{"id":1275,"date":"2026-11-22T02:19:15","date_gmt":"2026-11-22T07:19:15","guid":{"rendered":"https:\/\/vight.ai\/blog\/mean-time-to-resolution\/"},"modified":"2026-11-22T02:19:17","modified_gmt":"2026-11-22T07:19:17","slug":"mean-time-to-resolution","status":"publish","type":"post","link":"https:\/\/vight.ai\/blog\/mean-time-to-resolution\/","title":{"rendered":"Mean Time to Resolution: How to Measure and Improve It"},"content":{"rendered":"<p>Mean time to resolution is the average elapsed time between a ticket being opened and being closed, calculated by dividing the total resolution time of all resolved tickets by the number of resolved tickets. Incident MTTR averages 8.40 business hours in global benchmarking, with results ranging from 0.67 to 33.67 business hours.<\/p>\n<p>The most popular advice about this metric is to drive it down. That advice is incomplete. A falling mean time to resolution can come from faster technicians, or it can come from a quiet change in when the clock starts, when it stops, and which tickets get excluded from the export. Before an MSP celebrates a lower number, it should audit the definition that produced it. The math is easy. The measurement policy is where the surprises live.<\/p>\n<h2>Table of Contents<\/h2>\n<ul>\n<li><a href=\"#the-mean-time-to-resolution-formula-with-a-worked-example\">The Mean Time to Resolution Formula With a Worked Example<\/a><ul>\n<li><a href=\"#work-the-formula-on-five-tickets\">Work the formula on five tickets<\/a><\/li>\n<li><a href=\"#why-the-benchmark-is-not-a-target\">Why the benchmark is not a target<\/a><\/li>\n<\/ul>\n<\/li>\n<li><a href=\"#what-counts-as-resolution-time-and-what-does-not\">What Counts as Resolution Time and What Does Not<\/a><ul>\n<li><a href=\"#the-clock-has-two-ends\">The clock has two ends<\/a><\/li>\n<li><a href=\"#paused-time-is-a-policy-not-an-accident\">Paused time is a policy, not an accident<\/a><\/li>\n<li><a href=\"#a-name-is-not-a-definition\">A name is not a definition<\/a><\/li>\n<\/ul>\n<\/li>\n<li><a href=\"#benchmark-ranges-to-compare-your-numbers-against\">Benchmark Ranges to Compare Your Numbers Against<\/a><ul>\n<li><a href=\"#scope-changes-the-headline\">Scope changes the headline<\/a><\/li>\n<\/ul>\n<\/li>\n<li><a href=\"#why-averages-mislead-and-what-to-report-instead\">Why Averages Mislead and What to Report Instead<\/a><ul>\n<li><a href=\"#segment-before-you-celebrate\">Segment before you celebrate<\/a><\/li>\n<li><a href=\"#the-long-tail-is-the-story\">The long tail is the story<\/a><\/li>\n<li><a href=\"#the-monthly-reporting-template\">The monthly reporting template<\/a><\/li>\n<\/ul>\n<\/li>\n<li><a href=\"#where-resolution-time-leaks-in-msp-service-desks\">Where Resolution Time Leaks in MSP Service Desks<\/a><ul>\n<li><a href=\"#intake-gaps\">Intake gaps<\/a><\/li>\n<li><a href=\"#after-hours-demand\">After-hours demand<\/a><\/li>\n<li><a href=\"#handoffs-and-reopens\">Handoffs and reopens<\/a><\/li>\n<li><a href=\"#vendor-dependency\">Vendor dependency<\/a><\/li>\n<\/ul>\n<\/li>\n<li><a href=\"#how-msps-cut-resolution-time-without-losing-quality\">How MSPs Cut Resolution Time Without Losing Quality<\/a><ul>\n<li><a href=\"#fix-the-first-hour-not-the-last\">Fix the first hour, not the last<\/a><\/li>\n<li><a href=\"#the-ai-shortcut-has-a-quality-bill\">The AI shortcut has a quality bill<\/a><\/li>\n<li><a href=\"#protect-the-first-time-fix\">Protect the first-time fix<\/a><\/li>\n<li><a href=\"#test-one-lever-for-four-weeks\">Test one lever for four weeks<\/a><\/li>\n<\/ul>\n<\/li>\n<li><a href=\"#set-up-your-resolution-time-dashboard-this-month\">Set Up Your Resolution Time Dashboard This Month<\/a><ul>\n<li><a href=\"#the-five-steps\">The five steps<\/a><\/li>\n<li><a href=\"#how-to-read-month-one\">How to read month one<\/a><\/li>\n<\/ul>\n<\/li>\n<\/ul>\n<p><a id=\"the-mean-time-to-resolution-formula-with-a-worked-example\"><\/a><\/p>\n<h2>The Mean Time to Resolution Formula With a Worked Example<\/h2>\n<p>Mean time to resolution is the average elapsed time between a ticket being opened and being closed. The standard calculation is the sum of the resolution durations for all resolved tickets divided by the number of resolved tickets, as <a href=\"https:\/\/www.thinkhdi.com\/library\/supportworld\/2018\/metric-of-month-incident-mean-time-to-resolve\">HDI&#039;s metric-of-the-month analysis<\/a> sets out.<\/p>\n<p>In service-desk measurement, that elapsed time usually includes queueing, reassignment, waiting for customer information, technician handoffs, diagnosis, remediation, and closure. It is not just the minutes a technician spends with hands on the issue. That is why the metric captures workflow friction as well as technical effort.<\/p>\n<p><figure class=\"wp-block-image size-large\"><img decoding=\"async\" src=\"https:\/\/vight.ai\/blog\/wp-content\/uploads\/2026\/11\/mean-time-to-resolution-formula.jpg\" alt=\"A visual guide explaining the mean time to resolution formula with a clear worked example.\" \/><\/figure><\/p>\n<p><a id=\"work-the-formula-on-five-tickets\"><\/a><\/p>\n<h3>Work the formula on five tickets<\/h3>\n<p>Take five resolved tickets from a small MSP and calculate the number in order.<\/p>\n<ol>\n<li>List each ticket with its elapsed resolution time. T-1041 was a password reset at 2.5 hours. T-1042 was a printer driver fault at 6.0 hours. T-1043 was an email sync failure at 1.0 hour. T-1044 was a server disk alert at 14.5 hours. T-1045 was a VPN access request at 4.0 hours.