Google Ad Case Study: Review and Advice?

Hi All,

I'm currently working with 3 local businesses running their Google ad campaigns for them and would like some feedback on how I can go about improving some campaigns that I am running for them.

Company A & Company B are taxi companies operating in the local area. They're both owned by the same people but operate as two separate entities.

Company C is an online booking platform specifically for Airport Taxis.

Below are the metrics for all companies over the past 30 days

Company A:

• 5 campaigns with an overall CTR of 14.50%
• Average CPC of £0.23
• Conversion rate of 76% (conversions are: a phone call at £5 per phone call, app download at £5 or web booking recording as viewing the booking page with a value of 0.1 per view)
• Overall ad spend £1,456.47

Company B:

• 5 campaigns with an overall CTR of 15.53%
• Average CPC of £0.23
• Conversion rate of 73% (conversions are: a phone call at £5 per phone call, app download at £5 or web booking recording as viewing the booking page with a value of £0.10 per view)
• Overall ad spend £1,802.76

Company C:

• 5 campaigns with an overall CTR of 15.53%
• Average CPC of £0.25
• Conversion rate of 47.53% (conversions are: add to basket as £0.01 if they viewed the booking page and purchase as £10 if they went through to the thank-you page)
• Overall ad spend £831.70

Below is a breakdown of my ad strategy for companies A and B:

• 4 campaigns
  • 3 focusing on 8-hour intervals (Mon-Fri 08:00-16:00, 16:00-00:00 & 00:00-08:00) and the final just focusing on the weekend.
  • Campaigns focused in the city +20m radius
  • All campaigns share the same ad groups which are
    • Generic taxi words - keywords include: taxi near me
    • Dynamic ad group
    • specific services - keywords include: airport taxi, minibus taxi
    • Keywords have primarily been broad match except for the brand name which has been an exact match
  • Each ad group has one detailed search ad focusing on the keywords in the ad group as well as one call ad.
• 1 campaign
  • Focused on a particular area of the city the company wanted to grow in
  • 4 ad groups consisting of
    • generic taxi words
    • dynamic ad group
    • area-focused - keywords include: taxi in {area}
    • competitor focused keywords - keywords include: {competitor} taxi alternative
    • Keywords have primarily been broad match except for the brand name which has been an exact match
  • Each ad group has one search ad and one call ad also

The breakdown for Company C:

• 4 campaigns
  • Each campaign is focusing on an airport + 20-30 mile radius
  • Ad groups are
    • Generic airport taxi words - keywords include: airport taxi near me, cheap airport taxi etc
    • [airport of interest] words - keywords include: cheap taxis near {airport name}
    • dynamic ad group
  • Each ad group has 1 search ad with landing pages that include the targeted keyword.

Overall, companies A & B are performing well with very few complaints. the companies are getting a good number of phone calls through especially for the area-focused campaign that we have run.

Company C I can see is doing relatively well in terms of CPC and CTR, however, isn't converting well. Many people go through to the booking page but not a lot of checkouts. I have made the following suggestions to them:

• Implement an abandoned cart for people who have entered their details but didn't check out offering some discount code.
• Offer a discount code pop-up on exit intent
• Add a live chatbot function for any customer questions and to improve trust
• They don't currently have reviews, maybe start adding reviews on Google or Trustpilot to improve brand trust

Companies A & B have asked if there is any way lower ad spend while keeping the conversion the same or improving the conversion rates at their current ad spend. Bearing in mind the CTR and conversion rates, I said that there isn't a whole lot of room. I advised that they should scale as all campaigns are currently limited by budget telling me that there is room for growth.

