Google reviews to Tableau
Your business already reads Tableau, so a separate reviews dashboard is a link that gets opened twice. Reviewflowz publishes an OData feed, Tableau’s built-in OData connector reads it, and every Google review arrives as an ordinary row with the site it was left at attached.
Set up in under five minutes • No demo required
Google reviews 7,928 across 14 sites 
Tableau Rating by location
Every location row arrives with a latitude and a longitude, so this is a drag and drop.
Google reviews, into the Reviewflowz OData feed, into a Tableau extract.
Location is a column, and so is its latitude
A business with fourteen sites has fourteen reputations, not one. Google files every review under the site it was left at, and Reviewflowz writes that site onto every row. The location name sits next to the rating, the date, the language, the tags and the text, so slicing by site is a click, not a project.
Locations is its own table in the feed, not a string you clean up later. It carries the name, the city, the country, the postcode, the latitude and the longitude. Tableau reads those two as geographic roles on sight, which is why rating by site on a map is a drag and drop rather than a geocoding project.
That is the difference between a chart and an argument. Thirteen sites between 4.33 and 4.79 and one at 3.62 is not a mood. It is an address, and somebody can be standing in it on Saturday.
And it is not one table. Reviews sits beside Locations, your review profiles, a ratings history and a change stream, so one workbook can chart the rating over eighteen months and hold the reviews that moved it.
Google reviews Data Reviewflowz OData feed Locations
14 rowsEvery Reviews row names its site
Reviews
7,928 rows trimmed to eight of twentyLatitude and Longitude arrive as numbers, not as a place name Tableau has to guess at, so the geographic roles are right before the map is drawn.
Paste one URL. There is nothing to install.
Reviewflowz publishes an OData v4 feed, and Tableau has had an OData connector built in for years. Connect, OData, paste the service URL, and the tables appear. No add-on to approve, no web data connector to host, no Python script on an analyst’s laptop.
The credentials are one token you create inside Reviewflowz. Choose Username and Password, paste the token into the user name field, leave the password blank. That is the whole authentication story, and it ends when you revoke the token.
Tableau builds an extract from an OData source, which is the right shape here anyway. Eight thousand Google reviews is nothing next to a transactions table, so it finishes in seconds, and the published workbook opens instantly for whoever only came to look.
The table names are stable, and that matters because a saved workbook binds to them: Reviews, ReviewProfiles, Locations, RatingsHistory, ReviewSummary and ReviewChanges. The sheet you built in March still works in November, a low bar most exports fail.
Tableau
Google reviews via Reviewflowz
The connection produces an extract. Tableau Desktop supports extracts for OData.
The league table leadership will actually open
Rating by site, sorted, with the chain average drawn through it as a reference line. It takes about a minute to build and it changes the meeting, because a ranked list with a line through it does not need a narrator standing next to it.
Thirteen sites clustered between 4.33 and 4.79 and one sitting at 3.62 ends an argument rather than starting one. The question stops being whether Bristol feels worse and becomes what happened at Bristol, which is a question with an owner.
Put it in the workbook leadership already opens. A standalone reviews dashboard is a link in a Slack message: opened once out of politeness, once out of curiosity, then never. A sheet in the operations workbook is in front of them every Monday whether or not they came for the reviews.
And it stays true without maintenance. Every refresh feeds the same sheet, so the quarterly site review is that page with a different date range selected, not an export somebody reformats the night before.
Google reviews
Tableau
Google reviews, Reviewflowz OData feed, 7,928 rows. Axis starts at 3.0, not 0.
Volume tracks footfall, not a release calendar
Google review volume follows people through the door. It roughly doubles at the weekend, it jumps the week a campaign runs, and it falls off the week the car park is dug up. A monthly average flattens all three into one number that moves by 0.02 and tells nobody anything.
Reviews per day on a column chart, with that flat average laid over the top, is the whole point in one visual. The Saturdays stand a head above the Tuesdays, and the week the autumn menu launched stands above both. Nobody built a dashboard for that. Somebody added one sheet.
Which turns volume into a signal rather than a vanity count. A site whose weekend volume drops while its neighbours hold is a site where something changed, and you see that on a chart in September instead of hearing it in a district manager’s report in November.
The review date arrives as a real date on every row, so a Tableau date filter, a week-of-year grouping and a running total behave the way they do on any other table.
Google reviews
Pennyroyal Bakehouse, 14 sites
Tableau Google reviews per day
27 Jul to 6 Sep 2026 Sat, Sun Mon to FriSaturdays average 64 Google reviews a day and Tuesdays 27. The average of 38 sits between them and describes neither.
Read in Tableau Desktop from an extract of the Reviewflowz OData feed.
The map says where. The blend says why.
A rating by site is a symptom. The workbook already holds the causes: revenue per site, transactions per hour, weekend hours on the rota. Put the rating on one axis and the staffing on the other and the outlier stops being mysterious within about ten seconds of looking at it.
