/Timmy and Jen/Jenny/Jennifers of the world.
Bad news: It’s still not that great. Are you sure the data is accurate enough to import directly into your database? To weed out the wrong Mike Tyson or Amy Poehler, does your team have time to verify each one? How long does the verification take per record?
How much time will pass before you process the data or purchase another set? People switch jobs every 5.4 years, and younger generations are doing so more frequently, so a big chunk of the data you just paid for is out of date before long.
Even if the file is mostly correct, how many constituents were matched? 5 or 10%? EverTrue has a partnership with LinkedIn, which provides near real-time insights into LinkedIn while allowing us to use email and other data to tell Jane Smith from Jane Smith. Through this partnership, EverTrue routinely matches more than double the number of constituents who are identifiable through public data alone.
Scraped LinkedIn data also loses the wonderful context you have as a logged in user. When viewing LinkedIn constituent profiles in GivingTree, you can see shared connections—other constituents or individuals who can help you start a relationship.
Time Is Finite. LinkedIn Data Is Valuable. What Should You Do?
Rather than spend time in a vicious cycle of scraping, purchasing, checking, and entering data into your CRM, wouldn’t you rather segment by accurate title, company, and industry?
An advancement services director in 2015 who isn’t exploring all the options with a valuable professional data source is failing their institution. Rational human nature led you to consider all the above options. We think GivingTree provides better insights more efficiently, and just like fair trade coffee, is ethically sourced.