Marketing

This page turns Ascent Healthcare Texas's website, Google Search and website-form activity into a few numbers and trends, so we can see whether more of the right people are finding and contacting us. It contains no names or personal details, only counts.

What the numbers mean

Website sessions

One visit to the AHT website. A person who visits on Monday and again on Friday counts as two sessions. A visit ends after about 30 minutes of inactivity, so someone who comes back after lunch can count twice. This is the main "is anyone coming?" number.

Page views

How many pages were opened in total. A visitor who reads three pages adds three. If page views grow faster than sessions, people are looking around more.

Search impressions

How many times an AHT page was shown in Google results, whether or not anyone clicked. It measures how visible we are.

Search clicks

How many times someone clicked an AHT result in Google. Clicks divided by impressions is the click-through rate: how often being seen turns into a visit.

Form submissions

How many times someone filled in the "Website Contact Us" form. This includes spam. Only the count is kept here; what people typed is never copied to this page.

Change arrow (▲ / ▼ and a percentage)

Compares the period you picked with the same number of days just before it ("+21% vs previous 28 days"). It shows direction, not importance: with small numbers a handful of visits can move the percentage a lot.

Daily charts

One point per day. The dot and number at the right edge are the newest complete day. Hover over a chart (or click it and use the left and right arrow keys) to read any day's value. "Table" shows the same numbers as a list.

Channels

How visitors arrived. Direct: typed the address, used a bookmark, or came from somewhere that doesn't say where from (many email apps and chat tools). Organic Search: a Google or Bing result. Organic Social: a post or profile on LinkedIn or another network. Referral: a link on another website. AI Assistant: a link from an AI tool. Other channels appear when they occur.

Countries

Where visitors are, as "user-days": each person is counted once per day they visit, so someone who visits on three days counts three times. Unexpected countries are often automated traffic (search engines and scanners), not people.

Top pages

The most-viewed pages. "/" is the home page. After publishing new content, look here (and at the Organic Social channel) over the next few days.

Search clicks by domain

Which of our web addresses people clicked from Google. Our website is ascentoutpatient.com. If more addresses are ever added and forward to it, a click on any of them lands on the same website, but Google reports each address separately.

Data-freshness line (under the title)

✓ up to date, with the last day it covers. ! more than five days behind. ✕ the last automatic update failed (the numbers shown are older). ○ not connected yet (LinkedIn today).

How to use them

• Look at weeks, not days. At our current size, one day can swing a lot because of a single visitor or a bot. Use the 28-day view and the arrows to judge direction.
• Are more people finding us? Watch sessions and search impressions. Rising impressions with flat clicks means we're being seen but not chosen: look at page titles and descriptions.
• Is it the right kind of traffic? A healthy mix has growing Organic Search and Organic Social, not just Direct. A sudden spike that is all Direct, from one country, on the home page only, is usually automated.
• Are visitors reading? Compare page views to sessions and check Top pages. If people land on the home page and never open anything else, the next step may not be obvious.
• Are visitors contacting us? Form submissions are the end of the trail. Compare them with sessions over the same weeks, and remember some are spam.
• Checking a campaign or post. Note the date, then compare the daily charts and the channels before and after it.
• Our own use. Visits from the team count too (mostly under Direct), so expect some baseline that isn't customers.

Things to know before you rely on a number

• Today is left out on purpose: analytics data for the current day is incomplete. The newest day is yesterday. Search figures run about three days behind Google's own pages, and recent days can change slightly as Google finishes processing them.
• Days follow UTC (Greenwich) time, so a late evening in the US can fall on the next day's number.
• Website figures and Search figures count differently. Visits to ascentoutpatient.com, and to any address that forwards to it, are one website, counted together in the website numbers (sessions, page views, channels). Google Search lists each address on its own, because Google can show a result for any of them. Don't add the Search numbers of different addresses to the website numbers as if they were separate sites.
• LinkedIn will appear here once LinkedIn approves AHT's access. It will show page-level totals only (followers, views, post reactions); LinkedIn does not allow person-level details or matching people to our contact records.
• The page shows marketing activity only, not sales pipeline or deals; those stay in HubSpot.

Ask Claude about the numbers

With the Marketing connector added in Claude (ask your AHT administrator for access; the address is at the bottom of this page), you can ask in plain language and Claude reads the same numbers as this page, for example:

• "How did website traffic change over the last 30 days?"
• "Which pages were viewed most this week, and which channels sent the most visits this month?"
• "Did search impressions or clicks change after we published the new article on the 14th?"
• "Compare form submissions to website sessions for the last two months."
• "Were there any unusual days in the last month?"

Claude can also refresh the data on request (for people with that permission), and can report connector usage and the simulated bill the same way. It can't see names or personal details of website visitors, LinkedIn members or contacts, because none are stored (the connector usage section names staff by sign-in name only if the owner has switched per-person figures on), and it can describe what the numbers do but not why they moved, so treat its explanations as suggestions to check.