--- title: Match Management slug: match-management docTags: createdAt: 2025-06-10T15:21:54.377Z --- # Match Management Match Management is where you review and act on every record the integration has tried to match. Open it any time you want to check what got matched, what still needs a decision, and what could not be matched at all. You will find it in the Altrata Settings app, on the **Match Management** tab. ## What the three tabs do Match Management groups every eligible record into three buckets. The number next to each tab tells you how many records are currently sitting there. - **Matched** shows records that have been linked to an Altrata profile, either automatically or by hand. - **Possible Match** shows records where the system found one or more candidate profiles but was not confident enough to pick one for you. These need your review. - **No Match** shows records the system could not match to anything in the Altrata dataset. Every tab also lets you switch between Lead, Contact, and Account (and Person Account, if your instance uses it) using the sub-tabs near the top. ## Auto-match threshold You will see an **Auto-match threshold** indicator in the top right of every Match Management tab, set to 95% by default. This is the confidence score a match has to hit before the system links the record automatically. Anything at or above the threshold ends up in **Matched**. Anything below it ends up in **Possible Match** for your review. If you want auto-matching to be more aggressive (more records flowing straight into Matched without manual review), lower the threshold. If you want to be stricter and review more candidates by hand, raise it. Change the threshold from the **Field Mapping** tab in Altrata Settings, right next to Match Management. :::BlockQuote đŸŸĸ **Best practice:** the default 90% is a good starting point for most customers. Adjust it after your first round of bulk matching once you have a feel for how the system is performing against your data. ::: ## Matched tab The Matched tab shows every record that has been successfully linked to an Altrata profile. Each row shows: - **Salesforce Entity** with the record name and role - **Altrata Entity** with the matched profile name and role - **Match Score** as a percentage - **Match Type** showing how the match was made (Auto-match or Manual match) Use this tab when you want to spot-check matches after a bulk match run, audit the quality of recent auto-matches, or confirm that a specific record was linked correctly. The list is ordered by recently added by default, so the newest matches sit at the top. ## Possible Match tab The Possible Match tab is where most of your day-to-day match management work happens. These are records where the system found candidates but the confidence score did not clear the auto-match threshold. Each row shows your Salesforce record on the left, followed by up to **three possible matches for individuals and five possible matches for organizations**, each with a confidence percentage and the basic profile info (name, role, organization, public/private status, location). On the far right is a **Reject** option for the whole row. ### How to review possible matches For each row, do one of three things: 1. **Pick the right match.** Click the radio button next to the candidate you want. The system selects the highest match score record by default. 2. **Reject the row entirely.** If none of the candidates are correct, click **Reject** on the right. The record will move to No Match. 3. **Open more info.** Click **more info** on any candidate to expand additional detail (DOB, age, gender) so you can compare candidates more carefully. ### Confirming your selections in bulk Once you have made your selections, click the **confirm \[n] selections** button in the top right (where \[n] is the number of selections you have made). A confirmation modal will pop up asking you to confirm the action across all the records you selected. This batch confirm is the fastest way to clear a backlog of possible matches. You do not need to confirm row by row. :::BlockQuote ⚠ī¸ Confirmation is final. Once you confirm a batch, those records move into Matched and the linked Altrata data starts populating. If you change your mind later, you can manually unlink the record from the Salesforce record page. ::: ## No Match tab The No Match tab shows records the system could not link to anything in the Altrata dataset. Each row shows the Salesforce record on the left and a **No Match Reason** on the right. Common reasons you will see: - **No match was found for this record.** The system looked but did not find a candidate that met the matching methods you configured (see Field Mapping). - **Manually Rejected.** You or another user rejected all candidates for this record from the Possible Match tab. ### What to do with No Match records A record in No Match is not necessarily a problem. Sometimes the person or organization genuinely is not in the Altrata dataset, and that is fine. But if you expected a record to match and it did not, here is where to start: - **Check Field Mapping.