Customer Service Checklist Template

Fill in the first task before the desk opens: the team, the time the shift starts and who does what today. Every due date counts in hours from the shift start, so the checklist runs to your hours. Tick the channels the team answers today: each one adds its own opening task.

• Record the day, the shift start and who does what — the team or queue this run covers, the date and time the shift starts, which every due date counts from, the channels the team answers today, and the team lead who signs off the day, the queue coordinator, the QA reviewer, the knowledge base owner and the support manager who receives anything the team lead refers; later tasks are assigned from these fields
• Check the rota and who covers each channel — who is on shift, their hours and breaks, which channel each agent covers first, and who is off or late; move cover now if a channel would be left without anyone at a busy time
• Read the last handover and today’s known issues — the handover note from the last shift, promises due today such as call-backs, any known product issue, outage or release today, and the message on your status page or website if you have one, so every agent gives customers the same answer
• Open the phone lines — shown when the team answers the phone today: agents logged in to the phone system, the queue and opening-hours messages right for today, voicemail checked and each call-back request logged as a ticket, and any recorded announcement your team uses playing
• Open the shared inbox and web form queue — shown when the team answers email: the shared inbox or help desk queue checked for messages that arrived overnight, the auto-reply set to today’s hours, and anything misfiled or caught as spam by mistake moved into the queue
• Put live chat online — shown when the team answers live chat: the chat widget online on the website or app, agents set as available, the opening hours and offline message right for today, and chats left overnight turned into tickets
• Check social media messages and mentions — shown when the team answers on social media: direct messages and the public mentions your team replies to, as your social media policy sets, with any question about an account moved to a private message or a ticket before account details are discussed

Look at the whole queue before working it. Count what is waiting on each channel, find the tickets past or close to their SLA target and deal with those first, and move people to the channel that needs them. A ticket past or near its target adds a task to give each one an owner and update the customer.

• Count the queue on each channel — the tickets and contacts open on each channel, how many arrived since the last shift, how many are waiting on the team rather than on the customer, and the age of the oldest; leave a channel’s row blank if the team does not answer it
• Check tickets past or near their SLA target — every open ticket against your SLA targets for its priority, for the first response and for resolution if you set both: the tickets already past their target and those that will pass it before the end of the shift
• Give each late or at-risk ticket an owner and update the customer — shown when a ticket is past or near its target: an owner for each, a reply or an honest update to the customer with the time they will hear from you next, and a note of the cause when it is the queue rather than the ticket
• Move cover to the channels that need it — from the queue counts and the late tickets: agents moved between channels, or from email to the phone at busy times, and a note of it if the queue cannot be cleared today with the people on shift
• Link tickets about the same issue — contacts from several customers about one known issue linked to it, one agreed reply or macro used so every customer gets the same answer, and anything that looks like a problem affecting many customers passed to whoever runs your incident process

Urgent tickets and old ones are where customers wait longest. Check each has an owner and a next step, chase the teams you are waiting on, follow up the tickets waiting on the customer, and raise an escalation for any ticket that meets your escalation criteria.

• Review today’s urgent and high-priority tickets — each one with an owner, a next step and the time the customer was told they will hear back, with the priority checked against what the customer can and cannot do while they wait
• Review tickets open past your ageing limit — every ticket open longer than the limit you set for its priority, why it is stuck, who owns it and the next step; this template suggests one ageing limit per priority, kept the same every day so the count means something
• Chase tickets waiting on another team — tickets waiting on engineering, billing, delivery or a supplier, with the date each team gave and the update owed to the customer; where no date has been given, ask for one
• Follow up tickets waiting on the customer — tickets waiting for the customer to reply: a reminder sent as your policy says, and a ticket closed only after the follow-ups your policy sets, with a note that the customer can reopen it by replying
• Raise escalations that meet your criteria — any ticket that meets your escalation criteria, such as a critical problem, a customer whose own deadline is at risk or a request to speak to someone senior, raised now rather than at the end of the day, with a Customer Escalation Checklist opened for each so it has one owner

Check a sample of the day’s replies against the team’s service standard, the same seven points every day. The standard on the scoring task is this template’s suggestion; change it to your own. A reply below the standard adds tasks to give the agent feedback and to put right any wrong answer with the customer.

• Pick the QA sample — replies sent since the last sample, across agents and channels, chosen at random rather than by the reviewer, plus any reply a customer rated poorly; this template suggests about two replies for each agent on shift, which over a week shows each person’s work
• Score each reply against the service standard — each reply marked Met or Not met on the seven standards, and the number met out of seven; this template counts a reply as below the standard when the answer was inaccurate or the identity check was missed, or when two or more standards were not met
• Check every promise made in the sampled replies — refunds, credits, call-backs, replacement dates and fixes promised to customers, each logged where the team will see it, and each within what the team is allowed to offer
• Give feedback to each agent below the standard — shown when a reply fell below the standard: a short private conversation today with the reply in front of you, what met the standard as well as what did not, and anything the knowledge base or a macro got wrong noted for Phase 6 rather than blamed on the agent
• Put right any wrong answer with the customer — shown when a reply fell below the standard: where a customer was told something wrong or a step was missed, the queue coordinator makes sure the ticket’s owner sends a corrected reply today, with the ticket reopened if needed
• Read today’s customer ratings and comments — the satisfaction survey replies received today, if your help desk sends a survey: the comments on low ratings read in full, and a reply or call-back arranged where the customer asked for one or is still waiting

Log complaints and acknowledge them as they arrive; this phase is the day’s check that none was missed and that every refund or credit agreed today is on the record. A complaint today adds tasks to log it, acknowledge it and look for a pattern, and a refund or credit above the team’s limit adds an approval.