<\/li>\n<li>Add the durations. The total is <strong>28.0 hours<\/strong>.<\/li>\n<li>Divide by the number of resolved tickets, which is five. The mean is <strong>5.6 hours<\/strong>.<\/li>\n<li>Sort the same durations to find the median: 1.0, 2.5, 4.0, 6.0, 14.5. The middle value is <strong>4.0 hours<\/strong>.<\/li>\n<li>Compare the two. The mean of 5.6 hours sits above the median of 4.0 hours because one 14.5-hour server disk alert pulls the average away from the typical ticket.<\/li>\n<\/ol>\n\n<figure class=\"wp-block-table\"><table><tr>\n<th>Ticket<\/th>\n<th>Issue<\/th>\n<th>Resolution time<\/th>\n<\/tr>\n<tr>\n<td>T-1041<\/td>\n<td>Password reset<\/td>\n<td>2.5 hours<\/td>\n<\/tr>\n<tr>\n<td>T-1042<\/td>\n<td>Printer driver fault<\/td>\n<td>6.0 hours<\/td>\n<\/tr>\n<tr>\n<td>T-1043<\/td>\n<td>Email sync failure<\/td>\n<td>1.0 hour<\/td>\n<\/tr>\n<tr>\n<td>T-1044<\/td>\n<td>Server disk alert<\/td>\n<td>14.5 hours<\/td>\n<\/tr>\n<tr>\n<td>T-1045<\/td>\n<td>VPN access request<\/td>\n<td>4.0 hours<\/td>\n<\/tr>\n<tr>\n<td>Total and mean<\/td>\n<td>Five resolved tickets<\/td>\n<td>28.0 hours total and 5.6 hours mean<\/td>\n<\/tr>\n<tr>\n<td>Median<\/td>\n<td>Five resolved tickets<\/td>\n<td>4.0 hours<\/td>\n<\/tr>\n<\/table><\/figure>\n<p><a id=\"why-the-benchmark-is-not-a-target\"><\/a><\/p>\n<h3>Why the benchmark is not a target<\/h3>\n<p>HDI reported historical MetricNet global benchmarking data with an average incident MTTR of <strong>8.40 business hours<\/strong>, ranging from <strong>0.67 to 33.67 business hours<\/strong>. The same analysis named ticket backlog, user-population density, travel time, and incident complexity as major factors behind that spread. It also cited common service targets in which 80% of incidents are resolved the same day or next day, and 80% are resolved within 8 business hours or 24 clock hours.<\/p>\n<p>Those targets describe a distribution, not a single average. A single mean should not become a universal target. Segment the result by priority, complexity, customer, queue, and whether the ticket required escalation. A falling MTTR is meaningful only if reopen rates, resolution quality, and ticket mix remain stable.<\/p>\n<p>The figure also only means something when the same start and stop events are used every time. Tickets still open at export are excluded until they close. If you include a partial duration for an open ticket, you have changed the metric, not measured it earlier.<\/p>\n<p>If your service desk also reports customer-facing response times, Sift AI&#039;s guide to <a href=\"https:\/\/www.getsift.ai\/blog\/customer-support-metrics\">how to measure social care performance<\/a> is a useful cross-check on start and stop definitions, because the same discipline applies to any elapsed-time measure.<\/p>\n<p>Calculate the same figure for the last 90 days of closed tickets from a PSA export. Compare the mean with the median before drawing any conclusion.<\/p>\n<p><a id=\"what-counts-as-resolution-time-and-what-does-not\"><\/a><\/p>\n<h2>What Counts as Resolution Time and What Does Not<\/h2>\n<p>The arithmetic is the easy part. The clock definition decides the result. Teams may start the clock at ticket creation, incident detection, or acknowledgment. They may stop it at workaround, service restoration, ticket closure, or verified permanent resolution. A review of customer-support statistics published by <a href=\"https:\/\/stealthagents.com\/research\/customer-support-statistics-2026\">Stealth Agents<\/a> makes the point that these choices materially change the reported figure, so a lower MTTR can reflect a measurement-policy change rather than faster service.<\/p>\n<p>Pick one start event for the whole report. Then write down the stop event. If the service desk closes a ticket when a workaround is applied, the number measures workaround speed. If it closes only after the client confirms the fix, the number measures durable resolution. Both are legitimate. Mixing them across months is not.<\/p>\n<p><a id=\"the-clock-has-two-ends\"><\/a><\/p>\n<h3>The clock has two ends<\/h3>\n<p>Queueing before assignment belongs inside the elapsed time because it shows backlog and intake delay. Reassignment and handoffs belong inside it because they expose fragmented ownership. Waiting for customer information belongs inside it too, but report that time separately so it does not hide inside technician work time. Vendor dependency is outside the MSP&#039;s direct control, yet the client still experiences it, so keep it visible in a separate bucket.<\/p>\n\n<figure class=\"wp-block-table\"><table><tr>\n<th>Stage<\/th>\n<th>Counts toward MTTR?