A site at 3.62 running the thinnest weekend cover in the chain is not a coincidence that needs a consultant to interpret. Sixty-eight of its ninety-four one and two star reviews in the last ninety days mention the queue. The dot on the scatter and the sentence a customer wrote are the same fact at two zoom levels.
Reviewflowz does not hold your rota and does not want it. The blend happens in Tableau, against the sources already in the workbook, because that is where the operational data lives and where the person who reads it already works. Nothing has to be exported to a third place to be joined.
That is the case for putting reviews in Tableau rather than in a reviews tool’s own dashboard. A reviews dashboard can only ever show you reviews.
Google reviews 7,928 reviews · 14 sites
Tableau
Google rating vs weekend cover, by site
68 of the 94 one and two star Google reviews Bristol Harbourside took in the last 90 days mention the queue or the wait. Its weekend cover runs 18 per cent below the chain.
Google reviews arrive as a Reviewflowz OData extract. Staffing hours come from the rota export already in this workbook. Tableau blends them on site name.
One token per workbook, and you can take it back
A connection is a token: created in Reviewflowz, revoked in Reviewflowz. Give the agency running your autumn campaign one of its own, end the engagement, revoke it, and their workbook stops refreshing that afternoon. No shared password, no spreadsheet you cannot un-send.
A token carries the access of the person who created it. If your South West manager only sees four sites, so does every workbook built on their token. The permissions are the ones you already set up, not a second set that drifts away from them over a year.
Revoking is a state, not a deletion. The connection stays in the list with the date it was revoked and the date it was last used, so a question in December about August has an answer.
Setup is the same at four sites or forty. Connect the Google Business Profile, create the connection, paste the URL into Tableau. Under five minutes, with no call with a salesperson standing in front of it.
Google reviews Pennyroyal Bakehouse, 14 sites Tableau BI connections
3 active, 1 revokedOne token per connection, created and revoked here.
Group Tableau workbook
Created by Nadia R., group ops · Last used 7 Sep, 06:15
7 Sep 2026, 06:15
South West region
Created by Tom H., regional manager · Last used 6 Sep, 21:40
6 Sep 2026, 21:40
Agency, autumn campaign
Created by Priya S., marketing · Never used
Never used
Q2 board pack
Revoked 28 Aug 2026 by Nadia R. · Last used 5 Aug, 09:02
5 Aug 2026, 09:02
Audit trail kept
A token carries the access of the person who created it. A workbook built on Tom’s token can never show a site Tom cannot see in Reviewflowz.
The BI feed is on Premium. It is not part of the free trial.
Put the site league table in the workbook they already open
Set up in under five minutes. No demo required.
How do Google reviews actually get into Tableau?
Through an OData v4 feed that Reviewflowz publishes. You create a connection in Reviewflowz and copy its feed URL, then in Tableau choose Connect, OData, and paste it. Set Authentication to Username and Password, put the token in the user name field and leave the password blank. Tableau lists the tables and you drag in the ones you want. There is nothing to install: the OData connector ships with Tableau.
Extract or live connection?
Extract. Tableau Desktop supports extracts only for OData sources, so that is the connection you get, and it is the right one here anyway. Review volume is small next to your transactions table, the extract builds in seconds, and a published workbook backed by an extract opens instantly for the people who only came to look at it.
Which columns arrive?
One row per review with the rating, the review date, the location name, the review profile, the reviewer name, the title, the full text, the language, the tags, the topics and the owner reply with its date. Alongside it sits a Locations table with the name, city, country, postcode, primary category and the latitude and longitude, plus your review profiles, your ratings history, a review summary and a change stream of inserts, updates and deletes.
Do we have to geocode the sites to get a map?
No. Locations carries a real latitude and longitude per site, so Tableau assigns the geographic roles itself and the map draws from the numbers rather than from a name it has to guess at. That is the difference between a map you build on Tuesday afternoon and a mapping project you scope.
How does the workbook stay current?
You refresh the extract, and how that is scheduled is a setting in your own Tableau deployment rather than something we run. On the Reviewflowz side each Google Business Profile is checked about every couple of hours, so the feed already has the new reviews by the time Tableau asks for them.
What happens when a review is edited or removed?
It reaches the feed. A reviewer who edits a one star into a four star after you fixed the thing they complained about, and a review Google takes down, both arrive as events on the change stream rather than as a silence, so the numbers in the workbook do not quietly drift away from what is on the listing.
Does the first load include our Google review history?
Yes. Connecting the Google Business Profile loads its review history, so the trend line is full from the first extract rather than filling up over the next quarter. Bear in mind that volume tracks footfall, so the history spikes at weekends and after campaigns rather than on a release schedule.
Which plan includes the BI feed?
The BI feed is a Premium feature and it is not part of the free trial, so budget for that rather than finding out on day three. Everything else works the way the trial suggests: you connect your Google Business Profile yourself and see your reviews in minutes.
Do I need a demo to get started?
No. You connect your Google Business Profile and create the BI connection yourself, in under five minutes. If you get stuck, support is the people who build it.