As an independent book publisher, I’ve had reasonable success reaching parents and readers through Amazon Advertising. Of course, I would like to reach more readers for Sean Yeager Adventures, and this is where you can help. Here are some free tips and more detailed notes for how to use Amazon Advertising more effectively. These are gleaned from many hours of using the tools Amazon provide in the highly competitive markets of selling UK and US children’s books. In exchange, please view a book page on Amazon for one of the Sean Yeager titles and see what you think. You can even use the Amazon Look Inside feature to read a few pages. Thanks in advance for your help. Yes, the Sean Yeager books are good. Check them out, there are also free sample chapter e-books!
Now without any more ado here are 20 focused tips and below them are more detailed notes.
In short:
Glossary: Term(s) means the search terms (e.g. ‘Harry Potter’) you load and bid for in a terms Campaign. Target(s) means the products (ASINs) you load and bid for in a product based Campaign.
- Experiment to work out which type of advert delivers the best results for your book(s) and strategies – products/categories, terms, brand products/categories or brand terms.
- Run three or four Campaigns of different advert types/targets in parallel, but no more
- Be careful in accepting Suggested Bid ranges especially high ones (above 0.40 cents/pence)
- Start low with your bids and review their performance for a few days/weeks
- Adjust your term/target bids by small increments (up and down) and check the results weekly/monthly
- Stick to your break-even average cost per click range (CPC) as far as possible. This may be as low 0.12 pence/cents depending on your click to sales conversion rate and your net income per book sale
- Tune your campaigns regularly (bids, terms/targets) using weekly, monthly, year to date metrics in the Amazon Advertising tool.
- Base bid adjustment decisions on your historical campaign metrics (Sales, Clicks to Sales, Clicks, CPC, Impressions, DPV, as well as target/term relevance)
- Prioritize terms/targets that resulted in sales but not to the exclusion of your break-even CPC or having a breadth of terms/targets – sales can be random and unpredictable
- Keep in mind seasonal demand and raise bids for your best performing terms/targets in Mar/Apr and especially in Oct/Nov/Dec
- Use a naming convention for your Campaigns with date, ad type, and tactics in the name
- Aim for 500 to 900 targets/terms per Campaign and expect only 300 +/- to receive Impressions
- Review Amazon’s Suggested targets/terms monthly and sense check the best ones to use
- Keep adding refined targets/terms based on related books, new releases, and new bestsellers, plus refined or related ‘Search Terms’ that resulted in clicks
- Keep your bid search terms as brief as possible – two to three words ideally
- Use Negative Terms to weed out unwanted Search Terms that led to ‘bad’ or irrelevant clicks
- Use the Amazon website’s search feature to preview what the search results are likely to be for your chosen search terms & adjust your terms to be key, stem words (e.g. ‘book 8 years’ not ‘books for children aged 8 years’)
- Select ‘expanded’ product (ASIN) targets for wider related Impressions
- Use a mixture of current, classic, high selling and lower selling product targets (ASINS) for coverage (it can be surprising where impressions are cheapest & effective)
- Experiment with tactics for a defined period – say a month – review, regroup and start fresh Campaigns if needed. Use bulk Archive, bulk Adjust Bids and then Copy Campaign to speed this up.
If you have made it this far and you promise to help spread the word about Sean Yeager Adventures (go on, you know you want to), here are some more detailed notes. They are wordy, so have a coffee on standby.
The long version:
On bid prices and Suggested bid ranges:
Amazon appears to apply all manner of rules in their advertising algorithm/engine. In theory, you are bidding for targets/terms against others. In practice, there are clearly tiers of suggested bids which have little or no relation to the results you will see for your book adverts. It is probable that the engine behind Amazon Advertising has many layers of adjustments and gives bid ranges as approximate quotes or even default ranges. Indeed, you can/will see the exact same target/term quoted with different bid ranges in two different concurrent Campaigns. Amazon support will not answer this question in a meaningful way – ‘your Campaigns are competing with each other’ was the explanation I received. To be frank, that answer side-steps the observable facts. Not all Campaigns are considered equal.
Consider the following example seen repeatedly on Amazon US (Jan ’23) – low, mid, high bids: $0.69, $1.42, $3.43. As a frequently repeating pattern, it is obvious this is not based on actual bids from advertisers. Instead, the engine’s code is repeating this suggested bid range for a reason. Either it has evaluated a Campaign’s worth to Amazon and applied a tiered rate or it has no idea what the current bid price range is for the given target/term in question. Or both. Either way, beware! If your Campaign is being set at a low priority it follows logically that you will be given high-end suggested bids. Whereas, if you experience a run on sales, you will see the suggested bids fall like a stone – probably because your Campaign is at a higher level of priority within the engine and Amazon wants to promote more sales of your book. Make the most of this period, it might be brief.
From my experience, targets/terms can attract good impression counts per day (100’s) at bids of 0.10 to 0.15 cents/pence. It does not always follow that Amazon browsers (customers) are actually looking in large numbers at the targets/terms with the higher suggested bids, and only actual impression/click/sale numbers – by day, week, month – will prove with evidence which specific targets/terms are delivering the best results for your adverts. Hence, experimenting is necessary to find the sweet spot for your adverts/books. Also, this sweet spot will change by season, especially in Nov/Dec when bids will increase due to more advertising spend. (The majority of book purchases being in Nov/Dec.)