<\/th>\n<th>Why it matters<\/th>\n<\/tr>\n<tr>\n<td>Ticket created or incident detected<\/td>\n<td>Yes, as the start event<\/td>\n<td>One start event must be fixed for every report<\/td>\n<\/tr>\n<tr>\n<td>Queueing before assignment<\/td>\n<td>Yes<\/td>\n<td>Shows backlog and intake delay<\/td>\n<\/tr>\n<tr>\n<td>Waiting for customer information<\/td>\n<td>Usually yes, reported separately<\/td>\n<td>Hides inside technician time if not split out<\/td>\n<\/tr>\n<tr>\n<td>Reassignment and handoffs<\/td>\n<td>Yes<\/td>\n<td>Main signal of fragmented ownership<\/td>\n<\/tr>\n<tr>\n<td>Diagnosis and repair<\/td>\n<td>Yes<\/td>\n<td>The part most people picture<\/td>\n<\/tr>\n<tr>\n<td>Vendor dependency<\/td>\n<td>Yes, reported separately<\/td>\n<td>Outside the MSP&#039;s control but visible to the client<\/td>\n<\/tr>\n<tr>\n<td>Paused status such as client on hold<\/td>\n<td>Depends on policy<\/td>\n<td>Exclude or segment explicitly, never silently<\/td>\n<\/tr>\n<tr>\n<td>Workaround restores service<\/td>\n<td>Depends on definition<\/td>\n<td>Restoration is not always full resolution<\/td>\n<\/tr>\n<tr>\n<td>Closure followed by reopen<\/td>\n<td>Yes, and flagged by reopen rate<\/td>\n<td>A fast close can flatter MTTR<\/td>\n<\/tr>\n<\/table><\/figure>\n<p><a id=\"paused-time-is-a-policy-not-an-accident\"><\/a><\/p>\n<h3>Paused time is a policy, not an accident<\/h3>\n<p>A &quot;waiting on client&quot; status can stop the clock or keep it running. Someone has to decide which, and that decision has to be written into the PSA status model. If the status pauses the clock, the dashboard will show a better number than the client experienced. If it keeps the clock running, the number will show the cost of missing information.<\/p>\n<p>A fast closure followed by a reopen is the most expensive version of this problem. It flatters the original resolution time while raising customer effort and total labor. Track reopenings within a fixed period and treat them as part of the resolution record.<\/p>\n<p><a id=\"a-name-is-not-a-definition\"><\/a><\/p>\n<h3>A name is not a definition<\/h3>\n<p>Searchers often meet mean time to repair, mean time to restore, and mean time to recovery used for the same clock. Those labels can hide different stop events. Mean time to repair may stop when the device is fixed. Mean time to restore may stop when service is back. Mean time to recovery may stop when normal operations return. Map each name to the stop event your PSA uses before comparing reports or vendor claims.<\/p>\n<blockquote>\n<p><strong>Practical rule:<\/strong> Write the start event, the stop event, and the paused-time rules into the PSA status model before comparing any two months.<\/p>\n<\/blockquote>\n<p><a id=\"benchmark-ranges-to-compare-your-numbers-against\"><\/a><\/p>\n<h2>Benchmark Ranges to Compare Your Numbers Against<\/h2>\n<p>A benchmark number is a sentence with the subject and the scope cut off. Freshworks reported a median customer-support ticket resolution time of <strong>6 hours 15 minutes<\/strong> in its 2025 benchmark, based on approximately <strong>1.2 billion tickets from more than 32,000 companies<\/strong>, as summarized in <a href=\"https:\/\/www.pindrop.page\/blog\/time-to-resolution\">Pindrop&#039;s time-to-resolution overview<\/a>. That is a customer-support population, not an IT service desk.<\/p>\n<p>New Relic&#039;s 2024 observability research reported a median mean time to resolution of <strong>51 minutes for high-business-impact outages<\/strong>; <strong>39% of respondents<\/strong> reported an MTTR of at least 1 hour for such outages, according to <a href=\"https:\/\/newrelic.com\/resources\/report\/observability-forecast\/2024\/state-of-observability\/outages-downtime-cost\">New Relic&#039;s observability forecast<\/a>. That is outage restoration, not routine ticket closure.<\/p>\n<p>Those two numbers are not rivals. They measure different events on different populations. An outage clock may stop when service is restored. A service-desk clock may stop when the ticket is closed and the client confirms the fix. Comparing them tells you nothing useful about either team.<\/p>\n\n<figure class=\"wp-block-table\"><table><tr>\n<th>Benchmark<\/th>\n<th>Reported value<\/th>\n<th>Scope and caveat<\/th>\n<\/tr>\n<tr>\n<td>Freshworks 2025 median, customer-support tickets<\/td>\n<td>6 hours 15 minutes<\/td>\n<td>About 1.2 billion tickets from more than 32,000 companies; not limited to IT service desks<\/td>\n<\/tr>\n<tr>\n<td>New Relic 2024, high-business-impact outages<\/td>\n<td>51 minutes median<\/td>\n<td>Outage restoration, not routine service-desk tickets<\/td>\n<\/tr>\n<\/table><\/figure>\n<p><a id=\"scope-changes-the-headline\"><\/a><\/p>\n<h3>Scope changes the headline<\/h3>\n<p>A broad IT ticket population will include password resets, access requests, application faults, and infrastructure incidents. A desktop-support population will exclude some of those and include others. A customer-support population may include product questions that never reach a technician. The reporting period matters too. A quarter with a major migration will look worse than a quiet quarter with the same staff and the same process.<\/p>\n<p>Use a benchmark only when the scope, ticket population, and business-hour calendar match your own. If no published benchmark matches, use your SLA targets and your own trend line. A median and a mean from the same export are more useful than a headline figure from a different population.<\/p>\n<blockquote>\n<p><strong>Comparison rule:<\/strong> Put the scope beside every benchmark, then ask whether your tickets would have been counted in that population at all.<\/p>\n<\/blockquote>\n<p><a id=\"why-averages-mislead-and-what-to-report-instead\"><\/a><\/p>\n<h2>Why Averages Mislead and What to Report Instead<\/h2>\n<p>A single blended mean hides the answer. A password reset and a datacenter outage have different urgency, diagnostic effort, escalation paths, and acceptable resolution windows. Combining them can conceal poor performance on critical incidents while routine requests pull the overall average down, as <a href=\"https:\/\/www.easyvista.com\/blog\/itsm-metrics-and-kpis-a-practical-guide-to-measuring-what-matters\/\">EasyVista&#039;s ITSM metrics guide<\/a> explains.