• Check whether any complaints came in today — every channel, including replies to the satisfaction survey and public social media posts; this template counts as a complaint any contact where the customer says they are unhappy with the product, the service or how they were treated, whether or not they use the word
• Log each complaint with an owner — shown when a complaint came in: the customer, what they are unhappy about in their own words, the owner, when it was acknowledged and the date by which they will have an answer, in your complaints log as well as on the ticket
• Acknowledge each complaint and say what happens next — shown when a complaint came in: a reply today to every complaint not yet acknowledged, saying that you have it, who is dealing with it and when the customer will hear back, without arguing the point in the acknowledgement
• Look for a pattern in today’s complaints — shown when a complaint came in: the same cause, product, policy or agent behind more than one complaint today or this week; a pattern goes in the end-of-day summary, and the team lead decides whether to refer it to the support manager
• Log refunds, credits and goodwill gestures agreed today — each refund, account credit, replacement or goodwill gesture agreed today, its amount, the reason and who agreed it, checked against your refund policy and the team’s approval limit
• Get approval for refunds or credits above the limit — shown when a refund or credit is above the team’s limit: the support manager, or whoever your refund policy names, approves or refuses it before the customer is told it is agreed; if it is refused, the ticket owner tells the customer what can be offered instead

The day’s tickets show what the knowledge base and the saved replies do not cover. Write the gaps down while they are fresh, fix the ones that caused wrong answers first, and tell the team about today’s known issues and workarounds.

• List questions the knowledge base did not answer — from today’s tickets, chats and the QA sample: questions customers asked that no article answers, articles customers were sent that did not help, and searches by agents that found nothing
• Flag macros that gave a wrong or out-of-date answer — saved replies with a wrong price, policy, step or product name, and any the QA sample showed agents rewriting by hand every time
• Update or draft the articles and macros — the ones that caused a wrong answer today fixed first and the rest drafted with a date, each checked by someone who knows the product or policy before it goes live
• Share today’s known issues and workarounds with the team — a short note in the team’s channel with each known issue, the reply to use and what to tell customers about timing, so the next shift and tomorrow’s team start from the same answer

Nothing urgent should wait overnight without a name on it. Give every open urgent ticket an owner for the next shift, write the handover note, hand over the call-backs and follow-ups promised, and set each channel for out of hours.

• Give every open urgent ticket an owner for the next shift — each urgent or high-priority ticket still open with a named owner on the next shift or tomorrow, and the customer told when they will next hear from you
• Write the handover note — open urgent tickets and their owners, promises due on the next shift, known issues, anything waiting on another team, complaints in progress and changes to cover, short enough to read at the start of the next shift
• Hand over call-backs and promised follow-ups — every call-back or follow-up promised for a set time, with the customer, the time promised, the channel and who will make it
• Set each channel for out of hours — phone lines to the out-of-hours message or the next team, chat offline with the right message, the inbox auto-reply saying when customers can expect a reply, and social media messages handed over if someone covers them out of hours

Record the day’s numbers and a short summary, then the team lead reviews the day and signs it off. The sign-off is the gate: the report cannot be sent and nothing after it can be completed until it is recorded. The team lead also decides whether anything goes to the support manager.

• Record the day’s numbers by channel — contacts received, resolved and still open at the close on each channel, and the tickets that went past their SLA target, taken from your help desk and phone system at the same time each day so days can be compared
• Write the summary for the team lead — what went well, tickets past target and why, escalations raised, complaints, refunds above the limit, the QA result, knowledge base gaps and any problem with cover, with anything you think should go to the support manager
• Team lead reviews and signs off the day — the team lead checks the queue, the handover, the QA result, complaints and refunds, and records one decision: signed off, signed off and referred to the support manager, or not signed off until something is fixed; refer, for example, an urgent ticket past its SLA target, a pattern of complaints, a reply below the standard that reached customers or a cover problem the team cannot solve

Once the day is signed off, the report goes to the people who need it. A referral adds a task for the team lead to send it and one for the support manager to record what happens next. A day that is not signed off adds a task to fix what the team lead raised.

• Send the end-of-day report — the numbers, the summary and the team lead’s decision to the support manager and anyone else who receives it, and the handover note to the next shift if it has not gone already; if the day was not signed off, send it once the fix is done and the decision reads signed off
• Refer the issue to the support manager — shown when the day is signed off and referred: what happened, the tickets or complaints involved, what the team has done so far and what the team lead is asking for, sent today
• Fix what the team lead raised and ask for sign-off again — shown when the day is not signed off: put right what the team lead named, such as an urgent ticket with no owner or a missing handover note, then ask the team lead to change the decision on the sign-off task to signed off, or to un-complete that task and complete it again
• Support manager records the response to the referral — shown when the day is referred: what will happen, who owns it and by when, recorded by the next working day so the team lead can tell the team