Marketing Dashboards in Singapore: A Practical Looker Studio (Now Data Studio) Guide

Most Singapore SMEs we meet have the same reporting setup: a PDF that arrives on the 5th of the month, twenty slides long, full of numbers nobody argues with because nobody quite trusts them. The owner skims the first page, looks for the lead count, and files it. Eleven months later the same person asks why marketing is not working, and nobody can answer from the reports, because the reports were built to be sent rather than used.

A marketing dashboard is supposed to fix that. Done properly, it is the one screen you open on a Monday that tells you whether last week beat the week before, and where the money went. Done badly — and most are — it is the same twenty slides with a refresh button, plus a new failure mode where the numbers quietly stop updating and nobody notices for six weeks.

This guide covers what the free Google tool can and cannot pull, the caching and quota limits that break dashboards without telling you, what the whole thing genuinely costs here, and the specific ways dashboards mislead. It assumes you have done the groundwork in conversion tracking and UTM tagging. A dashboard is a window onto your data; it cannot fix data that was never collected properly.

First, the name changed: Looker Studio is now Data Studio again

If you have been searching for Looker Studio tutorials and finding pages that call it something else, you are not going mad. On 16 April 2026, Google renamed Looker Studio back to Data Studio — reversing the October 2022 rebrand that renamed Data Studio to Looker Studio in the first place. Google’s own documentation now carries the line “Looker Studio is now called Data Studio” at the top of the product pages, and the April 2026 release notes record the rebrand alongside a refreshed home page.

The stated logic is separation. Data Studio is the free tool for self-service reporting and ad-hoc exploration. Looker remains the enterprise business-intelligence platform built on governed semantic models. Four years of people confusing a free dashboard tool with a six-figure BI licence is a reasonable thing to want to end.

Practically, almost nothing changes for you: existing reports, data sources, sharing links and permissions carried over, and the paid tier is now called Data Studio Pro. What does change is your search results. For roughly the next year, guides and connector documentation will be split across both names, and some of the “Looker Studio” material you land on describes a menu that has since moved. When a tutorial does not match your screen, check its date before concluding you have broken something. We use both names in this guide where it helps you find things.

What a dashboard is for — and the three questions it must answer

Before opening any tool, decide what the dashboard replaces. For most SMEs the honest answer is: a phone call. Someone wants to know how things are going and would rather not ask. That means a dashboard has exactly three jobs, and every chart should serve one:

• Is the trend up or down? Enquiries, sales, revenue — this month against last month and against the same month last year. Not the last 30 days against the previous 30 days, which in Singapore quietly compares a Chinese New Year fortnight against a normal one.
• Where did it come from? Which channels produced the enquiries. This is the page that justifies or kills next quarter’s budget.
• What is broken right now? A landing page that stopped converting, a campaign that spent its budget in four days, a form that has produced zero submissions since Tuesday.

Anything answering none of those three is decoration. The most common dashboard failure we see is not technical — it is a twelve-page report whose first page is a wall of sessions, users, bounce rate and pageviews, none of which anyone has made a decision with. Our guide to reading a marketing report covers telling a working metric from a decorative one.

What Data Studio connects to free — and the Meta problem

This catches people out, and it is worth understanding before you commit a weekend to a build. Google provides free, first-party connectors to its own products. It does not provide one for Meta, and never has. If your marketing runs on Meta — as most of it does for a great many Singapore B2C businesses — you are choosing between a paid third-party connector, a manual export, or a dashboard with a hole in it.

The workarounds, honestly ranked. Paid partner connector is the least painful: several vendors sell Meta connectors on a monthly subscription that scales with how many accounts you connect. Budget it as a recurring line, not a one-off. Manual export to Google Sheets is genuinely free and genuinely tedious — someone exports the Meta report weekly, pastes it into a Sheet, and the dashboard reads the Sheet. It works, and it fails the first week that person is on leave. BigQuery as a middle layer is the grown-up answer if you have developer help, and it also solves the quota problem discussed below.