High-range suggested bids for targets/terms (£1/$1+) could mean a mixture of two things – that many browsers are searching for them or that many advertisers are bidding high for them. (Or that Amazon wants you to spend more on them – cynical me). While it is tempting to bid high for bestsellers, it does not automatically follow that your adverts will convert better at higher bid costs and break-even for you. Therefore, you need to work out which, if any, of the high demand/high bid range targets/terms you can safely bid higher for in your Campaigns. Only experimenting will tell you this, and therefore you need to build up some metrics to analyse. Start with low bid prices – 0.10 to 0.20 cents/pence even for a high-end suggested bid range, and nudge upwards if you see potential from the resulting metrics (impressions, clicks etc).
Example – If the term ‘Harry Potter book’ converts well for your books at say £0.60/$0.60 bids, great. If not, you should reduce your bids (to say 0.12 to 0.15) or even cull the term from your Campaigns altogether. Here common sense comes into play – do buyers of Harry Potter want your book because it is sufficiently relevant to their needs/tastes? Again, evidence in your Campaign historical metrics will tell you the truth. While most children’s authors would like their titles to sell to the Harry Potter buyers (parents/fans), it does not automatically follow that this is going to be borne out by actual click to sales figures. If it does, you will know a clutch of targets/terms to prioritise right away. If not, find other targets/terms, and there are thousands.
On break-even CPC and bid adjustments:
While it is tempting to bid £1/$1 for impressions and clicks on specific targets/terms, the value of the resulting clicks to you does not increase based on how much you bid. Instead, you will struggle to break even at higher bid costs unless your book converts at a high rate of sales to clicks (2 clicks to a sale for example). Only you can work out what that conversion rate is for your books (per Campaign type and bid type), based on your historical metrics, and it will vary by season – Nov/Dec being the primary buying season and hence the most expensive for winning clicks.
In addition, for Product adverts that appear in the ‘Also viewed’ sponsored section of Amazon search results pages (the ones that scroll left to right horizontally with up to 50 pages of adverts): Yes, if your ad appears on the first page you will stand a better chance of clicks and higher impressions. And, yes, you will have to bid higher to achieve this placement. However, the same break-even logic applies – what is the cost of these adverts relative to their return from sales? Spending £100/$100 per month for these product targets/placements) and achieving 10 sales per month at £2/$2 net income per book sale is not going to break even. (Clearly, 100 – 20 = a loss of 80.) And your thinking needs to be this granular or you are not going to be in control of your advertising spend and cost effectiveness.
From my experience, twenty ‘good’ clicks can be needed to achieve one sale. At best, in Nov/Dec, I have seen ten to twelve clicks being needed for one sale. The break-even bid cost (CPC) for a click can therefore vary but is typically in the range 0.10 to 0.15 cents/pence unless your sale to click ratio is far better. This is also based on how much net revenue you receive for a book sale. £2/$2 +/- per copy being a typical net return for a KDP sold print / e-book.
However, you can still bid lower than 0.10 and higher than 0.15 for specific targets/terms (except in Brand ad types with a 0.10 minimum bid) and achieve an average click cost within range, provided you monitor the results day by day or week by week. Yes, that is a lot of work, but it is your money you are spending, and you are a business with the aim of being profitable over time. (You might want to chant that line a few times.)
A note on Amazon Campaign level Bid Adjustments – don’t use them until you are clear on their outcomes. And then use them sparingly if at all. An uplift of 100% could literally mean paying twice the price for the same value click without any guarantee that the browser (potential customer) was about to purchase. To be frank, Amazon does not know who will buy in a given browsing session (how could they?), but they do know how to entice the advertiser to pay more for a single click. The house always wins and you won’t get a refund as the advertiser if the browser clicks but does not buy at a £3/$3 adjusted bid per click!
On tuning Campaign bids per target/term
What do I mean by tuning? Let’s start with some assumed targets and build forwards. Let’s say we want 2,000 impressions per day for a Campaign, we are aiming for 10 clicks per day, and we budget a £1/$1 per day spend for this Campaign for a period that is not Nov/Dec. Clearly, you can scale this up based on how much you are willing to risk, and spend more in Nov/Dec. However, this is a good starting point to learn and gather historical metrics for your adverts/books.
We should tune our bid prices weekly and monthly based on what targets/terms give the best – 1) sales, 2) relevance, 3) impressions, 4) clicks, 5) DPVs – detail page views (for Brand ads).
Sales – it goes without saying, if you convert well (low clicks relative to high sales) at a bid price for a target/term (within the same year) you prioritise this target/term and keep monitoring. But remember, sales can be random events – it does not automatically follow that a target/term will always sell your product at a profitable break-even bid price (CPC), you may still need to reduce these ‘proven seller’ bid prices over time if they fail to deliver on a break-even level. Either way, note what sells in a document, Amazon Advertising holds Search Terms for 65 days.