In terms of my questions, I want to see if someone could help with the following:

• To check performance, what key metrics (columns) do you have for:
  • Campaigns
  • Ad Groups
  • Keywords
• How could I go about lowering the ad spend?
  • I have looked at negative keywords but google is doing well at not matching the ad to any bad keywords that aren't relevant to the business. If there are any bad keywords, they only have 1 click for example so it makes me feel like it is a tedious/useless exercise to fine-comb through all of these search terms.
• For companies A & B, I have quite a lot of keywords in one ad group for example, in the airport taxi group, I have 72 keywords. Are these too many keywords for one ad group?
• When looking through the search terms, how do I know which keywords to start adding?
  • over the past 30 days, I am looking at the keywords that resulted in the best CTR and conversion rates and adding these but again, most keywords only have 3-4 clicks per word but accumulate over all the search terms to a significant amount. The ones that do have a high number of clicks are either words that I have already added or ones that are an exact match or phrase match with keywords that I have already added.

Any other advice on how I can improve my campaigns for these customers would be greatly appreciated! Feel free to ask any questions and I'll do my best to answer and help where I can.

It seems that the campign is setup quite well and the CTR, the results (conversions) also are quite good.

Shabzy said:

limited by budget

Click to expand...

You may increase the daily budget of the campaign, ofcourse only if you are seeking more conversions on a daily basis.

Trust WorkShop said:

It seems that the campign is setup quite well and the CTR, the results (conversions) also are quite good.

You may increase the daily budget of the campaign, ofcourse only if you are seeking more conversions on a daily basis.

Click to expand...

We have increased it for the particular local area targeted campaign which is performing really well. They don't really wish to increase it for the others but want me to somehow optimize the current campaigns further somehow.

Shabzy said:

We have increased it for the particular local area targeted campaign which is performing really well. They don't really wish to increase it for the others but want me to somehow optimize the current campaigns further somehow.

Click to expand...

If you are limited by budget then it sounds as though you can lower the CPC and/or the CPA (cost per action) to further optimise your campaigns.
Look at cost per action and impression share at a keyword level, focus on the keywords with lowest cost per action and most potential (missed impression share). Look to reduce bids and/or pause keywords with highest cost per action.

Lots of other little details you can look at (ad copy/assets/landing pages/bid modifiers), but this would be top of my list as a fundamental value mechanic.

Shabzy said:

• How could I go about lowering the ad spend?
  • I have looked at negative keywords but google is doing well at not matching the ad to any bad keywords that aren't relevant to the business. If there are any bad keywords, they only have 1 click for example so it makes me feel like it is a tedious/useless exercise to fine-comb through all of these search terms.
• For companies A & B, I have quite a lot of keywords in one ad group for example, in the airport taxi group, I have 72 keywords. Are these too many keywords for one ad group?
• When looking through the search terms, how do I know which keywords to start adding?
  • over the past 30 days, I am looking at the keywords that resulted in the best CTR and conversion rates and adding these but again, most keywords only have 3-4 clicks per word but accumulate over all the search terms to a significant amount. The ones that do have a high number of clicks are either words that I have already added or ones that are an exact match or phrase match with keywords that I have already added.

Click to expand...

Negative keywords or culling your keyword list wont do much really. It's pretty moot nowadays with all the improvements google have pushed through lately, the bot is fairly good at optimising on its own for the most part.

Your metrics look really good, you've clearly done a great job for them thus far

To be honest, it sounds like the customers are a pain in the arse. Spend more money or stfu

jacksi said:

Negative keywords or culling your keyword list wont do much really. It's pretty moot nowadays with all the improvements google have pushed through lately, the bot is fairly good at optimising on its own for the most part.

Your metrics look really good, you've clearly done a great job for them thus far

To be honest, it sounds like the customers are a pain in the arse. Spend more money or stfu

Click to expand...

Yes. I thought as much as so. I tried used manual bidding but the maximise conversions has been working really well for me.

Thanks Perhaps so haha, I'm working on trying to improve the conversion rates for company C and see what we can do there.

Is there anything you could suggest here?

Shabzy said:

Yes. I thought as much as so. I tried used manual bidding but the maximise conversions has been working really well for me.

Thanks Perhaps so haha, I'm working on trying to improve the conversion rates for company C and see what we can do there.

Is there anything you could suggest here?

Click to expand...

