TapCub is live — analytics, insights and live chat, free on one platform
Glossary

What is technical support? Tiers, metrics and a worked example

Technical support helps users solve technical problems with a product, in tiers, aiming to fix it on first contact. The three tiers, four metrics and a 1,000-ticket example.

Emma Collins Reviewed by Chloe Morgan Also called tech support Last updated Nov 19, 2026 Glossary

Quick answer

Technical support helps users solve technical problems with a product, through channels such as chat, email and phone. It is organized in tiers, so simple problems are fixed by generalists and hard ones reach specialists. The goal is to get the user working again, ideally on the first contact.

Formula: First-contact resolution = tickets solved at Tier 1 without escalation ÷ all tickets

What is technical support?

Technical support is the service that helps users of a product solve technical problems with it: they cannot log in, a feature fails, an install does not work, an integration returns an error. It is delivered through channels such as live chat, email, phone, remote sessions or a help centre, and it is usually organized in tiers so that simple problems are solved quickly by generalists and hard ones reach specialists.

Wikipedia describes tech support as a service that advises and assists registered users with problems related to technical products, and Techopedia calls it a subset of customer service. The useful way to hold the idea is by its goal. The goal is not to answer tickets; it is to get the user working again, ideally on the first contact, and to learn from the pattern of problems so fewer of them happen.

This page covers the model (tiers), the numbers that tell you whether it works, and a worked example. It is not a comparison of support providers; for the broader handling of requests, see the help desk guide for ecommerce.

How is technical support different from customer service?

The two overlap and share tools, but they answer different questions.

DimensionCustomer serviceTechnical support
Typical question"Where is my order?" "Can I change my plan?""The export fails with an error." "The script is not collecting data."
NeedsPolicy and account knowledgeProduct and troubleshooting knowledge
Usual first stepLook up the accountReproduce or diagnose the problem
Typical outputAn answer or an account changeA fix, a workaround or a bug report
Escalates toA supervisor or billingA more technical tier, then engineering

In small teams the same person does both. The distinction still matters for measurement: a billing question and an install problem have very different expected resolution times, and mixing them in one average hides both.

What are the three tiers of technical support?

The common model has three tiers (sometimes called L1, L2 and L3), with self-service often placed in front as an unnumbered first layer.

TierWhoWhat they handleEscalates when
Self-serviceHelp centre, FAQ, knowledge baseKnown, repeatable questionsThe user cannot find or apply the answer
Tier 1 (L1)Support generalistsGather details, identify the real problem, solve simple issues such as login problems and setup checksThe problem needs deeper product knowledge or a fix they cannot apply
Tier 2 (L2)Technical specialistsDeeper investigation; they review what Tier 1 already tried; diagnostics, configuration, remote sessionsThe problem is new, unsolved or needs a code change
Tier 3 (L3)Experts, often developers or people who know the back endThe hardest problems; develop fixes for new issues and test them before releaseRarely; the loop may end in a bug fix or a product change

Two rules keep tiers useful. Every escalation carries the history: what the user reported, what was tried and how long they have waited, so the next tier does not start from zero. And escalation is a tool, not a failure: a Tier 1 agent who correctly escalates a hard ticket in five minutes serves the customer better than one who spends an hour guessing.

Which metrics tell you whether technical support works?

Four measures cover most of what matters; more can be added later.

MetricDefinitionWhat it tells you
First response timeTime from the user's request to the first human replyWhether people feel heard; see first response
Resolution rateResolved conversations ÷ all conversationsWhether the team closes what it opens; see resolution rate
First-contact resolution (FCR)Tickets solved at Tier 1 without escalation ÷ all ticketsHow much self-service and Tier 1 can absorb
Escalation rateTickets passed up a tier ÷ tickets that reached that tierWhether tiers are sized and trained right; see ticket escalation

Add satisfaction (a short score after resolution) and reopen rate (tickets that came back) if you can. Do not judge a team on speed alone: a fast, wrong answer looks good on first response time while resolution stays low.

Worked example: where 1,000 tickets go

Sample data: a software product receives 1,000 technical tickets in a month. Here is where they end up.

StageTickets inResolved herePassed upRate
Tier 11,000640360FCR 64.0%; escalation 36.0%
Tier 2360250110Resolved 69.4% of 360; escalation 30.6%
Tier 31108525 still openResolved 77.3% of 110
Total1,00097525 openResolution 97.5%

How to recompute: FCR = 640 ÷ 1,000 = 64.0%. Tier 2 escalation = 110 ÷ 360 = 30.6%. Total resolved = 640 + 250 + 85 = 975, and 975 ÷ 1,000 = 97.5%. The 25 open tickets are the 110 that reached Tier 3 minus the 85 it resolved.

