The default conversion value that makes trade ROAS meaningless

Key facts

• 77% of trade businesses in the sample leave Google's default value of one dollar on every enquiry.
• 14% record no value at all, so there is no return on ad spend to read.
• 9% send real job value, and all of it came from won jobs uploaded after the fact; none came from website or call tags.
• In the median business, the biggest tenth of won jobs carried 41% of recorded revenue.

Definitions. A conversion value is the dollar figure attached to each conversion in Google Ads, and return on ad spend (ROAS) is total conversion value divided by cost.

Return on ad spend is conversion value divided by cost. If every enquiry is worth one dollar, conversion value is just the number of enquiries, and ROAS is enquiries per dollar of spend. It moves when the enquiry count moves, and it says nothing about whether the work those enquiries produced was a tap washer or a full re-pipe.

That causes three problems:

• ROAS comparisons mislead. A campaign that fills the diary with small call-outs shows a higher “ROAS” than one that brings in fewer, larger jobs, because the column rewards count, not value.
• Value-based bidding optimises the wrong thing. Switch to Maximise conversion value or a target ROAS strategy with a one-dollar default, and Google optimises for the number of enquiries while everyone believes it is optimising for revenue.
• Businesses cannot be compared. Two businesses that record value differently, or record revenue at different points, cannot be compared on ROAS at all. Nor can one campaign before and after its value setting changed.

There are two workable options. Leave value empty and bid on enquiries, judging the account on cost per enquiry and cost per won job. Or feed Google real job values. What does not work is the middle: a placeholder that looks like revenue and is not. Before either, make sure the enquiries themselves are counted correctly; our article on phone taps that inflate trade leads covers that first step.

Because in a trade business the price is set on site, often days after the enquiry. A form about a hot water system could become a small repair or a full replacement; a call about a switchboard could be a quick fix or an upgrade. At the moment the tag fires, nobody knows.

That is why, in SoudCoh's research sample, none of the real values came from website or call tags. They came from the job book: won jobs recorded against the click or call that produced them, and uploaded later as an offline conversion. An online shop can send a basket value at checkout. A trade business has to send value back once the job is won.

Some businesses consider a middle path: a fixed value per enquiry based on an average job. It reads better than one dollar in reports, but it still treats every enquiry the same, so it cannot tell bidding which services or suburbs produce the valuable work.

Because revenue in a trade business is concentrated. In the median business in SoudCoh's research sample, the biggest tenth of won jobs carried 41% of recorded revenue. A switchboard upgrade, a full re-pipe or a leak that became a bathroom renovation can move a month's return on ad spend more than any campaign change.

That changes how the numbers should be read:

• Judge campaigns on rolling quarters, not single months, once real value is flowing.
• Look at the median job as well as the total. A total can jump on one job; the median tells you whether the typical job changed.
• Watch the services that produce the big tickets separately. They are rare enough to be cut by mistake when a monthly review looks only at cost per lead.

Our article on why emergency searches are a small share of trade demand looks at how a budget is split between planned and urgent work.

Not necessarily, and cost per lead alone cannot tell you. It is the easiest number to optimise and the least connected to profit: the ad group that fills the diary with cheap small jobs is not automatically the one that pays for the ute. With only 9% of SoudCoh's research sample sending real job value, most trade businesses cannot yet see the difference. You do not need a sophisticated system to see the difference: a monthly spreadsheet that lists each won job, its value and the campaign or ad group it came from is enough to start moving budget on revenue per ad dollar rather than cost per lead.

This is the method we set up for clients, and every step can be done in-house. Nothing here changes how your team quotes or invoices; it only requires recording which enquiry each won job came from.

1. Capture the click on every enquiry. Store the Google click identifier (the gclid, or the gbraid and wbraid Google uses for some iPhone traffic) from the landing page address on each form submission, and attach it to tracked website calls. Keep it with the lead in your CRM or job system.
2. Record won jobs with a value and a won date. From your job-management system, a periodic spreadsheet or your invoicing, list each won job, its value excluding GST, and the enquiry it came from.
3. Create an offline conversion action for won jobs. In Google Ads, add a conversion action that imports conversions from clicks, and set it as secondary so it informs reporting without changing bidding yet.
4. Upload on a regular rhythm. Weekly suits most businesses. Google only accepts an imported conversion within the conversion window you set for the action, at most 90 days after the click, so do not let won jobs pile up.
5. Record how revenue arrives. Note whether values are entered live or imported from a sheet, and where the won date comes from. Two businesses, or two periods, that record revenue differently cannot be compared on ROAS.
6. Move budget on revenue per ad dollar only when the data is complete, over rolling quarters and never on a single month. Consider value-based bidding only after that.

If your forms collect an email address or phone number, Google's enhanced conversions for leads can match uploaded won jobs using that information, hashed, rather than relying only on the click identifier. Either way the rule is the same: value comes from the job book, not the web page. Our tracking work sets this loop up end to end.

Report the measures your data can support, and add revenue as it arrives:

Compare cost per enquiry with the Google Ads benchmarks for Australian trades we publish, and keep ROAS out of the conversation until real values are flowing. Our white paper From keyword to revenue sets out the full method, from the click to the won job. For trade-specific set-ups, see our marketing for plumbing businesses, electricians, HVAC businesses, roofing businesses and construction businesses, or talk to us about your own account.

What conversion value should a service business use in Google Ads?

Either no value, with bidding and reporting based on enquiries, or real job values uploaded from won jobs. In SoudCoh's research sample, 77% of trade businesses used Google's default value of one dollar per enquiry, which turns the ROAS column into a count of enquiries. A website cannot know a job's value when the form is sent, so real value has to come from the job book.

Is ROAS a useful measure for a trade business?

Only once real job values are uploaded. Until then, the ROAS column for most trade businesses is a count of enquiries dressed as dollars. Even with real values, judge it over rolling quarters: in the median business in SoudCoh's research sample, the biggest tenth of won jobs carried 41% of recorded revenue, so one large job can swing a month.

Should I switch to target ROAS bidding?

Not while every enquiry carries a placeholder value. With a one-dollar default, value-based bidding optimises for the number of enquiries while looking like revenue optimisation. Upload won-job values as a secondary conversion first, and consider value-based bidding only once several months of complete data have built up.

How do I get job values into Google Ads?

Capture the click identifier on each enquiry, record each won job with its value against the enquiry it came from, and upload won jobs as an offline conversion action, set to secondary at first. Upload regularly, because Google only accepts an imported conversion within the action's conversion window after the click.

Why doesn't my website send the job value automatically?

Because the price of a trade job is usually set on site, often days after the enquiry. When a form is sent or a call connects, nobody knows whether it will become a small repair or a large replacement. That is why, in SoudCoh's research sample, every real value came from uploaded won jobs, not from website or call tags.