The Service Business Google Ads Conversion Guide for 2025

For service businesses that invest advertising budget into Google Ads, one of the most important things to get right is the conversion set up. The reason for this is simple – To give Google Ads the data signals it needs to optimise most effectively.
The majority of businesses we encounter have issues with their conversion tracking. This mainly stems from their conversion tracking being set up in a legacy manner that once worked fine. Before 2020, sophisticated conversion tracking didn’t matter nearly as much as what it does now, since automated bidding was in its infancy, and manual bidding tended to drive better results. Back then, as long as you had conversion data to provide insight for you to manually optimise towards, that was enough.
However 2020 was a key turning point for automated bidding strategies. They began outperforming the manual strategies, which meant we as marketers began handing a lot more control over to Google. It no longer gave us an advantage by spending our time manually creating adjustments based on time of day, device or other demographics, as Google’s AI would get more conversions via automated bidding.
Since then we’ve seen the bid model get more and more sophisticated, significantly more so after 2023. Whilst on the surface you would think this only means positive results for businesses using Google Ads, we’ve seen many major issues most of the clients who we’ve taken on.
With all the compounding developments in machine learning, Google Ads is becoming continually more powerful with how it utilises AI to do what you ask it to. Program it in the right way and it can be incredibly powerful in getting you what you ask for, however program it slightly incorrectly and it can go very much in the wrong direction.
Conversion goal set-ups which worked well for service businesses a couple of years ago can be very inefficient today, which is why it is critical to ensure the conversions are set up in a manner that compliments how Google Ads works today.
The powerful Google Ads algorithm needs to be directed with data and logic
Typical service business lead types
For service based businesses, the typical leads that are driven from Google through their websites are a range of the following:
- Phone calls
- Lead Forms
- Live chats
- Brochure downloads
- Request callbacks
This is very different from ecommerce websites where users will primarily purchase on the website. Service based businesses will often have online purchase capability too, but this will be alongside the other types of leads.
With most service businesses, users will land on the website, digest the information before getting in touch with the business and going into the lead nurturing process. At this point the actual purchase still hasn’t happened, and there can be many touchpoints over a period of weeks or even months before the user finally makes the purchase. Tracking this information is key to how we value the leads (discussed later).
Ensure all key actions are tracked
The first thing to get right is to ensure that all of these actions are being tracked correctly. The users who undertake these actions to get in touch with you will likely have some shared characteristics or behaviours. It may be they are from a similar location, are a similar age range, have been on your site several times or have been on competitor sites etc. We want to ensure that all of this crucial insight is communicating properly to Google, so it can up-weight bids for these types of users, and drive more of these leads for you.
Form actions
Form actions are the most simple to get set up correctly (brochure downloads, request callbacks and enquiry forms). When prospective customers submit these leads,they are required to fill in their personal information on your website (name, email address, phone number etc.) after which they’ll press a button to submit this information. This button click or where it leads to is essentially how we track these form submissions.
When this button is clicked, a trigger can be set up to push an event into Google Ads or GA4. In the below example we have set up the GA4 conversion event to be triggered once a user reaches a thank you page after submitting the lead form. It makes most sense to have a thank you page as this not only provides a medium for you to communicate any next steps to the user, but it also makes it easier to track this journey with GA4.


