Brand Bidding - When You Should And Should Not You Consider Doing It

A question that will always be asked of performance marketing teams is ‘should we be bidding on brand terms?’ In the below we'll address when you should and shouldn't consider brand bidding and how you can test if it's driving profitability for your business.

DSC 6947-2-e1552328249858-o4q6qk2rk1g5eqjxuxuzeft2l01anq2sr27mw5e2moby Bethan Rainford 2 July, 2020

A question that will always be asked of performance marketing teams is ‘should we be bidding on brand terms?’ That is, when people search for your brand name in Google or Bing, is it worth having one of your paid ads show? It’s an easy - and good - question to ask but a harder one to answer. At its root, it stems from the reasonable desire not to pay for traffic that would have reached your website anyway. Most search channels will be set up so that brand bidding is on, and so the question is mostly framed in a way that wants to discover if you can increase profitability - in other words, is the money saved on clicks greater than the amount that might be lost on any downturn in sales (if there is one).

As we’ll examine later, it is quite a difficult test to run either in parallel or sequentially, which is why there is no consensus right answer (which there shouldn’t be; it will differ from band to brand). Before we get to that, however, let’s examine some cases where it is worth considering a test and when it’s better just to stick with what you are doing.

When should I bid on brand terms?

Testing turning off brand campaigns has an opportunity cost - it takes focus away from your performance marketing team at the expense of other optimisations, and to work well involves not running major sales, offers or similar merchandising efforts for up to a month. There are some scenarios, then, where it’s worth pushing back on the request in favour of activities that will drive growth in other ways. These include:

Your brand is new / not well known

If you have a relatively new brand, then bidding on your brand has a few beneficial effects:

  • People may not yet spell your brand name consistently, and Google may not yet pick up on spelling mistakes, so bidding on variations helps stop any leakage

  • If it’s the first time someone has searched for your brand, it provides them with the reassurance that they have found the right link. This is especially true if your URL isn’t just your brand name, but you’ve needed to add another term to secure the domain such as ‘-software.com’, or ‘-activewear.co.uk’. 

  • Studies have shown that the act of spending on advertising provides reassurance to customers that the brand is legitimate, something that’s especially useful for new brands 

Brands with similar names already exist

It’s very unfortunate - and unlikely! - that you have a similar brand name to a competitor in the same market as you. However, it’s very possible that there is crossover with a brand from a different market. For example, ‘Evans’ could refer to Evans Cycles and Evans Clothing - not in competition, but both being returned for the same search. In these circumstances it’s worth giving people a path at the top of the search return page to click through to the site they are actually looking for. It’s a very low risk strategy - people looking for the other brand will simply click on their ad instead and you pay nothing.

You’re not the first brand in the organic listings for your brand search

There’s a number of reasons that this may happen - you may have recently migrated your website, you may be having issues with your website’s SEO or it may simply be a case that you are a small brand being carried by a large retailer and their pages are listed first. Similarly to the previous example, in this case it’s worth bidding on brand terms to ensure that potential customers have the opportunity to click through to your website

Other businesses are bidding on your brand terms

This may be totally legitimate (see the example above where a large retailer may carry a smaller brand) or more nefarious, but either way, the presence of other business bidding on your brand terms can cause confusion in your customers and lead to them ending up on a website other than yours. It’s worth considering the levels of this - if it’s a few random listings from broad match campaigns it’s probably not much of an issue, but if there are consistently ads popping up then you need to be conscious of this.

The final advantage that paid search listings give you for brand searches is brand extensions which are controllable in a way that they’re not for organic listings. If you are working for a company that has offers and / or specific customer messaging that you want to use, then there can be excess value created - this should be evident in any test that you run.

Brand Bidding Medic

When should I consider not bidding on brand terms

So given what we know, when does it make sense to consider a test?

You already dominate search space

If your brand is well established and the organic search listings are filled with your domain, extension and social listings, then you have an opportunity to try to test turning off your brand terms, especially if you rarely see other brands popping up on the paid side

You have a low AOV (or low ROI on brand terms)

If you have a high average order value, then it’s unlikely that the effects of stopping spend on brand terms (which are generally pretty cheap) will have much impact on the bottom line. However, if your ROI (and per order profitability) are lower, then the cost of media makes up a larger percentage of your marketing profit and loss. The larger the slice that you are trying to reduce is, the more important - and more effective - it is to try to reduce it.

Brand terms make up a large share of your search spend

If a large share of your search spend is going on brand terms, the greater the opportunity you have to recycle some of that spend further up the funnel, whether that’s social or more generic search activities

There is a high proportion of existing customers clicking through paid search

If someone has already bought from you, the considerations around brand awareness are much diminished - if they are clicking on paid search ads it’s likely that they are doing this because it’s the first thing they see. There is a strong argument for not bidding on brand terms for audiences of existing customers - remarketing lists for search ads (RLSA) is very helpful in this regard.

How do I test turning off brand terms?

Broadly speaking, you need to run an A/B test. Assuming that you are currently spending, this involves turning it off and then turning it back on again at a later date. Even with RLSA, it’s very difficult to run a side by side test, especially as most of the value will be coming from new customers.

It’s also important to run the test for long enough to build up enough data and for customers to go through a full purchase cycle - in a fast moving goods environment two weeks should cover both requirements. 

This is where it gets tricky - demand levels are not static, but change by week in the month and month in the year. Additionally, you need to pick a time to do this where the business is comfortable not changing on site promotions - if one test group sees a more attractive set of offers, this can easily skew your data without being an effect of the test itself. 

On top of this, not all organic traffic is tracked as organic by Google in Google Analytics - a significant proportion will register as direct, or no source / medium. 

The best way to approach this, then, is to try to minimise the noise in the data and hope to see a signal that gives you an answer one way or the other. For a start, focus on the aggregate value generated by organic + direct + paid channels that enter through the home page. Then, assuming that average order value is fairly constant, you can normalise this against any changes in conversion rate from other channels and run this through your A/B testing calculator of choice.

Conclusions

If all of this sounds like it comes with a lot of uncertainty, it does. It’s a difficult test to run without creating a full econometric model (which comes with other issues), and frequently you will find inconclusive results. For this reason, it helps to have a strong feeling beforehand - based on the reasons above - as to whether you think your brand bids are creating excess value or not.

DSC 6947-2-e1552328249858-o4q6qk2rk1g5eqjxuxuzeft2l01anq2sr27mw5e2mo
Bethan Rainford
Bethan Rainford is the Paid Media Team Lead at Vervaunt and runs the day-to-day operations for the paid media side of the business. Bethan has spoken at a host of eCommerce events and has been working in paid media for over 5 years, previously managing accounts for the likes of BMW, Nintendo and MUJI.