** If the record is missing data on all four matching methods (First/Last/Organization, Email, LinkedIn URL, foundation brand ID), it cannot match. Add or correct the data in Salesforce and wait for the next match run. - **Check Data Filters.** If you have Data Filters configured, the record may be filtered out on purpose. Confirm the filter is what you want. - **Send a profile request.** If the record is commercially important and you believe the person or organization should exist in the dataset, you can request that Altrata research the profile. ### Sending a profile request On the No Match tab, you will see a **Send Request** button on rows where requesting a profile is supported. Click it to redirect to the Altrata web platform, where you can submit a profile request to the research team. :::BlockQuote ℹī¸ **Profile requests use credits.** Your organization needs available profile request credits in order to submit a request. If you are out of credits or unsure how many you have, contact your Altrata Account Manager. ::: The research team reviews requests and responds through the standard fulfillment process. Once a profile is created and the record is matched in a future match run, it will move out of No Match automatically. ## When to use Match Management There are a few moments when Match Management is most useful: - **Right after the initial bulk match.** Spot-check Matched, review Possible Match, and triage No Match. - **After each weekly auto-match cycle.** New records get added, no matched records get re-evaluated. - **When you have configuration concerns.** If a lot of records are landing in No Match or Possible Match, that is usually a sign your Field Mapping or Data Filters need a second look, not that the records are bad. - **Before sending profile requests.** Make sure the record really cannot be matched (and that you have credits available) before spending a request on it. ## Match Management best practices - **Work through Possible Match on a regular cadence.** Weekly works well, lined up with the automatic refresh. Leaving them unresolved means the Salesforce record stays unmatched and users will not see Altrata data on it. - **Look for clusters in No Match.** If a whole segment of records (one campaign, one region, one product line) is landing in No Match, the issue is almost always upstream in Field Mapping or Data Filters, not in the records themselves. - **Be deliberate with profile requests.** Credits are limited. Confirm the record is genuinely worth requesting before clicking Send Request. # Manual Matching Records may also be matched on a case-by-case basis. Should this be enabled, users can manually match **Leads**, **Contacts**, **Person Accounts**, or **Accounts**. ## Manually Link a Record ::::WorkflowBlock :::WorkflowBlockItem Navigate to the record. ::: :::WorkflowBlockItem Click the **Link to Altrata **button. If the record has already been matched, this button will display **Unlink from Altrata**. ![](https://api.archbee.com/api/optimize/OchO21fL_QwZk6Sym4QQl/hDnuOH-pW9hJjiivUfpwa_image.png) ::: :::WorkflowBlockItem A list of potential matches, if available, will be presented for your review. You may optionally filter by first name, last name, organization, or role title in the search bar. Select the desired match and click **Confirm**. ![](https://api.archbee.com/api/optimize/OchO21fL_QwZk6Sym4QQl/BThyZOzwEL5_pO-6IZU9r_image.png) ::: :::WorkflowBlockItem The Salesforce record will now be populated with Altrata data. ::: :::: **Manually Unlink a Record** ::::WorkflowBlock :::WorkflowBlockItem Navigate to the record. ::: :::WorkflowBlockItem Click the **Unlink from Altrata **button. ![](https://api.archbee.com/api/optimize/OchO21fL_QwZk6Sym4QQl/GezpfpQRWYGIn6Pj4ddY0_image.png) ::: :::WorkflowBlockItem Confirm the unlinking by clicking **Unlink** on the pop-up window. ![](https://api.archbee.com/api/optimize/OchO21fL_QwZk6Sym4QQl/RAMF8IIqp77047UxcVOJZ_image.png) ::: :::WorkflowBlockItem The Salesforce record will now have the Altrata data removed. ::: :::: # Altrata Matching FAQ Common questions about how the Altrata matching engine evaluates records, assigns scores and decides whether a record is an auto-match, a possible match or no match. ## Rule Evaluation Logic **Are the matching rules evaluated independently or combined?** The matching logic runs all rules at the same time. The best match found from any combination of rules is retained as the match. **Does satisfying a single rule result in an auto-match, or are multiple rules evaluated collectively?** A single rule can result in a match. Whether it becomes an auto-match or a possible match depends on the match score. **How is the overall match score calculated when multiple rules match?** There are different layers of matching. The first layer is an exact match, which scores 100. The next layer is fuzzy matching, where different weights are assigned to different data points. Those weights combine to produce the overall score. ## Rule Priority and Execution Order **Is there a priority order in which the rules are evaluated?