18 seconds. That is how fast the top 20% of teams in Freshworks’ 2025 benchmark send a first reply to WhatsApp support messages. How fast is yours?


Most teams cannot answer that question. They know support feels busy. They do not know the numbers.

That is where WhatsApp support SLAs and the right WhatsApp support metrics come in. An SLA is a promise. Metrics tell you if you keep it.

This guide shows you what to promise, what to measure, and how to read the numbers. It uses simple steps and real data. You can use it whether you have two agents or twenty.

 

Quick answer

The WhatsApp support metrics worth tracking are: first response time, average response time, resolution time, first contact resolution, SLA compliance rate, CSAT, customer effort score and NPS, backlog and workload, bot handoff metrics, and channel health (WhatsApp quality rating and delivery status). If you are just starting, track first response time, resolution time, and workload.

 

Table of Contents

 

What is a WhatsApp support SLA?

A service level agreement (SLA) is a promise about the level of service you will give. It uses numbers you can measure. Common ones are response time and resolution time.

There are two kinds:

  • External SLA: what you tell customers. Example: “We reply within 5 minutes, 9 AM to 6 PM.”
  • Internal SLA: what you set for your team. It is often tighter, so you have a safety buffer.

A metric (or KPI) is what you measure. The SLA is the target. The metric is the score.

SLA = the promise. Metric = the proof.

Why WhatsApp needs its own SLAs

You cannot copy your email SLA to WhatsApp. People use the two channels in very different ways.

Customers see WhatsApp as a fast chat. A survey of 400 consumers by Nextiva set the expected wait at one minute for chat. For SMS and tickets, it was five minutes. For email, it was 30 minutes.

WhatsApp also has a rule that email does not have: the 24-hour customer service window.

The 24-hour window is a built-in clock

When a customer messages you, a 24-hour window opens. Each new customer message resets the timer. Inside the window, you can reply with any free-form message. Outside it, you can only send approved template messages. This comes from Meta’s WhatsApp Business Platform rules.

 

Why this matters for SLAs

Say a customer writes at 10 AM Monday. Your team replies at 11 AM Tuesday. The window has closed. You now need an approved template to reopen the chat. So your resolution target should usually fit inside the window.

 

 

There is a cost angle too, and it changed recently. WhatsApp has billed per message since 1 July 2025. From 1 October 2026, Meta also bills service messages, which are free-form replies sent inside an open window. Each business phone number gets 1,000 free service messages a month. After that, each delivered reply is charged at a rate that varies by country. Utility templates sent inside the window are chargeable too. Messages customers send to you stay free.

So every extra back-and-forth can now cost money. Solving the issue in fewer messages protects both your SLA and your budget. Check Meta’s pricing page for the rates in your market.

Setting up the channel well makes every SLA easier to hit. If you are still building your setup, start with our complete guide to WhatsApp customer support. It covers the team, tools, and flows that sit behind the numbers in this article.

The 10 WhatsApp support metrics to track

Do not track everything on day one. Pick a few and track them well. Here are the ten, from most important to supporting.

1. First response time (FRT)

First response time is how long a customer waits for the first reply. It is the headline metric on WhatsApp. People expect fast answers here.

FRT = time of first reply − time of customer’s first message

What good looks like:

  • In Freshworks’ 2025 Customer Service Benchmark Report, which used 2024 data from its customers, “trendsetters” (the top 20% of performers) averaged 18 seconds on WhatsApp.
  • Freshchat’s 2023 report covered over 118 million conversations across 25 industries. In 80% of the industry verticals it analyzed, the first response came within 10 minutes.

 

Watch out

Does an auto-reply count? It depends on the tool and your policy. Nextiva, for example, says a plain auto-acknowledgement does not count. A reply must address the question. Track two numbers: first bot reply and first human reply. A fast “Thanks, we got your message” does not mean the customer got help.

 

 

2. Average response time (reply gaps)

FRT only measures the first reply. Average response time measures the wait after every customer message. A team can reply fast once and then go quiet. This metric catches that.

Use the median and the 90th percentile, not just the average. An average hides long waits. The 90th percentile shows what your slowest 10% of customers feel.

3. Resolution time

Resolution time is the time from the first message to the issue being solved. A fast first reply can still sit on top of a slow fix. This metric exposes that.

Resolution time = time issue marked resolved − time of customer’s first message

It changes a lot by industry and issue. In the Freshchat 2023 benchmark, resolution time ranged from 28 minutes to 11 hours across industries. So set targets by issue type, not one number for all.

4. First contact resolution (FCR)

FCR is the share of issues solved without the customer coming back. Chakra notes that high FCR generally goes with lower support costs and higher satisfaction. Having to repeat yourself is also a classic source of customer effort (see metric 7).