I think you've got most standard CRO techniques down in your original post mate

CRO is a weird one, theres a bunch of standard things to do like speed, offers, reviews, UI/UX, social proof etc that if you're not doing can push huge returns, most of those are pretty obvious, require pretty much no A/B testing and is usually fast to implement. Once you've got that covered though it's a really hard and tedious road for that last 2%, which I find in most cases is not the most efficient use of time/$$.

I'd just get the basics down, anything super obvious and call it a day. It sounds like they aren't big budget, most likely not going to go for more advanced stuff tbh

jacksi said:

I think you've got most standard CRO techniques down in your original post mate

CRO is a weird one, theres a bunch of standard things to do like speed, offers, reviews, UI/UX, social proof etc that if you're not doing can push huge returns, most of those are pretty obvious, require pretty much no A/B testing and is usually fast to implement. Once you've got that covered though it's a really hard and tedious road for that last 2%, which I find in most cases is not the most efficient use of time/$$.

I'd just get the basics down, anything super obvious and call it a day. It sounds like they aren't big budget, most likely not going to go for more advanced stuff tbh

Click to expand...

Yeah we're in the process of putting some things in place. I'm really trying to see what I can do myself so I can take this information and use it on future projects.

What sort of stuff would you say is advanced for instance?

Shabzy said:

Yeah we're in the process of putting some things in place. I'm really trying to see what I can do myself so I can take this information and use it on future projects.

What sort of stuff would you say is advanced for instance?

Click to expand...

Once you start running lots of A/B tests I'd say that's fairly advanced. Running multiple iterations of the UI/UX, homepage layout, CTAs, imagery, user journeys and conversion flows, user behaviour/psychology etc etc

If you've cracked off the basics then it's really just everything else. Small minor changes in volume to push the last couple %.

I've spent 7 figs on CRO 'experts' and in my experience, they just guess based on what's worked previously, run a test, draw a very loosely correlated hypothesis that's almost never been tested enough to be definitive and deliver. It's always been underwhelming.

I haven't found one yet that actually delivered measurable results, solely attributed to the tests they ran, that blew my socks off. (theres some guys im sure that are fantastic)

I'm personally of the mind that, the last 2% isn't really worth undertaking unless you've exhausted literally every other angle, it takes alot of effort, time and capital. There's almost always more efficient ways to spend all 3 of those things, business wise, that will generate more revenue

It may be worth you though, plugging in VWO into your underperformer and just analysing some of the user flows of non-converted customers and see if you can pinpoint any pain points

jacksi said:

Once you start running lots of A/B tests I'd say that's fairly advanced. Running multiple iterations of the UI/UX, homepage layout, CTAs, imagery, user journeys and conversion flows, user behaviour/psychology etc etc

If you've cracked off the basics then it's really just everything else. Small minor changes in volume to push the last couple %.

I've spent 7 figs on CRO 'experts' and in my experience, they just guess based on what's worked previously, run a test, draw a very loosely correlated hypothesis that's almost never been tested enough to be definitive and deliver. It's always been underwhelming.

I haven't found one yet that actually delivered measurable results, solely attributed to the tests they ran, that blew my socks off. (theres some guys im sure that are fantastic)

I'm personally of the mind that, the last 2% isn't really worth undertaking unless you've exhausted literally every other angle, it takes alot of effort, time and capital. There's almost always more efficient ways to spend all 3 of those things, business wise, that will generate more revenue

It may be worth you though, plugging in VWO into your underperformer and just analysing some of the user flows of non-converted customers and see if you can pinpoint any pain points

Click to expand...

Thank you for this thorough explanation. I felt like I was doing something wrong when it comes to this especially when you come across so called gurus who have insane conversion rates.

With company C not really having a presence on Google in terms of reviews, I'm gonna start to funnel reviews to both google and FB, run some local citations to rank higher and then maybe tweak some things on the website.

I've added some different contact methods such as callback, WhatsApp and Facebook messenger.

I was thinking about pop ups but I've heard it can be intrusive therefore causing more harm than good. Maybe a banner with a limited time discount code will be a better option in this case.