This GA4 conversion event is then pushed through directly into Google Ads. Alternatively you can set it up directly within Google Ads, but these ultimately track the same thing, and the benefit of doing it in GA4 is that you can also monitor these submissions from other channels too (e.g. organic, referral, direct, social etc.).
Conversion events can also be triggered without sending users through to a thank-you page, based on elements like the text in the CTA buttons, but we generally find it makes more sense to send users to thank-you pages. Therefore for lead form fills, callback requests and brochure downloads, the users should be sent to 3 different thank-you pages, with each of these triggering separate conversion events in GA4.
Inbound calls
Across almost all of our service based clients, we find that inbound calls are the strongest type of lead. When a user lands on your website, interacts with your content and then calls you, there is a considerable amount of intent. Additionally once the user is on a call with one of your call handlers, it is the first time in that journey where they are able to express any concerns and reservations to a human; no matter how good your website is, there is nothing quite like the human touch to sell your services.
With lead form fills and callback requests whilst there is still intent, you’ll find that the people have often had time to cool off and when you phone them back, most won’t answer. Furthermore, of the people that do answer, the conversion rate is usually significantly lower than inbound calls.
As call tracking is a more complex set up than tracking lead forms, we have encountered a lot of businesses who don’t have this set up. These aren’t just small businesses either; some of them have been major market leading private equity backed businesses that spend up to £1m per year on Google Ads, yet still haven’t been tracking their phone calls and optimising their PPC toward these. This means the best quality leads aren’t getting fed back into Google Ads, and instead the algorithm will only be directed to optimise for lead form fills.
Some businesses have attempted to counteract the lack of call tracking by adding a conversion goal when a user clicks the call button in the header on mobile. Essentially when they click this call button, it triggers an event in GA4 in the same manner as a lead form button click.
Whilst this does allow you to optimise for calls, it also means you are only optimising for mobile and disregarding any calls that have come through on desktop. On desktop the users will see the phone number on your website and have to physically type it in to their phone which will mean no button click actually occurs; therefore this won’t track calls.
The problem here is that desktop calls whilst usually fewer in quantity are usually higher in quality, yet Google won’t be fed any conversion data for this, and you’ll see CPCs drop for desktop as Google goes after the mobile traffic; we’ve seen this first hand on multiple occasions.
Additionally, you’ll not be able to put any measure of quality such as length of the calls, so the bid strategy will be optimising for the low quality calls as well as the high quality calls. However ideally you want a measure of quality, so you are only optimising for the high quality calls. The way you can do this is by getting call tracking set up.
Call tracking
The way call tracking works is to dynamically change a phone number on your website to a purely unique one for each individual user. So if your phone number is 0800-111-1111, it could change it to 0800-222-2222. Only one user would see this number, and when they call this number, it would be diverted back through to the original number in the back end.
Alongside showing a purely unique phone number to each user, the call tracking software will also capture the marketing channel and if they come via a PPC campaign it will capture the campaign, ad group, keyword and the google ads click ID; this is a number that specifically identifies the exact PPC click that brought the user to the website.
By having that unique phone number alongside the channel and PPC data, if somebody calls the number, the call tracking can not only monitor it, but also match it back to the click which drove it; essentially it means you are able to track exactly what PPC campaigns or keywords drove phone calls.
What’s more, depending on the call tracking, there are different measures of quality you can pass through too. For example, you could base this on the call length being over 4 minutes which would then only send a conversion event if the call lasted this length or more. This means you wouldn’t be passing back in those low quality calls and only the data you want would be sent into the Google Ads algorithm.
The most advanced call tracking set ups we work on are with our call tracking partner Call360. The way these work are by transcribing every call, using AI to identify keywords or keyword patterns that determine what the call was about, and then triggering conversions only if the call was a good quality sales call. It means the bid strategy won’t be optimising for irrelevant calls or existing customer calls, but just the target quality calls.
