Bulk SMS, WhatsApp, RCS, voice and email for businesses across India [email protected] Baner, Pune

How to Measure Campaign Communication Delivery Without Overclaiming Impact

By SMSNiti Team · · 9 min read

Measure campaign communication by what the records can actually show: messages or calls sent, delivery or answer status reported by the provider, key presses or replies where captured, and issues logged. None of these proves that a message changed anyone's opinion or vote. Define each metric before the first send and report it with its limits.

After a send, teams want a number that says it worked. The honest answer is that campaign communication reporting can show delivery and response, and cannot show persuasion. This guide explains what each common metric means, what it cannot prove, how to reconcile reports with your own checks and how to write a report that stands up to questions. It is part of our wider guide to political campaign marketing in India.

By Rupali Dahiwal. Last reviewed: 25 October 2026. This is operational guidance, not legal advice. No benchmark rates appear here, because any rate you publish should come from your own logs.

Start with the question the report must answer

Different readers need different answers. The operations lead asks whether the right version went to the right list at the right time. The finance lead asks what was spent and what was delivered. The campaign manager asks whether the plan worked. A single report cannot answer all three unless you decide in advance which questions it covers.

Write the questions down before the send. For example: did the approved message go to the intended list within the window, what share of messages the provider reports as delivered, what did recipients do where an action was captured, and what went wrong. Then pick the metrics that answer those questions and nothing more.

The metrics and what each one means

Sent or submitted

The number of messages or calls your campaign submitted to the provider. It shows what you asked for, not what happened. It is the base for every other figure.

Delivered (provider-reported)

For SMS, the number the provider reports as delivered to the recipient's handset or network. SMSNiti's bulk SMS page lists delivery status for every number. Delivery depends on the number, the network and the handset, so a delivered status means the network confirmed receipt, not that anyone read the message.

Failed or not delivered

Messages the provider could not deliver. Look at the reason where one is given, such as an invalid or unreachable number. A high failure share often points to a list quality problem, not a content problem.

Answered (voice)

Calls picked up. SMSNiti's voice page says call reports show who answered, how long they listened and which key they pressed. An answered call is not the same as a call heard to the end, which is why listening time is useful.

Listened for a stated duration (voice)

The time a recipient stayed on the call. Use the provider's definition. Do not assume that a long call means the message was understood or agreed with.

Responded

Key presses, replies or link clicks, where your design captures them. These show action, which is more informative than delivery. They still say nothing about opinion.

Issues

Complaints, wrong versions, wrong numbers, corrections and escalations. Track them with time, cause and fix. A short, honest issue log is one of the most valuable parts of a report.

What the numbers cannot prove

  • That a message was read or understood. A delivery receipt is a network event.

  • That it changed a view or vote. There is no way to infer this from delivery or key-press data.

  • That a response rate is good or bad. Without a baseline from your own earlier sends, a rate means little.

  • That one channel beat another. The messages, audiences and timing differ, so the comparison is rarely fair.

Never write that a message "reached" a number of voters unless you mean delivered to that number of handsets, and say so. Never present benchmark figures from other campaigns or vendors as expected results.

Reconcile the report with your own checks

Do not accept a report at face value. Compare it with what you can verify:

  1. Test handsets. Did the test messages and calls appear in the report with the statuses you saw on the phones?

  2. Counts. Does submitted equal the size of the list you approved, after suppression and duplicate removal?

  3. Version. Does the content shown in the report match the approved version number?

  4. Timing. Do the send times fall within the window you approved and outside any restricted period?

  5. Failures. Are failure reasons plausible, or do they suggest a list or setup fault?

If any of these do not match, investigate before you report. A report built on a wrong list or a wrong version is worse than a late report.

Build a simple reporting template

Use the same layout for every send so reports can be compared:

  1. Campaign, channel, message or call type and version number.

  2. Approved window and actual send times.

  3. List name, source and size after suppression.

  4. Submitted, delivered or answered, failed, with the provider's definitions.

  5. Responses captured, with the design that captured them.

  6. Issues, with cause, time and fix.

  7. Cost for the send.

  8. Notes on anything unusual and what you would change next time.

Ask the provider for definitions of every status before the campaign, in writing. Our communication brief checklist includes a reporting line for exactly this purpose.

