Research
Internet research from real orders
Konecteaze may earn compensation when you call, click, or order through links on this page. How we make money
Call Call a Konecteaze advisor — (855) 503-7272Konecteaze advisor line.Each study includes the method, a downloadable CSV, and a short FAQ. Dates for the underlying records are in the method section of each page.
The Internet Providers People Actually Choose (in Our Order Data)
Viasat (Exede) was the most-ordered internet provider, at 48.5% of internet orders (n = 59,859). This is order share in this dataset, not market share.
How Long Does It Take to Sign Up for Internet by Phone?
A typical phone call that ended in an internet order took 19.2 minutes with an agent, not counting time on hold (median).
Internet Installation Times: Which Days and Windows You Can Expect
46.5% of scheduled internet installs were booked in a morning window (36,008 windows that could be read).
How these studies were made
Each study page lists its own measure, record dates, and limits. These rules apply to every study here.
Where the records come from
These studies use multi-year call and order records from our affiliated sales call center, where agents help shoppers compare and order home internet and TV from many providers.
Call records include the outcome an agent recorded for the call, how long the conversation lasted, and the caller’s state. A conversation is a call with at least 60 seconds of speaking time. Test calls, answering machines, robocalls, wrong numbers, and dead-air calls are removed. Duplicate records of the same call are counted once.
Order records include the provider, the plan or package, the sale date, the scheduled and actual installation dates, and the order’s final status.
Calls about products other than home internet or TV, such as electricity or gas plans, are left out. Orders whose provider could not be identified are left out.
De-identification
Before the figures were calculated, names, phone numbers, email addresses, street addresses, account numbers, and agent identities were removed or replaced with one-way tokens. No recordings or transcripts were used. The published result is aggregate statistics only.
Small-number rules
Any count from 1 to 10 is hidden, and the cell is left blank. A percentage or median is published only when it rests on at least 30 records. When hiding one cell would let a reader recover it by subtraction, a second cell is hidden too. Results are not broken out below the state level, and no study publishes a state table.
How the numbers are calculated
A share is the records in a category divided by all records in that group, rounded to one decimal place. Rounded shares in one group can add to slightly more or less than the whole. A median is the middle value. Where the group is large enough, the table also shows the 25th and 75th percentiles.
Limitations
These are people who called our affiliated sales call center to shop for service. They are not a random sample of United States households, and the results are not national market share.
Where callers came from depends on advertising and which providers were offered, so the mix of states and providers reflects where our affiliated sales call center advertised.
Outcomes are recorded by people and can contain mistakes. Provider names are the names used when the order was sold. Providers rename plans and brands over time.
Disclosure and reuse
Konecteaze may be paid by providers when you order through links or phone numbers on this site. Payment does not change these figures or how they were calculated.
You may reuse the CSV tables with attribution. The suggested credit is in “How to cite this study” on each study page. The tables are offered under the Creative Commons Attribution 4.0 license.
Dates for the underlying records are in each study’s methodology.
Published and last updated .