<\/p>\n<p>That is the blended-average trap. The service desk can look acceptable overall while priority-one incidents miss their SLA every week. The owner sees one number, the client sees the outage.<\/p>\n<p><figure class=\"wp-block-image size-large\"><img decoding=\"async\" src=\"https:\/\/vight.ai\/blog\/wp-content\/uploads\/2026\/11\/mean-time-to-resolution-blended-mean.jpg\" alt=\"A conceptual diagram illustrating how blended mean averages can obscure significant IT service delivery problems.\" \/><\/figure><\/p>\n<p><a id=\"segment-before-you-celebrate\"><\/a><\/p>\n<h3>Segment before you celebrate<\/h3>\n<p>Report by priority first. Then add category, client, queue, and whether the ticket was escalated. The calculation per priority is the total resolution time for resolved tickets in that priority divided by their count, compared against that priority&#039;s SLA. This is the same formula, applied to a population that shares a service expectation.<\/p>\n<p>A mean and a median can diverge even when the underlying process is stable. PlotStudio AI&#039;s primer on <a href=\"https:\/\/www.plotstudio.ai\/articles\/distribution-of-the-data\">normality tests and QQ plots<\/a> explains how distribution shape changes what an average represents, which is why a long tail can move a mean without moving the typical ticket.<\/p>\n<p><a id=\"the-long-tail-is-the-story\"><\/a><\/p>\n<h3>The long tail is the story<\/h3>\n<p>A small number of complex cases can remain open for days and raise the mean sharply while the median looks fine. The worked example already showed the pattern: one long ticket moved the mean away from the typical ticket. In a real export, those long cases are often escalations, vendor dependencies, or tickets waiting for missing information.<\/p>\n<p>Report the median and the 90th percentile alongside the mean. The median tells you what the typical ticket experiences. The 90th percentile tells you what the unlucky client experiences. The mean tells you the total burden spread across the population.<\/p>\n<p>The short walkthrough below shows why a blended average is a poor management tool.<\/p>\n<iframe width=\"100%\" style=\"aspect-ratio: 16 \/ 9\" src=\"https:\/\/www.youtube.com\/embed\/ENSlgS45v-g\" frameborder=\"0\" allow=\"autoplay; encrypted-media\" allowfullscreen><\/iframe>\n\n<p>For the wider service-desk scorecard that sits around this metric, see <a href=\"https:\/\/vight.ai\/blog\/msp-kpis\">MSP KPIs<\/a>.<\/p>\n<p><a id=\"the-monthly-reporting-template\"><\/a><\/p>\n<h3>The monthly reporting template<\/h3>\n<p>For each priority, publish five things: the mean, the median, the 90th percentile, the count of resolved tickets, and SLA attainment. Add reopen rate as a sixth line. That set takes a few minutes to produce from a PSA export and answers the questions a single average cannot.<\/p>\n<p><a id=\"where-resolution-time-leaks-in-msp-service-desks\"><\/a><\/p>\n<h2>Where Resolution Time Leaks in MSP Service Desks<\/h2>\n<p>The hours do not disappear in one place. They leak at four points in the ticket lifecycle, and each leak has a different fix.<\/p>\n<p><a id=\"intake-gaps\"><\/a><\/p>\n<h3>Intake gaps<\/h3>\n<p>A request that omits the affected location, number of users, error text, or recent system changes forces a reply before diagnosis can begin. A benchmark described by <a href=\"https:\/\/www.techprocomp.com\/learn\/why-msps-response-time-is-lying-you-what-measure-instead\/\">TechProComp<\/a>, based on more than 50,000 help-desk tickets across more than 30 organizations over 14 months, found a median resolution time of <strong>4.4 hours for heavily AI-automated help desks versus 71 hours for help desks without automation<\/strong>. The same report identified missing submission details as a major cause of delay because engineers must request and wait for those details before starting diagnosis.<\/p>\n<p>Measure time from ticket creation to first technician action. If that interval is long, the problem is intake, not repair speed.<\/p>\n<p><a id=\"after-hours-demand\"><\/a><\/p>\n<h3>After-hours demand<\/h3>\n<p>Freshworks research reported that <strong>47% of IT-support tickets<\/strong> were submitted outside standard business hours, while responses to after-hours requests generally took at least one hour longer than responses logged during the working day. The data also showed that <strong>6% to 8% of after-hours tickets<\/strong> were escalated, roughly matching business-hours escalation rates, while SLA attainment remained <strong>two to five percentage points lower<\/strong> after hours, as <a href=\"https:\/\/itbrief.news\/story\/freshworks-data-shows-after-hours-it-support-gap-widens\">ITBrief reported<\/a>.<\/p>\n<p>Measure the after-hours share of tickets and the after-hours resolution time separately. Staffing is only one part of the response. Urgency capture and escalation handling matter too.<\/p>\n<p><a id=\"handoffs-and-reopens\"><\/a><\/p>\n<h3>Handoffs and reopens<\/h3>\n<p>A 2025 incident-resolution study found that lifecycle variables such as update count, reassignment count, reopen count, priority, and category were among the most influential predictors of resolution time. Its best random-forest model using aggregated lifecycle features achieved an <strong>R\u00b2 of 0.8318<\/strong> and a mean absolute error of <strong>60.67 hours<\/strong>, according to <a href=\"https:\/\/www.iieta.org\/journals\/isi\/paper\/10.18280\/isi.310219\">the IIETA study<\/a>. Repeated updates, reassignments, and reopenings are leading indicators of fragmented ownership, incomplete intake, or weak handoffs.