The pattern we see most often in Singapore SMEs with one marketer: native Google connectors for everything Google, one paid connector for Meta, and a Google Sheet updated monthly for anything else — walk-ins, phone enquiries, WhatsApp conversations. That last Sheet matters more than it sounds, because a great deal of Singapore demand never touches a web form. Our guide to tracking calls and WhatsApp leads explains why, and why Google’s call-forwarding measurement is not available here at all.

The freshness trap: your dashboard is up to 12 hours stale

A fact that surprises almost everyone building their first dashboard: Data Studio does not query your sources every time someone opens the report. It caches. And for Google’s own marketing products the cache window is fixed at twelve hours and you cannot change it. Google’s documentation is explicit that Google marketing and measurement products — Google Ads, Google Analytics, Campaign Manager 360, Search Console, YouTube Analytics — refresh every 12 hours, and that rate cannot be changed. Other sources are configurable.

Two practical consequences. First, stop using a dashboard as a live campaign monitor. To check whether this morning’s campaign is spending, open Google Ads. The dashboard is a reporting layer, not an alerting layer. Second, put the refresh timestamp on the report — one text box showing when data was last pulled prevents the expensive conversation where someone decides on Tuesday afternoon using Monday evening’s numbers without realising it.

Editors can now let viewers refresh data themselves, a June 2026 addition, via right-click or the three-dot menu. Turn it on; it converts “the dashboard is broken” emails into a click.

The quota wall that breaks SME dashboards

If your dashboard has ever shown a red configuration error on half its charts at 3pm and healed itself the next morning, you have hit the GA4 Data API quota — the most common technical failure in Data Studio reports, and almost never diagnosed correctly.

Every GA4 query costs tokens, and the number depends on the complexity of the request and the size of the underlying data. Google publishes the core quotas:

Read the concurrency row again, because it is the one that bites. A standard property allows ten concurrent requests, and every chart is a query. A page with twelve GA4 charts fires more than ten requests the moment it loads. Put fifteen scorecards across the top of your overview page because they look tidy, and you have built something that fails intermittently by design.

Worse, quota is consumed per property, not per report or per user. An agency with four people opening the same client dashboard on a Monday morning, plus a scheduled email delivery, plus the client refreshing it, all draw from the same 40,000 hourly tokens. It is why dashboards that tested fine collapse in month three.

Five fixes, in the order we apply them:

1. Reduce charts per page. A page with six charts is more usable and less likely to fail than one with eighteen.
2. Check what each chart costs. In edit mode, right-click a chart or the canvas for the Google Analytics token usage option. It shows consumption per component and what remains. Most people have never opened it.
3. Shorten default date ranges. A chart defaulting to twelve months with a daily breakdown is dramatically more expensive than the same chart over 30 days.
4. Cut unnecessary dimensions. Complexity drives token cost: a table broken down by four dimensions costs far more than one broken down by two.
5. Move heavy reporting to BigQuery. There is no Data API quota on a BigQuery table. You pay query costs instead, usually small at SME volumes, and you get unlimited history rather than GA4’s retention window — one of the settings our GA4 setup guide tells you to get right on day one.

A structure that works: four pages, in this order

Almost every good marketing dashboard we have built for a Singapore SME lands on the same four-page shape. The order matters, because it mirrors how someone actually reads.

Page 1 — Answer. Six numbers, no more: enquiries this month, enquiries same month last year, spend, cost per enquiry, revenue if measurable, and one trend line. Someone should be able to close the report here and know how the business is doing. Every scorecard shows a comparison, because a number without one is not information.

Page 2 — Channels. Where enquiries came from and what each channel cost. One table, one chart. This is where consistent UTM tagging earns its keep: without it, this page is a large “Unassigned” row and an argument.

Page 3 — Campaigns and pages. The operational layer — which campaigns and landing pages are working. The marketer uses it weekly; the owner opens it twice a year.

