Illustrative training scenario. Names, cities, experience and figures in this story illustrate the method. This is not a verified case study or a confirmed result of an XO network member.

A repeat client doesn't want to start the conversation from scratch every time, but that doesn't mean a travel agent needs to keep a detailed dossier. A useful preference profile is a short, verifiable and customer-driven record of what really influences travel choices. It helps prepare a hypothesis faster, but never replaces a new brief.

The value of such a profile is not in the number of notes, but in the quality of signals: what is mandatory for a person, where he is ready to compromise, what he liked about the last trip and whether circumstances have changed. The agent stores only the necessary context, notes the source and date, separates fact from speculation, and asks for confirmation of relevance before each selection.

Illustrative scenario: Olga: memory has become a system, not an archive of correspondence

Consider this illustrative training scenario: Olga, a travel advisor from Poltava, has been working with repeat clients for seven years. She is warm in communication and remembers details well, but in the high season useful facts were lost among the voice messages: who doesn’t like night flights, who cares about the gentle entrance to the sea, and who is willing to live further from the center for the sake of silence.

After one long re-selection, Olga created a short profile in three parts: confirmed prerequisites, flexible preferences and conclusions from the last trip. She stopped copying personal conversations and recorded only what affected the service. A year later the family returned. Olga showed them five saved items and asked what had changed. Two were no longer relevant, one became mandatory. The first list of options was shorter, and clients didn't have to repeat the whole story.

The profile should not prove that I know everything about the client. It should help me ask the next precise question and not force the person to repeat what is important.

Olga — character in the illustrative training scenario

Define the purpose of the profile

  1. Reduce repetition of basic questions for returning clients.
  2. Don’t forget the mandatory restriction that already affected the trip.
  3. Formulate an initial search hypothesis faster, rather than automatically repeating the previous trip.
  4. Provide minimal work context to colleagues if the client is temporarily being handled by another agent.
  5. Show the client what data is being used and give the opportunity to correct or delete it.

If a field doesn't help you accomplish at least one of these tasks, it probably isn't needed. The profile should not turn into a chronicle of life, a set of evaluative labels, or hidden analytics. The fewer points that are precise and relevant, the easier it is to use and maintain trust.

Store only context that affects the trip

  • The composition of travelers and the roles in the decision are only in the current version.
  • Essential requirements: accessibility, diet, routine, documents or transport conditions, only to the extent needed.
  • Preferences for pace, leisure format, location, type of accommodation and departure time.
  • Acceptable compromises: what the client is willing to sacrifice for the sake of budget, dates or quality.
  • Actual reaction after the trip: what you liked, what didn’t work and why.
  • Preferred channel and convenient communication window, if the client has indicated them himself.
  • The date of confirmation of each important point and the source - the client’s words or observation of the result.

Don't write down unnecessary personal information just because it was mentioned in a conversation. Medical, financial and documentary data require special care and should be included in the workflow only when a specific service cannot be provided without them. In a preference profile, a functional formulation without details is usually sufficient.

Divide the profile into four layers

  • Essential: do not offer an option that does not meet this requirement.
  • Desirable: Adds value but allows for discussion.
  • Anti-preference: past experience that the client asks not to repeat.
  • Hypothesis: An agent's guess that needs to be tested by questioning rather than being used as a fact.

Add confidence status

  1. Confirmed Now - The client has directly agreed on an item in the current conversation.
  2. Confirmed earlier - the fact was important on the last trip, but requires double-checking.
  3. Conclusion after the trip - based on feedback and written in your own words.
  4. Hypothesis - noticed by the agent, not confirmed by the client and should not control filters.
  5. Obsolete - saved only for decision history or subject to deletion according to process rules.

Status protects against the most common mistake: turning one statement into a permanent rule. The client may have chosen an early flight because of the price, not because he likes to get up at night. Without a date or source, the note quickly becomes false personalization.

Collect better data after your trip

  • Which of the choices really made the trip more comfortable?
  • What turned out to be less important than expected?
  • What compromise would the client not want to repeat?
  • What should be saved as a must for the future?
  • Has the mix of travelers or the format of your next trip changed?
  • Is it possible to write these findings into a short profile for the next call?

I will save three points for the next selection: a quiet room is required, proximity to the center is desirable, night flight is not suitable. Before a new search, I will show you the recording and clarify whether it is still relevant.

Example of a transparent message to a client

Separate client words from interpretation

  • Fact: “the client said that he was not ready for an overnight flight with a child.”
  • Conclusion: “for a family trip, a night flight is a deal-breakers until reconfirmation.”
  • Dangerous label: “difficult client, always refuses cheap flights.”
  • Fact: “After the trip, the client praised the silence and breakfast, but did not use the pool.”
  • Hypothesis: “on the next city tour we can test whether silence is more important than hotel infrastructure.”