Furthermore, it enables for categorisation of the call based on the actual service that was offered, triggering different conversion goals for different services. So if you have different services each with different values, you can set up your call tracking to separate out these services into different conversion goals. That way if you want to push up the amount of leads you get for one of the services, you can assign a higher valuation to that type of call which will instruct Google to bid up to get those types of calls for you.
Micro conversions
Where conversion numbers are quite low, it can be useful to set something up called micro conversions. The reason for this is simple; we want to feed the Google Ads algorithm with accurate data signals which allow it to optimise.
This is particularly useful for new accounts where there isn’t a history of conversion events as it enables them to get an initial boost of data for the bid model. Additionally it can work well for accounts that have low spends.
An example of a micro conversion could be the 1st step in an online booking journey. If there are a low number of online bookings, you could optimise for the initial step which is the user clicking into the booking journey. However as this would be just the first step in the customer journey, it is key that it is valued significantly lower than say a phone call.
Other examples of micro conversions are users who have been on site for over 5 minutes, or even having certain pages like the contact page visited. These are behaviours that you would expect from good quality prospects, so with a lack of other conversion data it is worth optimising towards.
Conversion valuation exercise
As mentioned earlier in the article it is key that you undertake an exercise to calculate or at least estimate the values of your different conversion goals. The way to do this is to try to work out what the conversion rate is for each of the different conversion events to an actual sale. If the data isn’t readily available, you will likely need to work with your contact centre to gather this information; they are the ones who will usually follow up on the different lead types, so are the team to begin with.
Below are some example conversion actions and some typical conversion rates from lead to sale. These can vary considerably by industry and company which is why the valuation exercise is key.
Calculating the total sale value (£1,000 in this example), we can work out from there what the value is of the individual conversion actions by multiplying the total sale value by the conversion rate. For example say a phone call converts to a sale at 20%, then multiplying 20% by the total sale value of £1,000 would be £200. Out of every 5 calls there would be on average 1 sale, and therefore the value of a lead would be £200.
Total sale value £1,000
Conversion Action | Conversion rate to sale | Lead valuation |
Phone calls | 20% | £200 |
Lead forms | 8% | £80 |
Live chats | 5% | £50 |
Brochure downloads | 1% | £10 |
Request callbacks | 6% | £60 |
It doesn’t matter what you use for the total sale value, what matters is how the lead valuations are valued proportionally against one another. Most of our clients offer a variety of services at different price points, and with service businesses there is ongoing revenue, so it is tricky to really determine an average sales value. Therefore in most cases we’d just assign the total value to be £1,000.
Conversion value bidding
Once the above conversion value bidding exercise is completed, you will essentially have calculated the approximate value each conversion holds.
As discussed earlier in the guide, most service based business PPC accounts we have taken on in 2024 and 2025 simply put all of these different conversion events together, and instruct Google to drive as many as them as possible via a maximum conversions bid strategy, regardless of how they convert to sale.
The problem with this is that if a brochure download is only worth £10 to the business, but an inbound call is worth 20 times more at £200, really you want to bid up for the type of people who are more likely to call opposed to filling out a brochure request form.
By communicating this to Google, we can make use of its powerful AI driven algorithm. It will look at the characteristics of people who phone up vs people who fill out brochures, and when it sees those same characteristics in other searches, it will up weight bids to capture these users. For example the characteristics could be users who have been looking at various competitor sites for the past 7 days, and are searching during weekdays at 11am. The phrasing of the search terms may be slightly different, and it may be that desktop is the device that drives the high quality calls. It may additionally find that users in the age range of 45 to 60, female and in particular postcodes are more likely to call.
The powerful algorithm considers all these factors and many more to determine how much to bid, but without communicating the value of the different leads, we just aren’t giving it the information it needs to do this.
The way in which we program it with this information is very simple once the valuation exercise is complete. The first step is to add in the conversion values to each of the conversion goals within Google Ads. The way to do this is to go to The first step is to add in the conversion values to each of the conversion goals within Google Ads. The way to do this is to go to goals/conversions/summary:

From here you can select each of the different conversion goals. Within each of them you have the option to assign a value of the conversion goal. Generally for service business goals you will want to use the same value for each conversion, so select this option and add in the pre-calculated value.

After this exercise is completed, you then need to go into the campaign settings and change the bid strategy to maximise conversion value. This ultimately says to Google to get as much conversion value as possible for the budget you put into the campaign.

Moving to target return on ad spend (TROAS)
Initially you will want to leave the target return on ad spend box unticked because Google Ads won’t have the necessary data it needs in order to utilise this.
However, after getting a good amount of conversion value data, it is worth testing this out on a campaign by campaign basis.
Before doing this, we need to first look at the search impression share metrics in the campaign view, in particular the search lost impression share to budget. In the below example, it shows the search lost impression share to budget is 39.50%. This means that for 39.50% of the time this campaign is eligible to show for a search term, ads aren’t being shown because the campaign has run out of budget.

In this example, by enabling TROAS, in theory it means that we should be able to reduce CPCs so that for the same budget we get more clicks. By doing this, our ads should show for more of the available searches, and see this metric of search lost impression share to budget come down.
This equation is how we go about calculating what the existing ROAS is. By looking at the past 30 days data (providing the campaigns have around 15 or more conversions each), we can calculate what the existing ROAS is.

In this particular example the cost is £23,751 and the conversion value is £17,250 which works out to a ROAS of 0.726

Now we know the ROAS of the campaign alongside the detail around how much search impression share we are losing to budget, we can use this to assign a target ROAS into Google Ads and attempt to do as we have discussed.
When assigning a target ROAS it is important to gradually do this in increments every couple of weeks as the bid model can be highly sensitive; if the changes are too drastic it can lead to a huge drop in traffic with campaigns barely showing, so we want to avoid this at all costs. Instead it’s best to look at what the ROAS has been, and initially increase that by 10%. A 10% increase to a 0.73 ROAS is 0.80, so this should be set and then monitored carefully over a period of a couple of weeks.
If it goes successfully, we would expect to see the actual ROAS of the campaign to increase to the target amount after running for a couple of weeks. Essentially it means that for the same budget as before, we would be getting a greater level of conversion value. In addition, the impressions limited by budget % would decrease but not to zero.
The risk of doing this is if the bids drop by so much that the campaign budget can no longer be spent. It is the reason why it is crucial to check the impressions limited by budget before undertaking this. If your weekly budgeted spend is £1,000, the risk is that it stops spending this and perhaps it only spends £500 per week.
However if it does get the budget spent and the impressions limited by budget are still relatively high, then after the 2 weeks it is worth increasing another 10%. Follow this exercise until it can’t get the budget spent, and that then is the profit maximising point.
Still, it is key to monitor this even after you have found this point because if in certain months the search volume is lower, then you may struggle to get the budget spent, and therefore you need to react to this market change and decrease the target ROAS. Furthermore, competitive changes over time may also force you to increase it.
Integration with Salesforce, Hubspot or other CRMs
The final section of this guide is relating to integrating your CRM into Google Ads. This is increasingly becoming a way to get some of the best returns from PPC, as it is a way to only optimise for the high quality leads and disregard the low quality ones. This integration is applicable to all form fills, but not inbound calls; for those calls it requires separate call tracking.
Within a CRM, you usually have lifecycle stages for which prospective customers move through the funnel following various interactions with the sales teams. The initial stage will usually be a lead/enquiry where the user has followed an action like submitting an enquiry form or a callback request. This is what we optimise towards in the previous section about form actions through Google Ads and GA4.
The problem with the initial lead from these lead forms is that there is no indication of quality. In most instances these people won’t answer the phone when you call them back, and have likely enquired with other competitors too. One of our clients used to get people looking for jobs filling out these forms, despite us adding hundreds of negative keywords which disabled any job seeker search traffic coming through the campaigns, and big prominent notices on the form that this form was for prospective customers and links taking the job seekers to the relevant sections. Ultimately this meant the bid model was optimising for both demographic sets of prospective customers and job seekers, which was what we didn’t want happening.
By integrating into the CRM, we can bypass this initial stage and optimise only once a prospective customer has done some action to indicate that they are good quality, such as having a conversation with the sales team, and the sales team have assigned this within the CRM.
Furthermore we can even pass data once the user has got further through the funnel and become a customer; essentially we’d have two conversion goals correlating with the CRM lifecycle stages of Sales Qualified Lead and Sale.
- Lead/enquiry
- Marketing qualified lead
- Sales qualified lead
- Sale
Both Salesforce and Hubspot offer integrations into Google Ads, but what is necessary to do is to ensure that the forms pass through the Google Click ID into the CRM via the form fill. All CRMs work slightly differently, so we haven’t focused this guide on the setup for each, more the reasoning for why it is key to get this setup in most instances.
Ultimately by setting up the bid model to work in this manner, we are programming the Google Ads algorithm with the most accurate data signals to give it the best chance of maximising ROI. The more we can do that, the more the bid model will bid up when it sees a search it considers most likely high value, and subsequently capture that search over your competitors.