Telecommunications

    Coding churn reasons and NPS comments in telecommunications

    An operator gets thousands of comments a month. Survey Coder Pro codes them with the same codebook every month, so you know whether "signal" or "billing" really went up.

    Try FreeContact Sales

    What makes open-ends hard at an operator

    High volume, continuous surveys, and technical vocabulary customers use in their own way.

    Transactional NPS after every call or technician visit produces thousands of comments a month that nobody has time to read.

    "No signal", "the internet keeps dropping" and "the wifi doesn't reach the bedroom" sound alike but are different problems for the network team.

    The port-out reason almost always combines price and service, and a single code per answer loses what explains the exit.

    If every month is coded with different criteria, the committee sees movements caused by the method, not by customers.

    How a telecom open-end works in Survey Coder Pro

    You generate the codebook from the first wave or load the existing one, and it is reused every month. The AI codes each comment with every reason it mentions and leaves only the doubtful ones for review.

    Nets by service (mobile, home broadband, TV) and by reason (network, billing, service, price).

    A codebook that stays stable month to month; new reasons are proposed and your team decides whether they go in.

    Codes as columns per respondent to cross with plan, region or tenure.

    With the .sav, a tabulation book with region or customer type in the banner and significance testing.

    Telecom Use Cases

    Applications for telecommunications operators

    Churn Analysis

    Identify true abandonment reasons and prioritize retention actions.

    Operator NPS

    Code promoter and detractor feedback by touchpoint and segment.

    Technical Support

    Analyze post-call feedback and technical service interactions.

    Plan Evaluation

    Feedback on new offers, plan changes and promotions.

    Onboarding

    Line activation experience and first 90 days.

    5G Adoption

    Perception studies, barriers and 5G migration drivers.

    Illustrative example

    Example: reasons for porting out to another operator

    Survey of mobile customers who requested to port their number to another operator. Here is part of the dataset after coding.

    Open-ended question

    What is the main reason you decided to switch operators?

    Responses and assigned codes

    ResponseCodes
    “There's no signal inside my house, I have to go out to the yard to make a call.”
    Indoor coverage
    “The other carrier offered me twice the data for less money.”
    Better competitor offerPlan price
    “They raised my rate without telling me and when I called they said it was the annual adjustment.”
    Price increaseNo advance notice
    “I called three times about roaming and they never fixed it.”
    Issue not resolvedPhone support
    “Data is painfully slow in the afternoon, I can't even watch a video.”
    Data speed
    “They billed me for a content bundle I never asked for.”
    Unrecognized charges
    “I switched because my whole family is on the other network and the family plan is cheaper.”
    Family / group planPlan price

    Codebook excerpt

    Net / code% of respondents who answered
    NET Price and offer46%
    Plan price27%
    Better competitor offer21%
    Price increase12%
    Family / group plan6%
    NET Network33%
    Indoor coverage18%
    Data speed17%
    NET Billing15%
    Unrecognized charges10%
    No advance notice7%
    NET Customer service14%
    Issue not resolved11%
    Phone support8%

    Multi-code question: one response can carry several codes, so percentages add up to more than 100%. A net counts each respondent once, even if they have several codes from that group.

    Example built for this page: this is not client data and the percentages are illustrative.

    For market research agencies

    • Monthly NPS and churn trackers: each wave inherits the codebook from the previous one.
    • You deliver the .sav or Excel file to the operator with codes as columns per respondent.
    • A tabulation book with region, plan or customer type in the banner and significance testing.
    • Your team reviews only the doubtful answers, which keeps a high monthly volume manageable.

    For CX and insights teams at operators

    • Transactional NPS comments (call, technician visit, store) get coded every month without an analyst reading them by hand.
    • The same codebook month after month keeps the series comparable: the dashboard shows changes in customers, not in method.
    • Export to Excel or .sav and cross it in your BI tool with region, plan or tenure.
    • Nets separate network, billing and customer service, so each team gets what belongs to it.

    Frequently asked questions

    Can it handle the volume of a monthly transactional NPS?

    Yes. Each month you upload the new dataset, it is coded with the existing codebook, and your team reviews the queue of doubtful answers, not every comment.

    How does it tell a coverage problem from a speed problem?

    The codebook has separate codes for each, and the AI assigns the one that matches what the customer says. If a comment doesn't make it clear, it is flagged as doubtful.

    What happens when a new reason appears, such as a competitor promotion?

    It is proposed as a new code. Your team decides whether it goes into the codebook; from then on it applies to later waves and the earlier series is left untouched.

    Does it work for mobile and home broadband in the same survey?

    Yes. The usual setup is one net per service with reasons inside each, or a service variable in the tabulation banner.

    Try the workflow on one of your own open-ended questions

    Upload a dataset with an open-ended question and see the codebook the AI proposes. The free trial includes 250 responses.

    Try FreeContact Sales

    Related resources