How does your matching and enrichment work? How do you prioritize the different inputs?
Leadspace algorithms use and rely on multiple data sources, and the sequences and weight of every input might change according to the required results.
For company enrichment, it might be one flow, for a lead it's another, and for site level matching it's a third, etc.
However, if we were to outline a general, simple lead enrichment flow, it would look something like this:
- First and Last Name
A full name is crucial for identifying the correct person and being able to separate between one John Doe and another with the same name in the same company.
(We can identify shortened names like Dave = David/ Bill = William, but we will always give more priority to the input, so we will bring back Dave if that is what we have received.)
Next, the company is also crucial: we rely on several layers and sources to identify and cross-check the results. However, with an incorrect or partial name, the results might be taken into a different direction. Getting partial name may cause our algorithms to find an actual company with that exact name, but different from expected.
For example - Using company input of "HMSA" - For some, it will mean "Hawaii Medical Service Association," for others "Hawthorne Math and Science Academy".
It's important to understand what's the right company as it will have an enormous impact on the results. The input company will affect the results for the title, contact details, address, etc.
The location is another great way for us to try and focus on the right company. In the example above, if the input location is Hawthorne, California, we will probably bring back "Hawthorne Math and Science Academy." This is a more complex process, of course, as we also cross reference if that person works for the company. It all comes back to the importance of having a correct input. It's not uncommon for a person to live in one state and work in another. Or even different countries. The location helps us focus on a specific region.
Email is the last fallback: we can use this data point to verify the company, and whether the email is valid or not, etc. In some circumstances, if a person’s first and last names are unavailable, email can be used for locating the person. This process, however, will yield fewer results.
Today, our accuracy for identifying the correct lead/company is ~90% and the highest in the industry. If at any point you identify a corner case for inexact/inaccurate matches, please bring these to our attention and email us at [email protected]. We rely on our clients in partnering with us in identifying these edge cases as a way to continue to improve towards the next iteration of our algorithms.