<\/p>\n<p>Measure reassignments per ticket and reopen rate within 14 days. Each handoff is a visible cost, not a neutral routing event.<\/p>\n<p><a id=\"vendor-dependency\"><\/a><\/p>\n<h3>Vendor dependency<\/h3>\n<p>Some tickets wait on a third-party vendor, a software publisher, or an upstream provider. The technician may have done everything possible, but the clock keeps running. Keep vendor dependency in its own bucket so it does not get mistaken for technician delay. Brittle integrations and legacy code make that dependency worse; Faberwork&#039;s note on <a href=\"https:\/\/www.faberwork.com\/latest-thinking\/legacy-code-just-got-worse-watch-out\">how AI tools worsen legacy code<\/a> is a useful warning for teams automating around old systems.<\/p>\n<blockquote>\n<p><strong>Ranking method:<\/strong> Pull last quarter&#039;s closed tickets, sum the hours in each bucket, and fix the largest one first. If reassignment is the biggest bucket, repair <a href=\"https:\/\/vight.ai\/blog\/ticket-routing\">ticket routing<\/a> before adding staff.<\/p>\n<\/blockquote>\n<p><a id=\"how-msps-cut-resolution-time-without-losing-quality\"><\/a><\/p>\n<h2>How MSPs Cut Resolution Time Without Losing Quality<\/h2>\n<p>The fastest way to cut resolution time is not to tell technicians to work faster. It is to remove the minutes that never required a technician in the first place. That means fixing intake, routing, and knowledge reuse while protecting the first-time fix.<\/p>\n<p><a id=\"fix-the-first-hour-not-the-last\"><\/a><\/p>\n<h3>Fix the first hour, not the last<\/h3>\n<p>Set an intake standard. Require location, number of users, error text, and recent changes before a ticket reaches a technician. Add automated follow-up that asks the client for missing details within minutes rather than at the next business day. Vight&#039;s ticket triage emails the client for missing details, records the response, and moves the ticket to a ready-to-work status, which targets the intake gap without adding a technician step.<\/p>\n<p>After-hours intake should capture urgency and create a complete ticket when staff are away. Knowledge reuse should surface earlier fixes for matching symptoms so technicians do not rediscover the same solution. Set a handoff limit: escalate after a defined number of reassignments and name one owner.<\/p>\n<p><figure class=\"wp-block-image size-large\"><img decoding=\"async\" src=\"https:\/\/vight.ai\/blog\/wp-content\/uploads\/2026\/11\/mean-time-to-resolution-msp-efficiency.jpg\" alt=\"A graphic infographic listing four best practices for Managed Service Providers to reduce resolution times efficiently.\" \/><\/figure><\/p>\n<p><a id=\"the-ai-shortcut-has-a-quality-bill\"><\/a><\/p>\n<h3>The AI shortcut has a quality bill<\/h3>\n<p>A 2025 analysis covering more than 2,000 ITSM systems and over 60,000 anonymized data points reported average resolution time of <strong>32.46 hours for organizations without GenAI<\/strong> and <strong>22.55 hours for GenAI users<\/strong>. Another report summarized the change as a <strong>17.8% reduction<\/strong>, or <strong>4.87 hours saved per ticket<\/strong>, according to <a href=\"https:\/\/www.solarwinds.com\/company\/newsroom\/press-releases\/state-of-itsm-2025\">SolarWinds&#039; State of ITSM 2025<\/a>. The gap does not prove that AI alone caused the improvement. Process maturity, incident mix, staffing, documentation quality, and willingness to automate all differ between those groups.<\/p>\n<p>AI-assisted intake can cut time lost to missing details and produce more complete records. Poor summaries, incorrect triage, or rapid workaround-based closure can raise reopenings and repeat contacts. A shorter average does not establish a better customer outcome.<\/p>\n\n<figure class=\"wp-block-table\"><table><tr>\n<th>Lever<\/th>\n<th>What to change<\/th>\n<th>Guard metric<\/th>\n<\/tr>\n<tr>\n<td>Complete intake fields<\/td>\n<td>Require location, user count, error text, and recent changes<\/td>\n<td>Wait time for client information<\/td>\n<\/tr>\n<tr>\n<td>Automated missing-detail follow-up<\/td>\n<td>Ask the client by email before a technician starts<\/td>\n<td>Time to first technician action<\/td>\n<\/tr>\n<tr>\n<td>After-hours intake<\/td>\n<td>Capture urgency and create a complete ticket when staff are away<\/td>\n<td>After-hours resolution time and SLA attainment<\/td>\n<\/tr>\n<tr>\n<td>Knowledge reuse<\/td>\n<td>Surface earlier fixes for matching symptoms<\/td>\n<td>Reopen rate<\/td>\n<\/tr>\n<tr>\n<td>Handoff limit<\/td>\n<td>Escalate after a set number of reassignments and name one owner<\/td>\n<td>Reassignments per ticket and escalation rate<\/td>\n<\/tr>\n<\/table><\/figure>\n<p><a id=\"protect-the-first-time-fix\"><\/a><\/p>\n<h3>Protect the first-time fix<\/h3>\n<p>Pair every lever with a quality metric. Track reopen rate, repeat-contact rate, customer-confirmed resolution, technician touch time, and gross profit per ticket. A shorter MTTR with rising reopens is not an improvement. It is a transfer of work from the first ticket to the second one.<\/p>\n<p>The question for MSP leaders is not how to lower MTTR. It is which minutes can be removed without reducing first-time-fix quality or margin. For the automation side of that question, see <a href=\"https:\/\/vight.ai\/blog\/help-desk-automation\">help desk automation<\/a>.