Relevance – Amazon is built on concepts that include relevance. We can also conclude that Amazon prioritises revenue for itself – which books sell and which adverts make money for Amazon through the cost of advertising. Beware of the latter, the house always wins.
Relevance means – based on actual metrics, is a browser for X going to have an interest in your product? Has Amazon seen sales go through for your book and these targets/terms? If so, great you are beginning to progress and your suggested bid price ranges may reduce over time for the performing Campaign. If not, it will probably require hunches and experiments to find and establish relevant targets/terms and earn better prominance on Amazon as a whole. Remember, with good sales comes organic placements and recommendations by Amazon. Also, intuition and honesty come into play. Spending a lot on advertising for targets/terms that are not closely relevant to your book may yield a lot of impressions and clicks, but it is sales that you need! Hence, as a strategy, stay relevant in your choice of targets/terms on the whole.
Impressions – Your book(s) need to be seen to sell. They may sell today or a few days later based on a browser’s click or two. And seen means visible on the main search results pages for other products. Impressions can be misleading in numbers terms. Page 40 impressions on ‘Also Browsed’ are not a lot of use for your aims. However, each target/term needs to deliver at least double-digit impressions to be of any value for your aims. A larger number can be good, provided the target/term is relevant and specific enough. To be clear, if a target/term attracts only one impression and a sale, guess what? You’ll naturally go after more impressions for the same target/term. And this can happen, although there is no guarantee this focused tuning will attract more sales due to the randomness of what was in the browser’s mind at the time they searched for a book – i.e. this click/sale could have been a fluke. It is better therefore to build up a set of targets/terms that reliably give good impression counts week after week as the core of your approach.
Also, gaining thousands of impressions by targeting the latest best-selling book might bring visibility. However, how many browsers clicked? And did any buy your book? These kind of tactics can be morale boosting but without sales conversion they are a distraction from your core tuning. You can also argue (and I’ve debated this inwardly) – if these targets/terms cost little, where is the harm? My advice is to try a few of these but don’t rely upon them in a scatter-gun manner. And bid low for these high flyers unless you see a healthy sales conversion rate (sales to clicks).
Clicks – There are good clicks and bad clicks. With Search Terms you can tune out the bad click results for your bid terms and to a limited extent for products and categories. Each bad – as in irrelevant Search Term that did not lead to a sale – is a cost burden and needs to be limited or culled. You can use Negative Terms to eliminate the worst bad clicks. You may need to cull or reconsider the words in too broad a bid search term as well. e.g. ‘Children’s Book’ as a ‘Broad’ term may lead to all kinds on unwanted clicks – childrens toy, railway book, book about West Africa etc. Sometimes, the matched Search Terms for clicks can be bizarre. I suggest limiting the use of Broad terms for this reason. I tried ‘Warriors’ once and it led to thousands of matches for a war veteran’s book, not the cat fantasy book series. Sadly, the resulting clicks did not lead to sales.
Good clicks should be nurtured, within reason. Good clicks lead to sales or the likelihood of a later sale. You should review the Search Terms for clicks and assess – Was this a close match? Were they looking for a book like mine to buy? Did they buy my book later? Good clicks are to be encouraged within the average break-even range you have worked out for your campaign. And beware, not all clicks are of equal value to you. A click that converts to sales is gold dust, a click that does not is a cost of sales. If no clicks for a specific target/term convert within say 20 clicks, pause that target/term and review your advert and product detail page. Something is wrong and needs to change – either your product’s presentation and blurb is too weak, or the target/term is not relevant enough to convert sales and should be culled or reduced substantially in bid price.
DPVs – For Brand adverts, Amazon provides metrics of Detail Page Views (DPVs) You should review the ratios of clicks to DPVs. Those clicks with low DPVs and low sales should have their bids reduced or culled. Because – if the browser did not proceed to review your book’s product page, they were not attracted enough by your brand advert to find out more. Either your brand advert was not interesting/relevant enough for the browser who had in mind that specific search target/term, or the browser was not minded to explore your brand further for other reasons. Obviously with adverts we want to attract good clicks and interested potential buyers. DPVs give an insight into which brand terms/targets are encouraging near misses. By the way, a target/term that converts sales will typically have a DPV value equal to or higher than the number of clicks it attracts.
In summary, by tuning your campaigns, weekly and monthly based on actual historical campaign metrics, you will gain insights to target more effective advertising – higher sales revenue, for lower average click costs, and lower costs per month. In the longer-term, this will help you reach break-even for your adverts/book sales. Clearly, if your book takes off in sales terms, you will be able to flex/loosen your controls. Whereas, if you achieve minimal sales, you may conclude after several months that your product is not commercially viable for advertising on Amazon Advertising and cut your losses. Either way, tuning is about gaining control of your campaigns instead of a splatter-gun approach of spraying adverts around, paying higher CPC amounts, and praying for sales.
I sincerely hope these tips and notes help you with your Amazon Advertising.
D M Jarrett
Sean Yeager Adventures - hidden gem exciting action series 
You must be logged in to post a comment.