InsightsAutomation

How to Build Your Own In-House Paid Search Technology Stack

Written by Wes Parker
Co-founder

Posted on: August 20, 2020
Bio

You’ve taken a leap of faith and decided to build out your own in-house team to run your paid search activity.

The next step is to build the right technology stack to support your in-house team and ensure that it can manage large-scale ad campaigns effectively.

Looking at the chart below, it’s clear there is a bewildering array of PPC tools to help you improve performance, some more useful than others.

We’re first going to talk about just how much marketing technology you should use to get the best results.

Then we’ll take a look at the different types of paid search management software you should consider and list the core types of software every in-house marketer should use.

We’re also going to highlight some of our favourite PPC management platforms to give you some suggestions on where to start.

I should, at this point, note that we’re not affiliated with any of the software companies mentioned in this article.

How much marketing technology should PPC marketers use?

How much is too much when it comes to paid search technology?

Being the tech enthusiasts that we are, we would argue that you can never have too much technology.

However, in reality, there are pesky things like budgets that get in the way.

How much should you spend on martech?

When it comes to how much you should spend on paid search technology, Avinash Kaushik suggests that you use what’s called the 90/10 rule, where you invest 90% of your budget in media buying and 10% in technology that will help you.

This will, however, vary quite considerably, depending on the size of your media budget.

10% of budget towards technology might be understandable for a media budget of £50k per year, considering that most of the paid search management tools would cost £5k per year in that bracket.

As media budgets grow to over £1 million per year, spend on technology might be closer to the 2-5% region, where you are spending 3% on bid management tech, then 1-2% on split testing, ad fraud, auditing and optimisation tools, for example.

How many tools should you use?

Brad Geddes suggests in his webinar Simplilearn Webinar by Brad Geddes: Time & Tool Management For PPCers that PPCers use around 14 tools on average.

It is worth noting that this does include tools for collaboration and project management, which, however important (we couldn’t operate without Trello), wouldn’t necessarily be called PPC tools.

As a general rule of thumb, from our experience of using dozens of different paid search marketing tools, we suggest using between five and eight tools that are specific to PPC, alongside the stack that you use for collaboration, such as G Suite, Trello and Slack, for example.

Bid management

If you’re running an account with tens of thousands of keywords or more, you need to use some form of automated bid management to keep on top of things.

Most bid management platforms, such as Marin, Kenshoo, AMO (Adobe Media Optimiser) and DCM (Double Click Manager), use rule-based, positional-based or portfolio-based algorithms. You can read more about the science behind bid management platforms here.

Generally, if you’re optimising within a fixed budget, using a portfolio-based algorithm provides the best results.

This is where keywords are clustered together and the software allocates budget to the best-performing clusters first to maximise revenue or conversions of the overall portfolio of a group of clusters.

If you’re targeting a set cost per lead or ROAS without a budget cap, using bid management software that has a rule- or positional-based algorithm works best.

If you’re finding that you’re falling below the minimum requirements for the four leading bid management software providers, you have three main options:

  • You can use the smart bidding options within Google that we’ve found to be very effective when using TCPA, TROAS and maximise conversions. This should generally be your go-to if you’re falling below the minimum requirements.
  • You may want to instead consider building your own bid management platform using Google Ads scripts. Dan talks more about that here. If we don’t have the volume to use smart bidding, this is our preferred option. Here is a list of 120 Google Ads scripts that you can use to automate your bidding.
  • You can look at some of the lesser-known bid management tools, such as Acquisio or AdScale.

Auditing software

If you’re running an account with tens of thousands of keywords and ads, you should consider using auditing software.

Auditing software allows you to pick out issues within your account and benchmark your setup against best practice in real time.

For example, you could set up a report that flags all ad groups that have more than one keyword if you’re using SKAG (single keyword ad group) structure.

Another report could be flagging all ad groups that only have one advert and are not running a split test.

Two popular products that you could consider are Search Squared and GOA.

Split testing software

Keeping track of split tests when you have a few hundred ads within your account isn’t too difficult.

But when you have a few hundred thousand ads in your account, that’s a very different story.