What to do with it: Tier 2 is carrying 360 tickets, 36% of all volume. Suppose a knowledge-base article for the most common Tier 2 problem lets Tier 1 solve 100 more tickets on its own. Then Tier 1 resolves 740 (FCR 74.0%), and Tier 2 receives 260 instead of 360, which is 27.8% less. The specialists get time back for the harder tickets, and users with that problem are answered faster. Changing one article moved FCR by ten points, which is why the ticket topics that reach Tier 2 are worth reading every week.

What are the common mistakes?

  • Measuring only speed. Fast first replies with low resolution mean customers wait politely and are not helped.
  • One average for every ticket. A password reset and an integration bug differ by orders of magnitude in time. Segment by topic or tier.
  • Escalation without context. Making the user repeat everything at Tier 2 is the quickest way to lose goodwill.
  • No path from Tier 3 back to Tier 1. A bug fix should become a known answer at the front line, or the same ticket will come back next month.
  • Treating self-service as optional. A clear help article often resolves a problem in zero agent minutes; keep it current.
  • Counting unresolved tickets as closed. Auto-closing silent tickets inflates resolution rate. Report closed-without-confirmation separately.

Technical support with TapCub and AskCub

TapCub's live chat can run the front line of technical support on your site, and its analytics tell you what was happening to the user before they wrote in. It is a live chat inbox, not a full ticketing system: there are no formal tier fields, service-level timers or phone channels, so the tiers in the example are handled by transferring conversations between your agents.

  1. Self-service first. In Live chat › Knowledge base, Q&A entries are sent as written when a question matches and cost no AI call; pinned entries appear in the visitor FAQ, and opening an answer does not create a conversation. Only "not solved" hands the visitor to an agent.
  2. Tier 1 with context. In Live chat › Inbox, each conversation arrives with the visitor's source, current page and their last two days of browsing. One click opens the visitor profile in analytics. Common replies are inserted from Live chat › Canned replies by typing /.
  3. Escalate by transfer. Transfer a conversation to a colleague with deeper knowledge; internal notes stay hidden from the visitor and carry what was tried. Who can see all conversations is set in Live chat › Agents.
  4. See the problem from the user's side. For web issues, Monitoring › Errors shows error groups with the affected sessions, pages and devices, which helps a Tier 2 agent reproduce what the user saw.
  5. Measure it. Live chat › Reports gives conversation count, resolution rate (resolved ÷ all conversations), average first response (human replies only; an AI reply does not count), satisfaction, agent workload including transfers in and out, and the knowledge-base hit rate. Offline messages are counted separately and are not part of resolved. Escalation rate and FCR are not built-in metrics; calculate them from transfers and conversations as in the example.

Sources

Was this explanation clear?

We use your answers to fix and expand this page.

Emma Collins

Product Analyst

Focuses on event analysis, conversion funnels, and retention, using clear metric definitions and structured analysis to explain product performance.

FAQ

Questions people ask

Short answers to the follow-up questions that come up most.

Still have a question?

Chat with the team behind TapCub — we usually reply within a few hours on business days.

What is technical support?

Technical support is the service that helps users fix technical problems with a product, such as login failures, errors or setup problems, through chat, email, phone or a help centre. It is usually organized in tiers, with escalation to specialists.

What are the three levels of technical support?

Tier 1 generalists gather details and solve simple problems. Tier 2 specialists investigate deeper issues and review what Tier 1 tried. Tier 3 experts, often developers, solve the hardest problems and build fixes for new ones. Self-service often sits in front.

What is the difference between technical support and customer service?

Customer service handles account, order and policy questions. Technical support handles product problems and needs troubleshooting knowledge. They overlap and share tools, but their expected resolution times differ, so measure them separately.

How do you measure technical support?

Track first response time, resolution rate, first-contact resolution and escalation rate by tier, plus satisfaction and reopen rate. Segment by topic so a quick password reset does not hide a slow integration bug.

See this number on your own site

Add one script tag for web and apps, and the first reports fill in within minutes.

See your data clearly. Find your growth.

Every click, backed by data. Install one line of code and see your first numbers in a minute.

No credit card · Free plans for analytics and chat · Cookieless analytics