Page 4 — Definitions. The page nobody builds and everybody needs. What counts as an enquiry. Which conversion actions are included. What attribution model and lookback window the figures use. When the data was last refreshed. What is knowingly missing — walk-ins, phone calls, WhatsApp.

That last page prevents more disputes than the other three combined, because two people arguing about whether leads went up are usually arguing about definitions without knowing it. When your Google Ads and GA4 numbers disagree — and they will — page 4 is where you explain that they count different things over different windows, as set out in our guide to attribution models.

Building it: the short version

The mechanics are easy; the thinking is the hard part. Create a blank report rather than opening a template — templates demonstrate features, and you will spend longer deleting charts than building six good ones.

Add data sources one at a time and rename every field as you go. “Sessions” becomes “Visits”. “Conversions” becomes “Enquiries (form + call click)”. If a field name would need explaining to your managing director, rename it now, because you will not do it later.

Build page 1 last. Do pages 2 and 3 first, see which numbers you keep looking at, then promote those six to the front. Dashboards designed top-down show what the tool makes easy; dashboards designed bottom-up show what the business needs.

Two 2026 features worth using. Cross data source filtering, shipped 15 January 2026, lets one date or campaign control filter charts built on different sources by overriding field IDs — solving the old annoyance of needing separate date pickers for Google Ads and Meta data. Component visibility controls, from the same release, show or hide individual charts for specific viewers, which is how agencies serve one report to a client and a fuller version internally without maintaining two files.

Then set a scheduled email delivery and put yourself on it. A dashboard nobody opens is a dashboard that has been silently broken for a month.

What it actually costs in Singapore

Data Studio itself is free. That is genuine, not a trial. The costs arrive from three other directions.

Note the Pro billing model, because it trips up agencies: it is priced per user per Google Cloud project. If you run a separate project per client, a five-client roster is five subscriptions, not one.

On build time, be honest. A first dashboard takes a competent marketer two to five working days once you include arguing about definitions. Singapore’s median gross monthly income for full-time employed residents was S$5,775 in mid-2025 including employer CPF, per the Ministry of Manpower, and the employer CPF rate for staff aged 55 and below is 17% on an Ordinary Wage ceiling that rose to S$8,000 from 1 January 2026. Three days of a mid-level marketer is not free — costing it properly is the same discipline we apply in the customer acquisition cost guide.

On grants: Data Studio is free, so there is nothing to claim on the tool. The Productivity Solutions Grant supports up to 50% of eligible costs, capped at S$30,000, and only for pre-approved solutions listed on the Business Grants Portal. A specific listed analytics product may qualify; a general dashboard build, an agency retainer or ad spend does not. EnterpriseSG is explicit that the company must directly apply for and manage the grant — third-party applications are not permitted. SDM is a pre-approved PSG vendor, but the application is always yours to make, and it must come before you pay for anything.

Five ways a dashboard lies to you

1. It shows only what connects easily. If Meta needs a paid connector and phone calls need manual entry, a lazy dashboard shows Google performance beautifully and everything else not at all — then concludes Google is your best channel. That is not a finding; it is a wiring diagram.

2. Totals double-count. Add Google Ads conversions to Meta conversions to GA4 key events and you get a number larger than your actual enquiries, because each platform claims credit for journeys the others also touched. Never sum conversions across platforms on one scorecard. Pick one source of truth for the headline and show platforms separately underneath.

3. Averages hide the distribution. An average cost per lead of S$60 can be one campaign at S$25 and another at S$300. Show the breakdown, not just the mean.

4. It flatters recent history. “Last 28 days” against “previous 28 days” produces alarming swings around public holidays. Singapore’s 2026 calendar is unusual — Vesak Day, National Day and Deepavali all fell on Sundays with Monday holidays in lieu, and Chinese New Year fell in mid-February. Year-on-year is the safer default for anything the owner reads.