<\/p>\n<p><a id=\"test-one-lever-for-four-weeks\"><\/a><\/p>\n<h3>Test one lever for four weeks<\/h3>\n<p>Change one lever for one client segment for four weeks. Keep a control group on the old process. Compare median resolution time, 90th-percentile resolution time, reopen rate, and gross profit per ticket. A controlled cohort tells you whether the change worked. A before-and-after chart on the whole business tells you what else happened that month.<\/p>\n<p><a id=\"set-up-your-resolution-time-dashboard-this-month\"><\/a><\/p>\n<h2>Set Up Your Resolution Time Dashboard This Month<\/h2>\n<p><figure class=\"wp-block-image size-large\"><img decoding=\"async\" src=\"https:\/\/vight.ai\/blog\/wp-content\/uploads\/2026\/11\/mean-time-to-resolution-dashboard-setup.jpg\" alt=\"A professional infographic outlining three steps to set up a resolution time dashboard for tracking service tickets.\" \/><\/figure><\/p>\n<p>A lower MTTR can reflect a faster service desk, or a changed clock. Before celebrating a drop, document the start event, stop event, and paused-time rules your team will use, then keep that definition consistent.<\/p>\n<p><a id=\"the-five-steps\"><\/a><\/p>\n<h3>The five steps<\/h3>\n<ol>\n<li>Write the start and stop events and paused-time rules into the PSA status model.<\/li>\n<li>Export 90 days of resolved tickets with priority, category, client, and reassignment count.<\/li>\n<li>Calculate mean, median, 90th percentile, ticket count, and SLA attainment for each priority.<\/li>\n<li>Add reopen rate and repeat contacts within 14 days.<\/li>\n<li>Set one target per priority based on the SLA, not a benchmark headline.<\/li>\n<\/ol>\n<p>A PSA export and spreadsheet are enough for a first version. Keep the definition fixed and the measures visible; a new data warehouse or BI tool is not a prerequisite.<\/p>\n<p><a id=\"how-to-read-month-one\"><\/a><\/p>\n<h3>How to read month one<\/h3>\n<p>A definition change can move the mean without changing how quickly technicians work. Record any change beside the result and retain the prior figure for comparison. For example, switching from ticket closure to client-confirmed resolution may raise the reported time, while pausing the clock for client responses may lower it.<\/p>\n<p>Schedule the first review 30 days after launch. Check that the clock definition stayed consistent, then interpret the metrics. A stable definition makes a flat result useful; a falling result under shifting rules does not show whether service improved.<\/p>\n<p>Vight transcribes and summarizes each service call, then posts a structured time entry to your PSA with speaker identification, the issue, steps taken, and resolution. It connects to your existing phone system and PSA, and teams are typically up and running within a day. Visit <a href=\"https:\/\/vight.ai\">Vight<\/a> to see how call and ticket records can support the dashboard you have defined.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Mean time to resolution is the average elapsed time between a ticket being opened and being closed, calculated by dividing the total resolution time of all resolved tickets by the number of resolved tickets. Incident MTTR averages 8.40 business hours in global benchmarking, with results ranging from 0.67 to 33.67 business hours. The most popular [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":1274,"comment_status":"closed","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[100,97,16,98,99],"class_list":["post-1275","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-msp-operations","tag-itsm-metrics","tag-mean-time-to-resolution","tag-msp-service-desk","tag-mttr","tag-ticket-resolution-time"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v28.8 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Mean Time to Resolution: How to Measure and Improve It<\/title>\n<meta name=\"description\" content=\"Mean time to resolution for MSPs: the formula, how to read it, measurement traps, and ways to cut it without hurting quality.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/vight.ai\/blog\/mean-time-to-resolution\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Mean Time to Resolution: How to Measure and Improve It\" \/>\n<meta property=\"og:description\" content=\"Mean time to resolution for MSPs: the formula, how to read it, measurement traps, and ways to cut it without hurting quality.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/vight.ai\/blog\/mean-time-to-resolution\/\" \/>\n<meta property=\"og:site_name\" content=\"Vight Insights | AI for MSPs\" \/>\n<meta property=\"article:published_time\" content=\"2026-11-22T07:19:15+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2026-11-22T07:19:17+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/vight.ai\/blog\/wp-content\/uploads\/2026\/11\/mean-time-to-resolution-performance-metrics.