This is where ad testing software, such as Optmyzr or Adalysis, really comes into its own.

Such apps not only allow you to track when thousands of ad tests have finished, but can even allow you to queue up ad copies to be tested once an ad copy test has finished.

Shopping stack (feeds/builds)

With shopping becoming ever-increasingly competitive, making sure you have a fully optimised feed isn’t optional.

Feed optimisation software allows you to optimise your product titles and descriptions, as well as flagging missing fields at scale across tens or even hundreds of thousands of SKUs.

From our experience, we’ve found that Shoptimized works well as a hands-on solution if you’ve got a limited budget, but Feedonomics is the gold standard.

Once you’ve optimised your feed, you’ll want to build out your campaigns.

For the best results, we generally suggest having one ad group per product to give you maximum control over bidding and negative keywords.

You can technically build out this structure with a spreadsheet, but when you’ve got a huge inventory, this can be a challenge.

We’ve found Optmyzr to be a great tool to build out these complex structures automatically at scale.

Reporting

Reporting is a boring, repetitive task that should always be automated.

When it comes to creating reports for management, if you have fairly basic reporting needs, we suggest that you use Google Data Studio, which is Google’s free reporting dashboard, and accompany it with a detailed writeup for the week’s or month’s activities and actions.

If you’ve got more complex needs, you will want to use third-party reporting for your Google Ads. We’ve found Supermetrics and ReportGarden to be good tools that you could consider testing.

They allow you to collate data from different search and social platforms, as well as enabling you to create bespoke reports to visualise the key data points for your business.

Click fraud prevention

Data by Pixelate suggests that click fraud on desktop devices could be as high as 24%,with figures from CNBC suggesting that around 20% of digital ad spend is wasted to click fraud.

Although click fraud prevention is widely discussed in the world of programmatic media, buying it also needs to be addressed in other channels, such as display and, to a lesser extent, search and social.

Click fraud prevention technology works by pasting code on your website that uses an algorithm to determine if the user on your site is a human or a bot.

It then adds the IP address to a block list within Google Ads, so that your ads never appear for that search again, reducing your losses to ad fraud.

In terms of ad fraud technology, you could consider tools such as PPC Protect or Click Guardian.

Bespoke software builds

When managing large-scale complex accounts, situations arise where you may want to build your own custom PPC software to automate repetitive tasks.

An example might be an FMCG brand, such as Just Eat, that has to create ads every time a new restaurant is added.

This is a prime example of when you should build your own PPC software to automate this repetitive task.

Building your own software also gives you the ability to integrate external data via APIs with Google Ads to make more complex bidding decisions.

For example, we used external crime rate and house price data to bid more aggressively in postcodes where house prices were above average, and there had been an increase in burglaries, for a high-end monitored alarms and CCTV client.

If building software isn’t your core business, you need to decide whether you should build your own tools or pay a monthly fee to rent them. To help you out, Baremetrics has created a handy calculator that you can use.

When it comes to building bespoke tech solutions, there are a few options:

  • You can fully outsource development of your platform by working with an agency that has its own technology department and an understanding of Google Ads (such as DemandMore). This is likely to provide the best results.
  • You can use code builder tools such as Betty Blocks, Google App Maker and WayScript to create your own Google Ads scripts.
  • If you’re going to be using a lot of automation and can afford it, you can hire your own in-house paid search developers, like those that Just Eat and Trainline have, which run very large-scale complex accounts.

Why we wouldn’t suggest investing in competitive intelligence tools

An article on PPC tools wouldn’t be complete without a mention of competitive research tools.

Although competitive research tools can occasionally find a keyword that you might have missed, for the vast majority of time when we have tested the different ones, we have found the data to generally be fairly inaccurate, so suggest steering clear of them.

Conclusions

To get the most out of your in-house team, you need to empower it with the right tech stack that allows it to efficiently manage large complex campaigns at scale.

As a bare minimum, I suggest implementing automated bid management and account auditing on search accounts, and feed management for shopping campaigns.

For those looking to get the best results, split testing software and their own custom-built solutions will allow them to get ahead of the pack.