FCR % = (issues solved in first contact ÷ total issues) × 100

On WhatsApp, “first contact” is a little blurry because chats continue over hours. Freshchat counts an issue as first-contact resolved if it is solved within the first session of 24 hours. That matches the WhatsApp window well. Also track customers who return about the same issue. And now that service replies can be billed, count messages per resolved chat. It shows how many replies each fix takes.

5. SLA compliance rate

This tells you how often you kept your promise.

SLA compliance % = (chats that met the target ÷ total chats) × 100

Example: You handled 200 chats. 180 got a first reply inside your 5-minute target. Your compliance rate is 90%.

ChatArchitect’s WhatsApp SLA guide suggests aiming for 95% of chats to meet their targets. Treat that as a starting point. Set your own based on team size and hours.

6. Customer satisfaction (CSAT)

CSAT asks one simple question after a chat: “How satisfied were you?” Customers answer on a 1 to 5 scale. WhatsApp quick-reply buttons make this easy to send.

CSAT % = (answers of 4 or 5 ÷ total answers) × 100

For a top-tier reference, Freshworks reported a 95.63% CSAT for its top-performing WhatsApp teams in the same 2024 data. Use it as a stretch goal. Compare your own score month over month first.

7. Customer effort score (CES) and NPS

These two look at the bigger picture.

  • CES: “How easy was it to solve your issue?” Often scored 1 to 7.
  • NPS: “How likely are you to recommend us?” on a 0 to 10 scale. NPS = % of promoters (9 to 10) minus % of detractors (0 to 6).

Effort deserves attention. Research by CEB, now part of Gartner, published in 2010 found that 96% of customers with a high-effort service experience became more disloyal. For low-effort experiences, it was only 9%. The study is older and was not WhatsApp-specific, but the lesson still holds for chat.

96% of customers with a high-effort service experience became more disloyal. Only 9% did after a low-effort one.

Speed helps, but easy matters more. If the customer has to repeat their order number three times, that is high effort, even if you replied in 10 seconds.

8. Backlog and workload

Track open chats, chats waiting for an agent, and the age of your oldest open chat. Also track how many chats each agent handles at once (concurrency).

Chakra’s guide to WhatsApp team inbox management names first response time, resolution time, and concurrency as the first three to set up. Together they catch unanswered chats, false efficiency, and silent overload. Heavy workload can lead to slower replies, so watch it early.

9. Bot and handoff metrics

If you use a chatbot, track these:

  • Bot resolution rate: chats solved with no human.
  • Handoff rate: chats passed to a human.
  • Handoff wait time: how long a customer waits after asking for a human.

YCloud gives a smart warning here. Do not push one number, such as bot containment, without checking that customers got the right help. A bot that “contains” a chat by confusing the customer is not winning.

10. Channel health: quality rating and delivery

This is the metric group most teams forget. Meta gives each business phone number a quality rating. It is based on customer feedback, such as blocks and spam reports, over the last seven days. Poor quality can reduce your messaging limits. Templates also get their own quality status, and low-quality templates can be paused.

Check these in WhatsApp Manager. Also watch your message status: sent, delivered, read, and failed. A high number of failed messages means your customers may never see your replies.

Metrics cheat sheet and starting targets

The targets below are starting points, not rules. Use them to set a first version. Then adjust with your own data.

WhatsApp support metrics cheat sheet: what each metric tells you and a starting target
Metric What it tells you Starting target
First response time (human) Do customers feel heard? Under 5 minutes in working hours. Tighten over time.
Reply gap (90th percentile) How long your slowest chats wait Track for 4 weeks, then set a target from your own data
Resolution time How long fixes take Set by issue type. Aim to finish inside the 24-hour window.
First contact resolution (FCR) Do customers need to return? Rise month over month
Messages per resolved chat How many replies each fix takes (and costs) Fall month over month
SLA compliance rate Do you keep your promise? Around 95% (a common suggestion, adjust to your team)
CSAT Were customers happy? Track the trend. Aim to beat your last month.
Backlog age Are chats getting stuck? No open chat older than your SLA
Bot handoff wait Is the human handoff smooth? Same as your human first response target
WhatsApp quality rating Is your number in good standing? High (green)

Note: These targets are starting points, not industry rules. Adjust them with your own data.

 

How to write your WhatsApp SLA, step by step

Step 1: Pick your hours

Be honest about when humans are online. Write it down. Then set an away message for other hours. The WhatsApp Business app includes greeting and away messages for this. Business API tools can do the same through automation.