** There is no formal priority order, but the engine always runs exact matching on the data first. Fuzzy matching only runs on records where no exact match is found. When multiple data points have exact matches: - If those exact matches all point to the same Altrata profile, the record is auto-matched - If those exact matches point to different Altrata profiles, the record goes to Possible Matches so a human can decide which one is correct ## Exact Match vs. Fuzzy Match **Does the matching engine use exact matching or fuzzy matching?** Both. Exact matching runs first. If no exact match is found, fuzzy matching runs on the record. **Are spelling variations, abbreviations, or formatting differences considered?** Yes. The engine handles examples like: - "Bombardier Aviation" vs. "Bombardier" - "John Smith" vs. "Jon Smith" - "Marc Fisher Footwear" vs. "Marc Fisher" The score assigned depends on how close the fuzzy match is. Smaller differences produce higher scores, larger differences produce lower scores. ## Nickname Handling **Does the matching engine recognize nicknames?** Yes. The engine maintains a set of name synonyms it uses during matching. Common examples include: - Bob ↔ Robert - Bill ↔ William - Jim ↔ James - Mike ↔ Michael Work is underway to also ingest alternate names into the matching logic in future releases. **How do nickname matches affect the match score?** When synonyms apply, the engine evaluates the record against every reasonable name combination. For example, if an Altrata profile exists for "Robert Smith," the engine will attempt to match it against: - Robert Smith - Bob Smith - Rob Smith - Bert Smith Each combination is scored alongside the other data points on the record (organization, email, address, etc.), using the model's defined weights and fuzziness tolerances. ## Organization Name Matching **What weight is assigned to Organization Name within the matching algorithm?** Organization Name is weighted differently depending on which object is being matched: - **Organization matching (Account object):** Organization Name carries high weight, because it is the primary data point available for matching organizations - **Person matching (Lead, Contact, Person Account):** Organization Name carries lower weight, because it is one of several data points **Can a record be auto-matched when the person's name matches but the organization differs significantly?** Yes, as long as another strong data point is present. Examples: - A legacy foundation brand entity ID (BoardEx, RelSci, Wealth-X) that links the record to a known Altrata profile - An email address that matches an Altrata profile If the only data points on the record are name and organization, and the organization name differs significantly, the record cannot be auto-matched. **Under what circumstances would such a record be classified as an Exact Match rather than a Possible Match?** It will not be classified as an Exact Match unless there is another data point that produces a high-confidence match. Name alone is not enough. ## Address Matching **Does address matching require all address components to match?** No. ZIP code alone or a partial address is enough to produce a match. The more fields that overlap, the higher the score. **How are missing address attributes handled?** The engine uses only the address fields that are submitted. There is no imputation or fallback logic for missing fields. **Is partial address matching supported?** Yes. ## Auto-Match Threshold at 100% **What exactly constitutes a 100% match?** A 100% match is an exact match on one of the following combinations, with no fuzzy matching applied: - First Name + Last Name + Organization Name - First Name + Last Name + ZIP Code (for US addresses, only the first 5 digits are used) - Email - LinkedIn URL **Does the match need to satisfy all of the configured attributes?** No. Any one of the combinations above is enough. **Can records that rely on fuzzy matching still score 100%?** No. Only exact matches can achieve a score of 100. ## Possible Match Logic **When Auto-Match Threshold is set to 100%, are records scoring exactly 100% automatically linked?** Yes. **Does any match below 100% appear under Possible Match?** Not quite. Only matches with a score of **61% or higher** appear in Possible Match. Anything below 61% is classified as No Match. **Are there additional rules or thresholds that affect this behavior?** No. ## Match Score Transparency **Is there a way to see a detailed score breakdown for a matched record?** Not today. The matching model returns an overall score for each match, but it does not expose a per-field breakdown (for example, Name 40%, Organization 30%, Address 20%, Email 10%). A detailed breakdown can be calculated manually by applying the model's defined weights, but that information is not captured at the field level in the app.