Evaluative characteristics such as “capricious”, “thrifty” or “loves luxury” do not help well in choosing a service and create the risk of biased communication. Rewrite them into observable conditions: budget limit, minimum room level, acceptable transfers, quiet requirements, or willingness to change dates.

Make your profile transparent and manageable

  • The client understands why preferences are saved.
  • Only the context necessary for the service is included in the profile.
  • Every significant item has a date and source.
  • The client may ask to see, correct or delete the entry.
  • Access is limited to employees who need the context for their work.
  • The profile does not contain copies of documents, payment information and unnecessary personal correspondence.
  • Storage and cleaning periods are in accordance with internal regulations and applicable requirements.

Don't promise lasting memory and don't hide it. The short phrase “save these preferences for next time?” creates the right expectation. If the client refuses, the usefulness of the ongoing consultation should not be diminished.

Profile template on one card

  • Renewal date and responsible agent.
  • The current composition of travelers is marked “to clarify” before a new trip.
  • Up to five mandatory conditions.
  • Up to five flexible preferences and their order of importance.
  • Two or three acceptable compromises.
  • Two anti-preferences with context when they emerged.
  • A short summary from my last trip.
  • Hypotheses in a separate block, no more than three.
  • Consent status, next update date and link to original dialogue if necessary.

Start a new brief with testing, not silent application.

  1. Show the client a short summary of the saved points.
  2. Ask what has changed in the travelling party, budget, dates and priorities.
  3. Remove or downgrade obsolete terms.
  4. Confirm the must-haves and rearrange the order of the desirables.
  5. Only then use the profile as a search filter.
  6. Commit the new version so that the team doesn't revert to the old one.

A good conversation starter sounds like this: “Last time, silence, a direct flight and a short transfer were important to you. How much of this remains relevant and what has changed?” This is caring without the illusion that the agent knows the person better than himself.

Don't turn personalization into pressure

  • Don't unexpectedly mention details that the client didn't expect to see saved.
  • Don't make sensitive assumptions about health, income, family or status.
  • Do not replace the choice with the phrase “you always take this option.”
  • Do not use your profile for mass mailings without clear reason and consent.
  • Don't hide alternatives just because they are different from your last trip.
  • Do not transfer the preferences of one member to the entire family or group.
  • Don't pass off the algorithm's or agent's hypothesis as the client's word.

Share context securely within the team

  • A colleague needs a summary, not a complete archive of correspondence.
  • Place mandatory conditions above wishes and hypotheses.
  • Indicate what has already been confirmed in the current request and what needs to be asked again.
  • Add the nearest next step and deadline so that the profile does not replace the live deal card.
  • Once the master agent is returned, keep one current version with no parallel notes.

Update and clean your profile

  1. After each trip, offer to confirm two or three findings.
  2. Before making a new selection, double-check the prerequisites.
  3. Once in a specified period, find cards without updating.
  4. Mark obsolete and remove what is no longer needed for the service.
  5. Merge duplicates and save the source of the last change.
  6. Follow the customer's request for correction or deletion according to the accepted process.

Measure benefit, not volume of data

  • How many repeat clients confirmed the profile before a new selection.
  • Has the time from the first message to an accurate list of options been reduced?
  • How many mandatory conditions were preserved and not lost during the transfer of the case.
  • How often a client corrects an outdated item is an indicator of the need for updating.
  • Has the number of irrelevant options in the first selection decreased?
  • How many cards have been cleared of unnecessary or outdated data.
  • Are there any complaints about the unexpected use of personal context.

Pilot for 14 days

  • Select ten repeat customers with recent travel history.
  • Create cards only from verified working facts.
  • Limit essential requirements and preferences to five items in each section.
  • Send the client a short summary and ask them to correct any inaccuracies.
  • Conduct one re-brief with an open profile check.
  • Compare the time and quality of the first list of options with the normal process.
  • Remove fields that no one uses and enforce access rules.

Final profile checklist

  • Each point affects the tourism service.
  • Facts, conclusions and hypotheses are separated.
  • There is a date, source and relevance status.
  • The client knows about the profile and can manage it.
  • Before a new selection, the data is rechecked.
  • Unnecessary personal and sensitive information is not stored.
  • The team measures accuracy and usefulness, not the number of fields completed.

A good preference profile is not about trying to remember a person forever. It's a neat bridge between two consultations: it preserves what's important, reveals the unknown honestly, and starts a new conversation with respect for past experiences. The more transparent and shorter such a card, the easier it is for the agent to be attentive without intrusiveness.