Intercom Help
intercom.com backs 41 GrowthDex tactics. This page exists so readers and crawlers can follow the evidence trail from source to tactic.
Original source
Cited GrowthDex tactics
- Audience-gated public article answers by segmentApply audience rules to public help articles so Fin only uses each answer for the customer segments that should actually see it.
- Brand-matched AI help-center content scopingScope AI support answers to the active brand's help center and duplicate critical docs across brands when needed, instead of assuming the model will search every knowledge base.
- Collection-description scan copy for support routingWrite collection descriptions like routing copy, because a short line of context often decides whether a visitor opens the right support section or starts guessing.
- Collection-only search gate before Help Center launchDo not call a help article live until it sits inside the right collection, because orphaned articles are harder to find and cannot be searched in the Help Center.
- Customer card sort before Help Center IA lockTest your support categories with real customers before freezing the Help Center structure, because the internal taxonomy is usually worse than the team thinks.
- Help-center homepage section order by jobReorder articles and collections on the help-center homepage around the first job the visitor needs done, not the internal information architecture.
- Highest-volume question first in each help collectionPut the most asked question at the top of each support collection so the page answers the likeliest job before the visitor has to scan the rest.
- Locale-picked help-center homepage articlesCurate a homepage strip of published help articles per locale instead of assuming the collection grid will route everyone to the right answer.
- Localized help-center homepage CTA cardsUse a localized homepage content card to push one feature, webinar, or support action without sending every market to the same generic banner.
- Messenger-brand-linked help center suggestionsLink each Messenger brand to the right help center so article suggestions inherit the same brand context as the support conversation.
- Private help-article login redirect before publishSet a real login-page redirect before you hide help articles behind a user-only gate, so the reader lands on the answer after authentication instead of at a dead end.
- Same-workspace 301 map after help-center migrationKeep a same-workspace 301 map for imported or renamed help-center URLs so old article paths continue carrying traffic and trust after a migration.
- Single collection home for each public help articleGive every public help article one deliberate collection home instead of scattering the same answer across multiple support buckets.
- Unlisted public article preview before search releasePublish a help article as an unlisted public page first so the team can review the live URL before the page becomes searchable in the help center or crawlable by search engines.
- Workflow-step sequencing inside help collectionsOrder help articles by the real task sequence when the reader is trying to complete a workflow, so the collection doubles as guided onboarding instead of a static archive.
- Action-led help article titles and meta descriptionsWrite help article titles around the task and keep the description crisp, because the same copy often decides both internal search relevance and the public search snippet.
- Analytics priority queue for help-center translationTranslate the help articles real visitors already read first, instead of trying to localize the whole support archive at once.
- Custom-domain gate for private help-center articlesTreat private help articles as a real distribution surface only after they live on a custom domain that supports audience targeting.
- Friction-triggered help article promotionTrigger the relevant help article from the product right before a known friction step so the answer meets the job while the user is still trying to finish it.
- Help-center search terms in title, description, and bodyMatch help article titles, descriptions, and body copy to the exact terms users search so the right answer wins inside your own help center before they bounce away.
- High-traffic help articles linking to low-traffic answersUse your busiest help articles to internally link readers into overlooked but relevant answers, with anchor text that promises the job rather than just naming the topic.
- One help article per search jobSplit support content into one article per customer question so search, linking, and recirculation work on a clear job instead of a blended catch-all page.
- Product-limitation FAQ with best alternative pathWhen the product cannot do the requested job, publish a help article that says so plainly and routes the reader to the closest workable path instead of leaving support to repeat the same dead-end answer.
- Related articles block on every public answerShow a small set of related articles on every help page so one solved question can naturally hand off to the next one without sending the reader back to search.
- Search-intent collection copy in every help-center languageWrite collection names and descriptions in the words customers actually search for in each language, because those labels often become the public search snippet.
- Shared article library across multiple brand help centersReuse one article across multiple brand help centers and let internal article links resolve in the active help center, instead of duplicating the same doc for every brand.
- Translated collection gate before multilingual help-center launchDo not call a help center multilingual until each visible collection has a translated title and at least one translated article behind it.
- Contact and attribute preload before support history importCreate the customer contacts and required attributes before importing historical threads so every migrated record lands with the right identity and fields attached.
- Cutover runbook and rollback plan before support launchWrite the first migration run and the formal cutover as a runbook with a rollback path before launch day so the switch does not depend on memory or Slack improvisation.
- Drain-and-fill cutover before final support history batchMove new support traffic to the new tool first, work both systems temporarily, and migrate the remaining closed history after the old queue drains.
- Full transcript reply instead of message-by-message replayImport the thread as one created record plus one full transcript reply when the buyer mainly needs readable history, not a perfect recreation of every old event.
- History structure choice before support importDecide whether imported history should become conversations or tickets before scripting the migration so the buyer sees one coherent system instead of a messy hybrid.
- Automatic forwarding and domain auth before support cutoverSet up automatic forwarding and authenticate the support domain before the inbox migration goes live so the new workspace can handle real mail without looking improvised.
- Batch test migration on sample records firstRun the first real migration on a small sample set in a test workspace so the switch learns on safe data before touching the live queue.
- Cross-domain help-center 301s before docs moveBefore moving a help center to a new host, ship domain-level 301 redirects from the old support domain so rankings and bookmarked answers survive.
- Delta-date cutover for support migrationChoose a fixed delta date, move closed history first, then bring over the remaining open and pending work at formal cutover.
- Help-center collection link cleanup after domain switchAfter a help-center domain switch, audit collection and index pages separately instead of assuming the article redirects covered the whole docs surface.
- Legacy docs redirect map during help-center migrationWhen migrating a help center or docs corpus, ship permanent redirects from every legacy URL so search traffic, backlinks, and in-product links survive the move.
- Simplest-form history import for reporting continuityMigrate historical support data in the lightest useful format so buyers keep reporting continuity without turning the switch into a transcript reconstruction project.
- Single indexed help center during knowledge syncWhen syncing one public help center into another system, keep only one version indexable so the migration does not create duplicate content.
- Workflow exceptions before API-led migrationBefore importing conversations by API, disable or exempt the automations that would treat imported records like fresh customer activity.