Anyone who didn't checkout, we'll run an abandon cart email campaign to help those convert.

I think you need to understand what people are looking for.

im pretty sure that if they are at airport, they want quick taxi service and not to go thru checkout etc.

if its advanced booking you should target people outside that airport, before they arrive.

lets say, you have taxi booking in London
Target everything outside london and keywords taxis in london

so this way you should get people who will arrive in London and did booking before arrival.


I hope i did understand campaign correctly

Just to reiterate, if you are limited by budget there is room to decrease CPCs and therefore CPAs. You seem to be ignoring the most basic of supply/demand equations!

kdog said:

Just to reiterate, if you are limited by budget there is room to decrease CPCs and therefore CPAs. You seem to be ignoring the most basic of supply/demand equations!

Click to expand...

I don't understand what you're trying to say?

CPC's are based off ad rank so I have tried to improve landing page experience, wrote ads that correlate and I'm currently using Google maximise conversions for campaign instead of manual bidding.

I could reduce CPC by pausing words that aren't performing well but that means I'm potentially losing out on some search volume with those words if they are paused.

A lot of the words aren't driving crazy search volumes and are only getting one or two clicks per say so not enough volume to say for sure if they should be paused.

kdog said:

kdog said:

Just to reiterate, if you are limited by budget there is room to decrease CPCs and therefore CPAs. You seem to be ignoring the most basic of supply/demand equations!

Click to expand...

Click to expand...

That's not usually the case.

I've never had a campaign not be limited by budget and I run multiple campaigns with extremely large budgets

Nearly always the recommendations are counter intuitive to what you/the client actually wants.

If ROA is your aim and you have campaigns that run @ large ROA's, it will nearly always be limited by budget

The system will usually recommend an increase of 20% on max conversion campaigns (which perform the best), but that 20% will only return @ 1.5:1 or some other low number. In which case your budget is better spent on alternative campaigns

Essentially there's a sweet spot for max conv campaigns between budget / ROA.

Problem is that max conversion campaigns out perform ROA based campaigns (in producing ROA) by a country mile. So your giving the system mixed signals, your telling it you want maximum conversions, so it totally disregards ROA and will recommend that you increase your budgets to gain conversions at any cost. Which is generally, the opposite of what you want, especially if you're selling a product, what you want is ROA above all else.

Shabzy said:

I don't understand what you're trying to say?

CPC's are based off ad rank so I have tried to improve landing page experience, wrote ads that correlate and I'm currently using Google maximise conversions for campaign instead of manual bidding.

I could reduce CPC by pausing words that aren't performing well but that means I'm potentially losing out on some search volume with those words if they are paused.

A lot of the words aren't driving crazy search volumes and are only getting one or two clicks per say so not enough volume to say for sure if they should be paused.

Click to expand...

Have you looked at impression shares and individual conversion rates of keywords? This should make it very clear if there is room for more efficiency (limited by budget suggests this is the case).

If keywords that convert at efficient rates aren't getting over 90% impression share, then these should be prioritised over keywords that don't convert. This is generally what the automated bidding strategies aim towards, but if you split out and test your best keywords, you can generally find some value. You can also look at position on page for your top converting keywords, abs top/top/below fold makes a big difference to click share. Could be some wiggle room to boost conversion volume on some cheaper keywords. Lots of things to look at!

kdog said:

Have you looked at impression shares and individual conversion rates of keywords? This should make it very clear if there is room for more efficiency (limited by budget suggests this is the case).

If keywords that convert at efficient rates aren't getting over 90% impression share, then these should be prioritised over keywords that don't convert. This is generally what the automated bidding strategies aim towards, but if you split out and test your best keywords, you can generally find some value. You can also look at position on page for your top converting keywords, abs top/top/below fold makes a big difference to click share. Could be some wiggle room to boost conversion volume on some cheaper keywords. Lots of things to look at!

Click to expand...

What parameters would you say mean a keyword is doing well?
E. G.
>50% conversion rate
>50 clicks
>15% CTR

Something that would help me filter the keywords quickly