Articles & Case Studies
Semantic Contact Search vs Title Matching
In a DiscoLike search across 300 medical pain clinics, semantic matching returned 745 candidates across 274 titles. A comparison using 10 common titles matched 89 of those candidates.
· updated October 6, 2026
A client wanted “executives and senior decision makers” at medical pain clinics. That sounds like a straightforward title search until you inspect how people in those businesses describe their roles.
We compared a short title list with a semantic description across the same 300 companies. The difference illustrated a retrieval gap worth investigating.
The comparison: ten titles and one persona
We asked ChatGPT to suggest ten titles for the audience. It returned CEO, Chief Executive Officer, Owner, Founder, President, Medical Director, Chief Medical Officer, Chief Operating Officer, Managing Partner, and Executive Director.
We then checked how those titles would match the candidates returned by DiscoLike’s semantic search for the client’s audience description.
This comparison uses that particular ten-title list. Expanding the list or changing the matching rules would change the baseline.
What the search returned
| Measure | Result |
|---|---|
| Companies searched | 300 medical pain clinics |
| Semantic-search candidates | 745 |
| Distinct candidate titles | 274 |
| Candidates matched by the ten-title comparison | 89 |
| Candidates outside that comparison | 656 |
The semantic result set was approximately 8.4x the size of the ten-title match set. About 88% of the returned candidates were outside the short-list comparison.
These are candidate counts before independent role qualification. To measure campaign-ready coverage, check the additional contacts’ current roles and buying responsibilities.
Why the additional titles matter
The wider set included titles such as Doctor, MD, Physician, Managing Director, Chief Financial Officer, and Director of Business Operations.
Some may be valuable contacts, depending on the offer. Others may not own the buying decision. A title like Physician can indicate a relevant practitioner without establishing purchasing authority.
Semantic search broadens the candidates you can inspect. Qualification determines which belong in the campaign.
Run a fair comparison for your own persona
Use the same company set for both methods. Save the title rules and the semantic description, then deduplicate people before counting.
Review candidates found only by semantic search. Check current company affiliation, responsibilities, and relevance to the product you sell. Report both incremental candidates and incremental accepted contacts.
Also inspect people missed by semantic search if you have a known reference list. A larger result set does not establish perfect recall.
Choose the method for the requirement
Exact title matching is useful when the title itself is mandatory. Semantic matching is useful when related responsibilities appear under different names. Company and geography constraints can support either method.
Try DiscoLike contact search with a familiar buyer persona and evaluate the additional contacts before expanding your outreach.
Related posts:
