Last week I looked up 145 people in Apollo and Seamless.AI, 2 of the contact tools sales teams pay for, and checked every answer against something I could prove.
It's the same check I used to do by hand. I'd find the right person at a company, open them on LinkedIn to see if it was still true, and type them into a row. About 10 minutes a contact, on a good run. This time the tools did the looking and I did the checking.
If your reps write an opening line around a job title, you should know that about 1 in 4 of those titles was wrong or missing in each tool. And if someone changed jobs this year, one of the 2 tools almost never noticed.
01What I checked them against
I needed answers I could prove, so I built 2 lists first.
The first was 96 people listed on their own employer's website: a team page with their name and title, and for 21 of them, the email the company publishes. If the company says that's their VP of Sales today, that's the answer.
If the company says that's their VP of Sales today, that's the answer.
The second was 49 people whose move to a new company was announced in a press release between May and October. I looked each of them up at their old company, the way a rep would if that's what's in the CRM, and checked whether the tool had caught up.
Both tools got the same people, looked up the same way, on the same day. I'm an Apollo Certified Partner, so you should also know that neither company knew about this test, paid for it or saw it before it went up. The whole method is on the results page.
02About 1 in 4 titles were wrong or out of date
Of the people each tool found, Apollo had the right title, or close to it, for 75%. Seamless had 73%. Close counts as right: "Founder" for "CEO and Founder" is fine.
In the rest, an executive VP showed up as a VP. A regional VP showed up as a director. A head of people showed up as a production assistant. In one tool, 2 people's current job was a Forbes council membership.
A head of people showed up as a production assistant.
Reps personalize on title, so a wrong one is often the first thing a prospect reads from you.
03The emails were mostly fine
21 of the people had an email their company publishes on its own site.
Apollo matched 17 of them exactly and Seamless matched 18. Most of the misses were a different format, like a first initial and last name where the company uses first names. One had the wrong first name entirely.
If you're choosing a tool on email accuracy alone, the 2 tools are close enough that it shouldn't decide anything.
04The 49 people who changed jobs
Of the 49 people who'd announced a new job, Apollo showed 33 at the new company. Seamless showed 1. Looked up at their old company, Seamless still had 34 of them sitting there, weeks or months after the announcement went out.
I expected the misses to be the recent moves, the ones from October that nobody could have caught yet. They weren't. Of the moves more than 4 months old, Seamless still showed all 13 it found at the old company, and Apollo showed 6 of 18.
Looked up at their old company, Seamless still had 34 of them sitting there, weeks or months after the announcement went out.
A rep sequences someone who left in June. Your CRM keeps them as the contact at that account. And the person who might buy from you at their new company never hears from you, because your tool still has them at the old one.
05Test yours this week
You don't need my lists to do this. Pick 20 people you can check: 10 from customers' team pages and 10 who announced a new job in the last 6 months. Look each one up in your tool the way your reps do, and count 3 things: titles that are wrong or missing, emails that don't match the one the company publishes, and job changers still shown at the old company.
It takes about 20 minutes. If you're choosing between 2 tools, run the same 20 names through both trials before you sign anything.
06What I'd do with the answer
If 5 or more of your 20 titles miss, have reps confirm the title before they write a line around it. If your job changers are still sitting at the old company, re-check the person before any record goes into a sequence, however old the record is. Age didn't make the data any fresher in my test.
Either way, have someone open the person on LinkedIn before a row goes into a sequence.