5. Revenue figures include GST. If e-commerce revenue reaches the platform from a GST-inclusive order total, every revenue and return-on-ad-spend figure is inflated by 9% that was never yours: a reported 4.0 return on ad spend is really nearer 3.67. Strip it at source or annotate it on page 4 — the same trap covered in ROAS versus ROI.

When a dashboard is the wrong answer

Two situations where we tell clients not to bother yet. If your conversion tracking is not trustworthy, a dashboard makes bad data prettier and more persuasive, which is worse than a spreadsheet nobody believes — fix tracking first. And if you spend under roughly S$2,000 a month across one or two channels, the platform interfaces plus a monthly Google Sheet will serve you better than a dashboard you check twice; the overhead of maintaining a report exceeds the value of automating it at that scale. Revisit when you add a third channel or a second person who needs the numbers.

Frequently asked questions

Is Looker Studio still free after the Data Studio rename?

Yes. The free tier is unchanged: unlimited reports and data sources, and free native connectors to Google products. The rename on 16 April 2026 was a branding change, not a pricing change. The paid tier is now called Data Studio Pro and starts at about US$9 per user per month, billed per Google Cloud project on annual terms.

Why does my dashboard show a quota error in the afternoon and work again the next day?

You are exhausting the GA4 Data API token quota. Standard properties get 40,000 tokens per hour, 200,000 per day, and 10 concurrent requests. A page with more than about ten GA4 charts, opened by several people, will hit it. Reduce charts per page, shorten default date ranges, check per-chart token usage via right-click in edit mode, and move heavy reporting to BigQuery.

How do I get Facebook and Instagram Ads data into Data Studio?

There is no free native Google connector for Meta. Your options are a paid partner connector, a manual weekly export into a Google Sheet the dashboard reads, or piping data into BigQuery. Most Singapore SMEs with meaningful Meta spend end up paying for a connector; budget it as a recurring cost from the start.

How fresh is the data in a Data Studio dashboard?

Less fresh than most people assume. Google’s own marketing products — Google Ads, GA4, Search Console, YouTube Analytics — refresh every 12 hours and that rate cannot be changed. Google Sheets defaults to 15 minutes, and BigQuery can go down to one minute. Use the platform interfaces for live campaign monitoring and the dashboard for reporting, and display the last-refreshed time on the report.

Can I claim a marketing dashboard under PSG?

Data Studio is free, so there is nothing to claim on the tool. PSG supports up to 50% of eligible costs, capped at S$30,000, and only for pre-approved solutions listed on the Business Grants Portal — a specific listed analytics product may qualify while a custom dashboard build, an agency retainer or ad spend does not. EnterpriseSG requires the company to apply for and manage the grant itself, before committing to any spend.

Should I use a dashboard or my agency’s monthly report?

Both, doing different jobs. The dashboard answers “how are we doing” on demand. The monthly report should explain what changed, why, and what happens next — the analysis a chart cannot provide. If your agency’s report is only numbers you could read off a dashboard, you are paying for data entry rather than thinking.

Where to start

Build one page before you build four. Put six numbers on it, each with a year-on-year comparison, and use it yourself for a fortnight. You will learn quickly which numbers you actually look at and which you added because they were available — a discovery worth more than any template. Then add the definitions page. Then, and only then, worry about connectors and automation.

If you would rather have this built properly and handed over — tracking verified first, definitions agreed in writing, numbers reconciled against actual sales — our performance marketing team does exactly that for Singapore businesses. See the measurement work it supports in our client case studies and the wider picture in the performance marketing guide. If you are still deciding which numbers deserve a place at all, start with the GA4 reports worth checking and the customer lifetime value figure that gives every other number its meaning.

Last updated 1 August 2026. Written by Adrian Tan and the SDM team. Sources: Google Cloud documentation for Data Studio (product overview, release notes, manage data freshness), Google Analytics Data API v1 quota documentation, Ministry of Manpower income statistics, CPF Board contribution rates from 1 January 2026, and EnterpriseSG Productivity Solutions Grant guidance.