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"1672\" \/>\n\t<meta property=\"og:image:height\" content=\"941\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"Nathan Spitalny\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Nathan Spitalny\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"16 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/vight.ai\\\/blog\\\/mean-time-to-resolution\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/vight.ai\\\/blog\\\/mean-time-to-resolution\\\/\"},\"author\":{\"name\":\"Nathan Spitalny\",\"@id\":\"https:\\\/\\\/vight.ai\\\/blog\\\/#\\\/schema\\\/person\\\/d256cba1d191205aed6403d4604a2794\"},\"headline\":\"Mean Time to Resolution: How to Measure and Improve It\",\"datePublished\":\"2026-11-22T07:19:15+00:00\",\"dateModified\":\"2026-11-22T07:19:17+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/vight.ai\\\/blog\\\/mean-time-to-resolution\\\/\"},\"wordCount\":3296,\"publisher\":{\"@id\":\"https:\\\/\\\/vight.ai\\\/blog\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/vight.ai\\\/blog\\\/mean-time-to-resolution\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/vight.ai\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/11\\\/mean-time-to-resolution-performance-metrics.jpg\",\"keywords\":[\"ITSM metrics\",\"mean time to resolution\",\"MSP service desk\",\"MTTR\",\"ticket resolution time\"],\"articleSection\":[\"MSP Operations\"],\"inLanguage\":\"en-US\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/vight.ai\\\/blog\\\/mean-time-to-resolution\\\/\",\"url\":\"https:\\\/\\\/vight.ai\\\/blog\\\/mean-time-to-resolution\\\/\",\"name\":\"Mean Time to Resolution: How to Measure and Improve It\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/vight.ai\\\/blog\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/vight.ai\\\/blog\\\/mean-time-to-resolution\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/vight.ai\\\/blog\\\/mean-time-to-resolution\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/vight.ai\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/11\\\/mean-time-to-resolution-performance-metrics.jpg\",\"datePublished\":\"2026-11-22T07:19:15+00:00\",\"dateModified\":\"2026-11-22T07:19:17+00:00\",\"description\":\"Mean time to resolution for MSPs: the formula, how to read it, measurement traps, and ways to cut it without hurting quality.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/vight.ai\\\/blog\\\/mean-time-to-resolution\\\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/vight.ai\\\/blog\\\/mean-time-to-resolution\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/vight.ai\\\/blog\\\/mean-time-to-resolution\\\/#primaryimage\",\"url\":\"https:\\\/\\\/vight.ai\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/11\\\/mean-time-to-resolution-performance-metrics.jpg\",\"contentUrl\":\"https:\\\/\\\/vight.ai\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/11\\\/mean-time-to-resolution-performance-metrics.jpg\",\"width\":1672,\"height\":941},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/vight.ai\\\/blog\\\/mean-time-to-resolution\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/vight.ai\\\/blog\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Mean Time to Resolution: How to Measure and Improve It\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/vight.ai\\\/blog\\\/#website\",\"url\":\"https:\\\/\\\/vight.ai\\\/blog\\\/\",\"name\":\"Vight Insights | AI for MSPs\",\"description\":\"AI Automation, MSP Growth Strategies &amp; Smarter IT Operations\",\"publisher\":{\"@id\":\"https:\\\/\\\/vight.ai\\\/blog\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/vight.ai\\\/blog\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/vight.ai\\\/blog\\\/#organization\",\"name\":\"Vight Insights | AI for MSPs\",\"url\":\"https:\\\/\\\/vight.ai\\\/blog\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/vight.ai\\\/blog\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/vight.ai\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/11\\\/vight-logo-square.png\",\"contentUrl\":\"https:\\\/\\\/vight.ai\\\/blog\\\/wp-content\\\/uploads\\\/2026\\\/11\\\/vight-logo-square.png\",\"width\":800,\"height\":800,\"caption\":\"Vight Insights | AI for MSPs\"},\"image\":{\"@id\":\"https:\\\/\\\/vight.ai\\\/blog\\\/#\\\/schema\\\/logo\\\/image\\\/\"}},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/vight.ai\\\/blog\\\/#\\\/schema\\\/person\\\/d256cba1d191205aed6403d4604a2794\",\"name\":\"Nathan Spitalny\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/850f7fc4c9dff63a9b775252efe7c5b5bf78f8a543e3791de6c9775be5539970?s=96&d=mm&r=g\",\"url\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/850f7fc4c9dff63a9b775252efe7c5b5bf78f8a543e3791de6c9775be5539970?s=96&d=mm&r=g\",\"contentUrl\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/850f7fc4c9dff63a9b775252efe7c5b5bf78f8a543e3791de6c9775be5539970?s=96&d=mm&r=g\",\"caption\":\"Nathan Spitalny\"},\"sameAs\":[\"http:\\\/\\\/127.0.0.1\"],\"url\":\"https:\\\/\\\/vight.ai\\\/blog\\\/author\\\/user\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Mean Time to Resolution: How to Measure and Improve It","description":"Mean time to resolution for MSPs: the formula, how to read it, measurement traps, and ways to cut it without hurting quality.