Brand promotion: this article introduces Guangsuan's Google Ads management services.
A SaaS growth team closes the month with two reports that rarely agree. Finance holds ad spend by campaign and day. The CRM holds qualified pipeline by close date. When the numbers differ, the reason is often fixed before download: rows without a stable key, two calendar rules in one comparison, or different views of what should count as a conversion.
Short answer for a closed-month check: note customer and campaign numbers plus every split that makes a row unique, note range with account zone and whether conversions sit by click date or by conversion date, tag primary actions apart from secondary steps, and turn micro amounts into currency in one logged place. Use CSV or XLSX for review and Sheets for shared review, keep JSON for an API or warehouse path, and test one action in one campaign group before any bid move. A quick screenshot for a colleague does not need this routine.
Decide what the CRM comparison must answer
Write one sentence first, for example which campaigns produced qualified demo requests in October. That sentence picks the range, the grain and the action to carry forward. A campaign total answers how much was spent. A campaign by day by device file answers where and when it was spent. Those two grains cannot share one key without overwriting valid rows.
On a Campaigns, Ad groups, Keywords, Search terms or Ads view, set the range and columns before using the download control at the top of the table. In the Reports area, use a predefined view or a custom table when day, device, geography or conversion splits must sit together.
For repeated or large pulls, query with GAQL through the Google Ads API rather than clicking. The rows carry only the fields named in the query, such as campaign number and name with clicks, conversions and cost in micros.
Freeze identity and row grain
Names change, numbers stay. Keep the customer number in the file. Keep the numeric campaign number beside the campaign name. Keep each split that defines a row, such as date, device, network part and conversion action where relevant.
For work beyond spend and clicks, carry search terms beside bid keywords so intent stays apart from the term bought, and add device, network, location and hour only when the opening question needs them. Skip a single file with every column turned on.
Keep file names that record account, report, range, date rule and pull time. Leave the raw download untouched and do analysis in a copy so the same slice can be replaced on retry instead of doubled.
Freeze range, zone and date splits
The download button is easy to find. The mistakes happen in two places: choosing a date range that does not match the question you are asking, and reading the conversions column without knowing which conversion basis it uses.
Exports are dated in the account's time zone, which may not match your calendar or your CRM timestamps. A request logged late on the last evening can land in different months on each side with no error on either side. Write the zone on the file itself. Use the segment control to cut by day, week or month rather than joining several separate ranges later. A monthly view can hide a weak week, while a daily pull over a long span turns hard to read, so export the grain you will actually analyse.
Freeze which conversions count
Google Ads can attribute and report conversions on more than one basis, and the default column is not the only valid view. Conversion-time placement puts the event where it happened. Click-time placement puts it where the click happened. A September click that turns into an October signup sits in different months under those two rules. Both rules are legitimate until they are mixed, so keep one rule on both sides of a period comparison.
Before you build a report on conversions, open the conversion settings and write down: the basis (conversion or click time), which actions are primary, the counting method and the attribution model. Changing any of them shifts history, so the note belongs beside the file.
Secondary actions — often micro-conversions such as a page view or a click-to-contact — are reported separately and should be labelled as such in your export. Only primary steps feed the Conversions column used for bidding.
If you rely on offline or CRM-confirmed leads, remember the export only shows what Google was able to record; sales-confirmed numbers live in your own system. Keep clicks, platform events, qualified enquiries and closed revenue as four separate tallies.
Keep money, format and file handling consistent
Google Ads supports CSV, CSV for Excel, TSV and XLSX, and can send a file to Google Sheets or by email on a schedule. Use CSV for scripts and stores, XLSX for sheet review, and the Sheets route when several people must see the same live file.
Open CSV through Data > From Text/CSV, force UTF-8, confirm the delimiter, turn cost and clicks from text into numbers, and check one date row before trusting month-day order. Add the cost column and compute your own ratios in the sheet rather than relying only on exported averages, because aggregated averages can hide the segments you actually need to act on.
Cost arrives in micros in the account currency, so apply the divide-by-one-million step once and note where it happened. Files in different currencies should not have raw totals added together.
For the field routine behind this list, see Guangsuan's guide to exporting Google Ads data and reading dates and conversions.
Where JSON fits in this workflow
Reporting downloads do not list JSON. XML is near it in the menu but serves a different reader. When a downstream tool asks for JSON, take one of two owned paths. Call the REST form of the Google Ads API and receive JSON rows, or land data in a warehouse and emit JSON there for a recurring CRM match. Turning a CSV download into JSON on your own machine is allowed, but that script becomes part of reporting: keep it versioned, name an owner, and test it on a small slice checked by hand.
A narrow pass before any bid change
Start with one primary action, one campaign group and one closed period. Line up basis, range and zone on both extracts. Hold ad-sourced records apart from organic and referral entries. Review counting behaviour and repeated uploads before calling a remainder unexplained.
As a hypothetical teaching illustration with invented figures, not a client outcome: a file records 51 ad-tracked contacts in a month. Sales accepts 18 as qualified under a written definition, 13 dated inside that month and 5 in the next month, with 3 reaching order. Each tally answers a separate stage, so none cancels another.
When few contacts ever qualify, look at query intent and page fit before moving bids. An export is only useful if the campaign structure behind it is sound — clean naming, one intent per campaign, and conversion tracking that matches how your business actually receives leads.
When Guangsuan can help
If reports stay hard to read because structure is tangled, or turning extracts into choices takes too long, Guangsuan's Google Ads management service covers account structure, search terms, landing pages, conversion checks and reporting. Scope and fees sit in a contract, with no fixed enquiry promise. Details are on the Guangsuan Google Ads management service page.
For a first look, bring the action list with primary flags plus one annotated extract that shows basis and zone.
FAQ
Click time or conversion time for month end?
Pick the rule that answers the question and keep it on both sides, with window and account zone filed with the extract.
Does an openable file mean a matchable file?
No. Layout says nothing about grain, units, date rule or which actions were summed.
Can a missing term mean hidden rows?
Yes. Very quiet rows can drop from a filtered view, so confirm filters and grain before saying the term never matched.