Step 2: Sort chats by priority

Not every chat is equal. A payment failure is more urgent than a product question. Here is a sample you can adapt:

Sample WhatsApp support SLA by priority: first human reply and resolution goal
Priority Example issue First human reply Resolution goal
High Payment failed, order stuck 2 minutes Same day
Medium Delivery update, refund status 5 minutes Within 24 hours
Low Product question, feedback 15 minutes Within 24 hours

Note: This is a sample template, not an industry standard. Change the numbers to fit your team, hours, and issue types.

 

Step 3: Decide when the clock stops

This is where many SLAs break. Write clear rules for:

  • Waiting on the customer: pause the clock.
  • Waiting on another team (warehouse, finance): track it as its own status.
  • Outside working hours: do not count it against the team.
  • Reopened chats: decide if they start a new clock.

YCloud’s guide suggests measuring time awaiting agent, customer, specialist, or external system as separate pieces. That shows where the delay really sits.

Step 4: Set a safety buffer

Promise customers a little less than your internal target. If your team aims for 3 minutes, promise 5.

Step 5: Use automation for the first touch

A bot can answer in seconds, 24 hours a day. It can collect the order number and send the right help article. Then a human takes over with full context. That protects your first response time and saves agent effort.

The key is a clean handoff. The customer should never have to start over. Our full guide to running WhatsApp support shows how to combine bots, live agents, and team inbox rules so the SLA you write here is one you can actually keep.

Step 6: Build one simple dashboard

Keep it to 6 to 8 numbers. Show them by day, agent, and issue type. If a number does not lead to an action, remove it.

Common mistakes to avoid

  • Looking only at averages. One 3-hour wait hides inside a 4-minute average. Use median and 90th percentile.
  • Counting the auto-reply as a response. Split bot and human numbers.
  • Chasing speed alone. Fast but wrong answers raise effort and repeat chats. Pair speed with FCR and CSAT.
  • Forgetting the 24-hour window. A great SLA means little if the window closes before you resolve the chat.
  • Waiting for a weekly report. By the time you read it, the problem may have already cost you customers. Use live alerts for chats close to breaching.
  • Ignoring quality rating. Blocks and reports can shrink your reach.
  • Using metrics to police agents. Live numbers should help agents get support in the moment. They should not feel like surveillance.

How often to review your numbers

  • Live: chats close to breaching, backlog, bot handoff queue.
  • Daily: first response time, workload, quality rating.
  • Weekly: SLA compliance, resolution time, FCR, top issue types.
  • Monthly: CSAT, CES or NPS trends, SLA targets. Ask: do the targets still fit?

Start with the first three metrics: first response time, resolution time, and workload. Add one new metric each month. In three months, you will have a full picture without overload.

Frequently Asked Questions (FAQs)

  • What is a WhatsApp support SLA?

    A WhatsApp support SLA is a promise about how fast and how well you will help customers on WhatsApp. It sets targets for things like first response time and resolution time. It also says when the clock starts, pauses, and stops.

  • What are the most important WhatsApp support metrics?

    Start with first response time, resolution time, and agent workload. Then add first contact resolution, SLA compliance rate, CSAT, bot handoff rate, and WhatsApp quality rating.

  • What is a good first response time on WhatsApp?

    Customers expect fast answers. In a Nextiva survey of 400 consumers, the expected wait was one minute for chat and five minutes for SMS. Freshworks reports that its top-performing teams average 18 seconds on WhatsApp. A practical starting target is under five minutes in working hours.

  • Does a chatbot reply count as first response time?

    It depends on the tool and your policy. Nextiva, for example, does not count a plain auto-acknowledgement; the reply must address the question. The best practice is to track the first bot reply and the first human reply as two separate numbers.

  • How does the 24-hour window affect my SLA?W

    Inside the 24-hour window, you can send free-form replies. After it closes, you can only send approved templates. Each new customer message resets the 24 hours. So try to resolve chats inside the window.W

  • Are replies inside the 24-hour window still free?

    Not always. From 1 October 2026, Meta gives each business phone number 1,000 free service messages a month. After that, free-form replies in the window are charged per delivered message, at a rate that varies by country. Messages from customers stay free. Check Meta's pricing page for your market.

  • What is the WhatsApp quality rating?

    It is a rating Meta gives your business phone number, based on customer feedback such as blocks and spam reports. A low rating can lower your messaging limits. You can check it in WhatsApp Manager.

  • How often should I review WhatsApp support metrics?

    Watch live alerts all day. Check first response time and workload daily. Review compliance, resolution, and FCR weekly. Review CSAT and your SLA targets monthly.

Final thoughts

Good WhatsApp support is not only about being fast. It is about being fast, clear, and right, and knowing it with numbers.

Pick a few metrics. Write a short SLA. Set clear clock rules. Review weekly. Then improve one thing at a time.

Start small, stay consistent, and let your own data guide each change.