• Every day or two: If you have less than a minute and a half 
  • When they are strapped for time, our current users usually go through duplicates in today’s arrivals.
• Every day or two: If you only can spend two to three minutes
  • Our current users typically look for duplicates in today’s arrivals, then they would look at one or more of the following:
    • Arrivals in the next seven days
    • In house today
    • Departures for today
    • Reservations made in the last seven days
• Once a week: when you have five to 15 minutes 
  • When our current users have a bit more time, they would often also look at:
    • Arrivals in the next 30 days
    • Reservations made during the past 30 days

That said, you can tailor your efforts based on your business needs & primary focus: 

• If your focus is on streamlining the arrival process and identifying possible repeat visitors before they arrive, you'll likely want to focus on near-term upcoming arrivals
• If your focus is on optimizing delivery of pre-arrival emails sent a week in advance to ensure guests have the optimal stay, you'll likely want to focus on arrivals a week ahead 
• If your focus is on offering promos and incentives to return to the hotel to past guests, you'll likely want to focus on last week's or last month's checked out guests   

Who should clean up duplicates?

Depending on your team’s needs and environment, you may find some commonality with one of these approaches:

• Some have given the responsibility to various levels of hotel management (usually GM, AM, GSM, FOM)
• Others have incorporated this into their reservation process, giving reservation and front desk agents the responsibility to merge guest profiles
• Some of our customers have seen this as a great fit for night audit

When should duplicate cleanup happen? 

We’ve seen and heard about some patterns for when people are doing this:

• First thing in the day/shift
• When there is typically a bit of daily breathing time (in some properties, this seems to be between check out and check in time)
• At night
• A less busy day (for weekly processes)

When to use “Show exact matches only” 

Use the exact matching rule set if:

• you want to be sure suggestions are rigorous/strict with little tolerance for data entry errors or misspellings
• you want to limit the chances of us suggesting possible matches that aren’t actually the same person

You may miss out on some records that are actual matches, but you can be more certain the suggestions are actually the same person.

When to use “Allow up to one misspelling” or “Allow up to two misspellings”

Use one of these fuzzy matching rule sets if:

• you want to find exact matches but you also want to find records that include data entry errors or misspellings in the name and/or key contact information.
• you don’t mind waiting a bit (it takes a while longer to run through the larger set of rules and run the fuzzy matching algorithm on the data).
• you don’t mind if we suggest some records that may not actually be the same guest.

If you see suggestions that are pretty clearly not the same person, we recommend you go to View Detail screen and mark the record as not a match. When you do that, we won’t suggest the records as a possible match for each other in the future.

In general, we recommend using Profile Match & Merge as a daily or weekly process, progressively broadening searches by date as time permits. 

When there is a large set of guest data for the system to analyze, we'll show the results as they come in (instead of waiting until all results are analyzed) so you can work though results as more are loading. We'll show progress of analyzing the guest data, e.g., "Looking for matches: 12/230 guests" at the top of the screen so you get a sense of overall progress.

In special cases, you may want to do a larger clean-up of duplicate data as a special project. 

Upcoming arrivals

One example was a customer who was planning their 2018 loyalty and perks program and wanted to clean up all arrivals for 2018 in one fell swoop so they would have a clear picture of guests who were returning in 2018. Our user chose arrivals and set a custom date range for January, dealt with those records, then moved on month by month through the year. It took her one morning to run through all the arrivals for 2018 (for a 45 room upscale property that focuses on repeat guests that has approximately nine years of guest data in RoomKeyPMS).

Historical data

If you want to focus instead on cleaning up historical data, we suggest using "Checked out" or "Reservations made" combined with date ranges in the past. We suggest starting with smaller date ranges, e.g., Last seven days, then moving to larger date ranges, e.g., Last 30 days or a custom date range of a month/quarter/year. Continue to expand the custom date ranges as you go until you've covered all historical data.