Reading failures without blaming the content

When messages fail, work through causes in order:

  • List problems: invalid, disconnected or wrongly formatted numbers.

  • Setup problems: template or sender mismatch, unregistered link or number.

  • Timing problems: sending outside permitted windows.

  • Network or handset problems: outside your control, but visible in patterns.

Fix list quality first, because it improves every later send. Our bulk SMS setup checklist covers the test steps that catch setup problems before release.

Reporting for voice campaigns

Voice reports give more detail per call but need the same caution. Separate attempted calls from answered calls and from calls that played to a stated point. If your campaign captured key presses, report each option separately and say what the option meant. Do not report a retry as a new contact unless the provider counts it that way. For how to plan the call flow that produces these reports, see how to plan a bulk voice campaign.

A hypothetical report summary

This is an illustration of format, not real data. A short summary for one send could read: "Venue-change notice, version 3, sent in the approved window to the registered-attendee list after suppression. The provider reports the delivered and failed counts, and the failures are mostly invalid numbers. One recipient reported a wrong venue name, which was traced to a merge field and fixed before the second batch. No claim is made about opinion or attendance." The figures would come from the provider's report, and the summary would attach the report and the approved version.

Hold a short review after each send

A report is most useful when someone acts on it. After each send, hold a short review with the content owner, the channel operator and the compliance contact. Walk through four questions: did the approved version go to the intended list within the window, what do the delivery and response records show, what issues arose and what will change before the next send. Write the answers in the report and name an owner for each change.

This review is also where you decide whether to repeat, adjust or stop a type of message. Base that decision on what the records support. If delivery was poor because the list was stale, clean the list. If recipients pressed a key you did not plan for, check whether the script was unclear. If complaints were concentrated in one district or language, look at the version that went there. Avoid reading more into the data than it contains.

Keep reporting separate from persuasion claims

It is tempting to turn a good delivery report into a story about influence. Resist that. A campaign’s internal reporting should describe operations accurately, so that decisions about lists, timing and wording rest on facts. Claims about persuasion or turnout need evidence of a different kind, and delivery logs do not supply it. Keeping the two apart protects the campaign from overclaiming, and it protects the team from being measured against numbers that the data cannot support.

Keep the records

For each send, file the approved version, the approval record, the provider report, the issue log and the invoice together. These records support your own review and any expenditure documentation. The Election Commission's note of 20 March 2026 refers to Section 77(1) of the Representation of the People Act, 1951 and to statements of expenditure on internet and social media campaigning within 75 days of completion of the elections. Check what applies to your role and election.

Common mistakes

  • Claiming persuasion from delivery data. The data cannot show it.

  • No definitions. Terms like "delivered" mean different things across providers.

  • Reporting before reconciling. Check counts and versions first.

  • Quoting benchmarks from elsewhere. Use your own logs.

  • Hiding issues. A frank issue log builds trust.

  • Different formats every time. You cannot compare sends.

Frequently asked questions

What does a delivered status mean in an SMS report?

It means the network or provider reported the message as delivered. It does not show that the recipient read it.

Can I measure whether a campaign message changed votes?

Not from delivery, answer or key-press data. Do not claim it.

What should a voice campaign report include?

Calls attempted, answered and listened, key presses by option, issues and cost, using the provider's definitions.

How do I check that a report is accurate?

Reconcile it with test handsets, the approved list size, the version number and the send times.

Should I share reports publicly?

That is a decision for the campaign. If you do, share only what the records support, with definitions and limits.

Sources checked

  • Election Commission of India, Press Note No. ECI/PN/038/2026 dated 20 March 2026. Checked 24 October 2026.

  • SMSNiti bulk SMS service and bulk voice call service pages, for the report fields described. Checked 25 October 2026.

To see what reports look like for your campaign, ask SMSNiti about bulk SMS delivery reports and request the status definitions in writing.