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/vight.ai\/blog\/mean-time-to-resolution\/","og_locale":"en_US","og_type":"article","og_title":"Mean Time to Resolution: How to Measure and Improve It","og_description":"Mean time to resolution for MSPs: the formula, how to read it, measurement traps, and ways to cut it without hurting quality.","og_url":"https:\/\/vight.ai\/blog\/mean-time-to-resolution\/","og_site_name":"Vight Insights | AI for MSPs","article_published_time":"2026-11-22T07:19:15+00:00","article_modified_time":"2026-11-22T07:19:17+00:00","og_image":[{"width":1672,"height":941,"url":"https:\/\/vight.ai\/blog\/wp-content\/uploads\/2026\/11\/mean-time-to-resolution-performance-metrics.jpg","type":"image\/jpeg"}],"author":"Nathan Spitalny","twitter_card":"summary_large_image","twitter_misc":{"Written by":"Nathan Spitalny","Est. reading time":"16 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/vight.ai\/blog\/mean-time-to-resolution\/#article","isPartOf":{"@id":"https:\/\/vight.ai\/blog\/mean-time-to-resolution\/"},"author":{"name":"Nathan Spitalny","@id":"https:\/\/vight.ai\/blog\/#\/schema\/person\/d256cba1d191205aed6403d4604a2794"},"headline":"Mean Time to Resolution: How to Measure and Improve It","datePublished":"2026-11-22T07:19:15+00:00","dateModified":"2026-11-22T07:19:17+00:00","mainEntityOfPage":{"@id":"https:\/\/vight.ai\/blog\/mean-time-to-resolution\/"},"wordCount":3296,"publisher":{"@id":"https:\/\/vight.ai\/blog\/#organization"},"image":{"@id":"https:\/\/vight.ai\/blog\/mean-time-to-resolution\/#primaryimage"},"thumbnailUrl":"https:\/\/vight.ai\/blog\/wp-content\/uploads\/2026\/11\/mean-time-to-resolution-performance-metrics.jpg","keywords":["ITSM metrics","mean time to resolution","MSP service desk","MTTR","ticket resolution time"],"articleSection":["MSP Operations"],"inLanguage":"en-US"},{"@type":"WebPage","@id":"https:\/\/vight.ai\/blog\/mean-time-to-resolution\/","url":"https:\/\/vight.ai\/blog\/mean-time-to-resolution\/","name":"Mean Time to Resolution: How to Measure and Improve It","isPartOf":{"@id":"https:\/\/vight.ai\/blog\/#website"},"primaryImageOfPage":{"@id":"https:\/\/vight.ai\/blog\/mean-time-to-resolution\/#primaryimage"},"image":{"@id":"https:\/\/vight.ai\/blog\/mean-time-to-resolution\/#primaryimage"},"thumbnailUrl":"https:\/\/vight.ai\/blog\/wp-content\/uploads\/2026\/11\/mean-time-to-resolution-performance-metrics.jpg","datePublished":"2026-11-22T07:19:15+00:00","dateModified":"2026-11-22T07:19:17+00:00","description":"Mean time to resolution for MSPs: the formula, how to read it, measurement traps, and ways to cut it without hurting quality.","breadcrumb":{"@id":"https:\/\/vight.ai\/blog\/mean-time-to-resolution\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/vight.ai\/blog\/mean-time-to-resolution\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/vight.ai\/blog\/mean-time-to-resolution\/#primaryimage","url":"https:\/\/vight.ai\/blog\/wp-content\/uploads\/2026\/11\/mean-time-to-resolution-performance-metrics.jpg","contentUrl":"https:\/\/vight.ai\/blog\/wp-content\/uploads\/2026\/11\/mean-time-to-resolution-performance-metrics.jpg","width":1672,"height":941},{"@type":"BreadcrumbList","@id":"https:\/\/vight.ai\/blog\/mean-time-to-resolution\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/vight.ai\/blog\/"},{"@type":"ListItem","position":2,"name":"Mean Time to Resolution: How to Measure and Improve It"}]},{"@type":"WebSite","@id":"https:\/\/vight.ai\/blog\/#website","url":"https:\/\/vight.ai\/blog\/","name":"Vight Insights | AI for MSPs","description":"AI Automation, MSP Growth Strategies &amp; Smarter IT Operations","publisher":{"@id":"https:\/\/vight.ai\/blog\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/vight.ai\/blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/vight.ai\/blog\/#organization","name":"Vight Insights | AI for MSPs","url":"https:\/\/vight.ai\/blog\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/vight.ai\/blog\/#\/schema\/logo\/image\/","url":"https:\/\/vight.ai\/blog\/wp-content\/uploads\/2026\/11\/vight-logo-square.png","contentUrl":"https:\/\/vight.ai\/blog\/wp-content\/uploads\/2026\/11\/vight-logo-square.png","width":800,"height":800,"caption":"Vight Insights | AI for MSPs"},"image":{"@id":"https:\/\/vight.ai\/blog\/#\/schema\/logo\/image\/"}},{"@type":"Person","@id":"https:\/\/vight.ai\/blog\/#\/schema\/person\/d256cba1d191205aed6403d4604a2794","name":"Nathan Spitalny","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/secure.gravatar.com\/avatar\/850f7fc4c9dff63a9b775252efe7c5b5bf78f8a543e3791de6c9775be5539970?s=96&d=mm&r=g","url":"https:\/\/secure.gravatar.com\/avatar\/850f7fc4c9dff63a9b775252efe7c5b5bf78f8a543e3791de6c9775be5539970?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/850f7fc4c9dff63a9b775252efe7c5b5bf78f8a543e3791de6c9775be5539970?s=96&d=mm&r=g","caption":"Nathan Spitalny"},"sameAs":["http:\/\/127.0.0.1"],"url":"https:\/\/vight.ai\/blog\/author\/user\/"}]}},"jetpack_featured_media_url":"https:\/\/vight.ai\/blog\/wp-content\/uploads\/2026\/11\/mean-time-to-resolution-performance-metrics.jpg","_links":{"self":[{"href":"https:\/\/vight.ai\/blog\/wp-json\/wp\/v2\/posts\/1275","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/vight.ai\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/vight.ai\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/vight.ai\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/vight.ai\/blog\/wp-json\/wp\/v2\/comments?post=1275"}],"version-history":[{"count":1,"href":"https:\/\/vight.ai\/blog\/wp-json\/wp\/v2\/posts\/1275\/revisions"}],"predecessor-version":[{"id":1280,"href":"https:\/\/vight.ai\/blog\/wp-json\/wp\/v2\/posts\/1275\/revisions\/1280"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/vight.ai\/blog\/wp-json\/wp\/v2\/media\/1274"}],"wp:attachment":[{"href":"https:\/\/vight.ai\/blog\/wp-json\/wp\/v2\/media?parent=1275"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/vight.ai\/blog\/wp-json\/wp\/v2\/categories?post=1275"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/vight.ai\/blog\/wp-json\/wp\/v2\/tags?post=1275"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}