How to Measure WhatsApp Campaign Performance: A KPI Guide
To determine whether a campaign is successful, you first need to decide what you consider success. Most businesses look at the numbers after a campaign and say, “Not bad”; but when no goal has been set in advance, that statement means nothing. This article focuses more on setting up a measurement system than on reading dashboard reports: which metrics to choose, how to calculate them, and how to test them. For an explanation of the screens on the dashboard, see our article on campaign analysis.
We’re not providing industry averages here. Such metrics vary widely depending on the source and may not represent your audience. What you should really be comparing is your own past campaigns.
Write Down Your Goal First
Before every campaign, write down a one-sentence note: “With this campaign, I want X people to do Y, and I’ll measure this using Z.” Examples:
- I want a portion of customers who haven’t visited in the last three months to book an appointment; I’ll measure this by the number of appointments.
- I want to identify people interested in the new product; I’ll count those who receive the “interested” tag.
- I want to understand customer satisfaction; I’ll collect survey responses.
Which Metrics Should You Track?
| Metric | Formula | What does it indicate? |
|---|---|---|
| Delivery Rate | Delivered / Sent | List and Link Quality |
| Open Rate | Opened / delivered | Recognition and Timing |
| Response rate | Respondents / Sent | Message appeal |
| Positive response rate | Interested individuals / Respondents | Alignment between the offer and the audience |
| Conversion rate | Sales / messages sent | Whether the campaign was effective |
| Bounce rate | "UNSUBSCRIBE" / Sent | Audience frustration |
| Invalid number rate | Not on WhatsApp / total target | List cleanup |
Let’s calculate this with an example. The numbers below are for illustrative purposes only; they are not real data: Messages were sent to 200 people, 20 people responded, 5 people made a purchase, and 2 people replied with “STOP.” The response rate is 10%, the conversion rate is 2.5%, and the opt-out rate is 1%.
Select a key metric
Determine a primary metric for each campaign. For appointment reminders, the number of confirmations is key; for sales campaigns, sales are key; and for customer re-engagement, responses are key. The others are supplementary information.
Add a protective metric
Churn rate and complaints serve as protective metrics. Even if sales are increasing, a rapid rise in churn requests indicates you’re exhausting your list. These numbers are also critical for account health.
Calculating Return on Investment
Campaign costs aren’t just software fees. Consider the following items:
- The portion of panel or message costs allocated to the campaign.
- The amount of the discount offered.
- The time spent by staff writing and responding to messages.
- Preparation of visuals and content.
Here’s a simple calculation: (Kampanyadan gelen kâr - Kampanya maliyeti) / Kampanya maliyeti. Using profit instead of revenue yields more accurate results.
Linking Sales to the Campaign
If a WhatsApp sale takes place in-store or over the phone, it becomes difficult to link it to the campaign. Some methods:
- Campaign-specific code: Track sales using the code included in the message.
- Special keyword: Phrases like “Text KIŞ” filter out responses.
- Manual tagging: In the WapFlow sales funnel, manually mark respondents as sales.
- Asking a question: The question “Where did you hear about us?” still works for small businesses.
If multiple channels are running simultaneously, it’s hard to pinpoint the single cause of a sale. Don’t treat the results as definitive proof—view them as a guide.
Testing: Single Variable
To determine whether something works, change only one thing at a time.
- Divide your audience into two similar groups.
- Change only one variable: the opening line, the send time, or the offer.
- Run the two campaigns during the same time period.
- Compare the key metric.
- Use the winner in your next campaign.
In the guide, you can create two separate groups and launch two campaigns, then compare the “Selected” items in the sales funnel side by side. A difference of one or two people in small groups could be a coincidence. Don’t draw conclusions from insignificant differences.
Monthly Tracking Table
A simple table displayed outside the dashboard shows the trend between campaigns. The following columns are sufficient:
| Date | Campaign | Audience | Goal | Sent | Reply | Sales | STOP | Note |
|---|---|---|---|---|---|---|---|---|
| Example | Winter Products | Past Buyers | Sales | ... | ... | ... | ... | Unlinked text tested |
The "Note" column is the most valuable one. It helps you remember what each test yielded a few months later. You can download the sales funnel report as a PDF and add it to these records.
Factors That Skew Measurements
- People with "read" notifications turned off: The read rate will appear lower than it actually is.
- People with their phones turned off: Delivery may be delayed; do not measure too early.
- Negative responses: “Don’t write” is also a response; it increases the response rate. Check the tags.
- Response time: The response rate in WapFlow counts responses received within 7 days; those received later are not included in this count.
- Automatic replies: If the recipient’s business account sends automated messages, this does not indicate genuine interest.
- Special periods: Do not compare holiday or sale weeks to a normal week.
Making Decisions Based on Numbers
The purpose of measurement is not to generate reports, but to make decisions. If the response rate is low, change the message. If the bounce rate is high, narrow your audience. If the response is good but sales are weak, review your response process. We’ve summarized where to look for each indicator in the table in the analysis section. You can refer to the campaign guide to improve your copy or use templates for pre-written text.
Measuring Response Quality
Numbers don’t tell the whole story. Once a month, read a few random responses from your latest campaign and ask yourself these questions: Did the customer understand the message? What are they asking? Which objections are recurring? These reviews often provide the best ideas for your next campaign. Automatic tagging is also reviewed during this process; if you see misclassified responses, correct the keywords.
Measure response time as well. How long did it take for the person to get back to you after they replied? Comparing periods with quick responses to those with slow responses within the same campaign helps you determine whether the issue on the sales side lies with the copy or the team.
Customer Value
Instead of focusing solely on one-time sales, look at the same person’s subsequent purchases as well. A customer who places a small first order can become valuable over time. You’ll need to use your own sales records for this information, as the dashboard isn’t connected to your payment or order system.
Three Questions to Ask in a Meeting
Hold a brief monthly review and answer just three questions: Which campaign performed best this month, and why? Where did the most exit requests come from? If we were to test just one thing next month, what would it be? Record the answers in a tracking table. This brief habit prevents reports from gathering dust in a file.
Frequently Asked Questions
What is a good response rate?
There’s no universal number. Calculate the average for your own campaigns and aim to exceed it.
When should I review the results?
Check for delivery and errors on the first day. Wait a few days for responses and sales.
Does tracking work for small lists?
Yes, but focus on the people, not the percentages. Reading responses one by one on a small list is more instructive.
Are automated reports enough?
No, not if you don’t track sales. You’ll need to complete the final step yourself to see the campaign’s financial results.
