Well, THANKS, ebay---Results of PLS Ad Rate Forecasting Tool Test

[my-cottage-books-and-antiques]

This is the result of ONE test with the new "forecasting tool" (announced in the last seller update, and in gradual roll out right now), so, take it for what its worth LOL

However, I've been saying for a long time that sellers who chase ever higher rates for Promoted Listings Standard are probably giving ebay too much money (assuming they make a sale).

I suppose, depending on how you interpret these results, this confirms what I've been saying.

I just tried the new forecasting tool.

I set up (but never launched) a campaign for the 347 listings I do not currently have in a campaign---a very quick look at the first page of the listings suggests ebay provides suggested rates of about 9% to 13% for these items. My parameters were: Continuous campaign, Dynamic type with caps.

Here are the results of the forecasts with various caps:

2%   >   1%

3%  > 16%

4% > 27%

5% > 36%

6% > 43%

7%>50%

8%>55%

9% >60%

10% >63%

11% > 65%

12% >67%

13% > 68%

The following all yielded the same as 13%, which is to say, 68%:

15%, 17%, 20%, 25%, 45% and 75%.

Again, this is one test, your mileage might (probably will) vary, but a few TENTATIVE conclusions:

1. Going beyond the highest suggested rate is probably a waste of money...there's no real benefit in this case in choosing, say, 20% instead of 13%. The forecast is the same. Even for 75%.

2. ebay provides no benchmarks for us to interpret the actual value of these numbers (even assuming the forecasts are relatively accurate). Remember, the forecast is predicting impressions, NOT views or sales. So, at 9% I'm looking at 60%, at 13% I'm looking at 68%. How meaningful is that difference in terms of sales? Maybe not very meaningful at all. But it certainly is meaningful in terms of how much you end up paying ebay if you get a sale.

If I were actually going to set this up as campaign, personally I would probably choose 3 or 4% to see what actually happens with that percentage (in terms of views/sales). The very highest I would likely go is 7%, because I'm just not at all confident that the difference between, say, 50% at 7% and 68% at 13% is worth paying more for.

I don't have time to run other tests, changing the listings included, setting a duration, etc., but will try to do some more testing when I get a chance. 

[selsa84]

I did a few tests as well and the higher I went the worse the sales got for me. My nest sales days were ALWAYS below the suggested ad rate. I also noticed no matter what it was at the sponsored ads and see similar ads on my  listings never showed more than one of my items, even when I went double of the suggested ad rate. The system is broken and search being the way it is is also broken. I have bought a few items and before doing so searched all over eBay for cheaper ones and after check out it showed me 3 cheaper ones with the EXACT item title I used to search. 

[the-hook-and-the-loop]

My apologies, I hadn't taken my medication yet that the doctor prescribed for me for logging into eBay.

Interesting results. I imagine different categories will have quite different results. 

[simply-the-best-for-you]

@my-cottage-books-and-antiques  Thanks for the screen shot.  I DO have it, I remember seeing it when I set up my last campaign, last week, but I didn't click on it.  I appreciate the info, I'll have to take a closer look, esp since I plan to make some changes soon. 

[simply-the-best-for-you]

---

@the-hook-and-the-loop wrote:

My apologies, I hadn't taken my medication yet that the doctor prescribed for me for logging into eBay.

Interesting results. I imagine different categories will have quite different results. 

---

Exactly.  I wouldn't be so quick to say that high rates are foolish.  I think your category & what your competitor's are doing is very important.  I say this as one who sells in one of the most highly saturated categories. 

[my-cottage-books-and-antiques]

Boy, is my face red! 

Apologies all around.

While the numbers I provided were accurate, I had not thought through the implications of using Dynamic Ad Rate....so, when I said, anything above 13% didn't move the needle...well, duh, of course not, because with dynamic rate, ebay uses their suggested rate (if my cap is high enough). So, with 13% the high suggested rate (in this case), even if I capped at 75%, ebay was actually going to use 13%, and so the forecast percentage at 75 is the same as 13, since ebay would actually be using 13, not 75. (As best I can figure)

Today I tried another test, this time using fewer items admittedly (only the "recommended" items, about 232 items) BUT I chose fixed price, rather than dynamic. Again, it looks like average suggested rates were in the 9 to 13% range.

So, with these FIXED price rates:

2% > 1%

4% > 27%

9% > 60%

13% > 74%

15% > 80%

20% > 91%

25%>100%

Anything higher yields 100%.

So, I emphasize, this is just one test. But I'll offer this thought:

ebay has always said that suggested rates are sort of the "sweet spot"---balancing the increased chance of a sale with the cost to the seller. (Of course, ebay is NOT factoring in the other costs a seller has, including cost of acquisition).

So, here, ebay is basically saying 13% will get me 74%, which, in ebay's view, has a good chance of leading to a sale, while only costing me 13%. 25% will get me 100%, but that will almost double my cost just to go from 74% to 100%.

So, when using fixed rate, yes, a higher percentage will increase your forecasted percentage  (at least up to a point....even here, we see that there would probably be no point in going higher than 25% in this case, which gets us to 100% forecast.

All of which begs the question: the forecast is only about impressions, not views/sales, and so we're back to my original point, which is ---we have no way of knowing whether , say, 74% at 13% really provides a substantial benefit when compared to , say, 60% at 9%?  Or even 27% at 4%? 

When I get time, I'll continue to play around with this, and encourage those who might be so inclined to do so as well. 

My stuff is single quantity long tail stuff, and, when PLS first arrived, it did benefit me at a reasonable price. I'm much less sure of the value today, but will continue to both test the new tool and run real tests with actual campaigns. 

[my-cottage-books-and-antiques]

by the way, again, just looking at this one test, but ebay seems to be break down results as "modest, average, good and excellent" In this test at least, the Suggested Rates were in the "good" section...which indicates that "excellent" , while presumably beneficial, probably costs more than its worth. I'd guess ebay views modest as "why bother?" and average as "OK, we'll give you some credit, you're at least trying" and "good" is the general sweet spot area. 

[cardxcraft]

If I had access to the forecasting tool I would experiment but mine has an "NA" note.  

It would be nice to get some reproducibility across categories and from different users.  

When I create a new automated pl campaign, the default for me is always 9%, which is agrees with your data...which appears to be the greatest benefit.  

[mcdougle4248]

I ended my current campaign and started a new campaign to see this new tool in action. My results are in line with what @my-cottage-books-and-antiques has already reported here in this thread. To reach 100%, I had to take the ad rate up to 25%. A 25% ad rate will not work for my items. So, I settled for using a Dynamic campaign that is forecasted in the "Good" range at 9% - 15.8% ad rates. This new campaign will end on August 31st and may prove to be too expensive. I may need to reevaluate my strategy at the end of the month.

[my-cottage-books-and-antiques]

@mcdougle4248  One thing I want to stress is that there is NO evidence that, as an example, doubling the forecasted rate from 50% to 100% doubles your chance of making a sale. It's important to figure out what rate is acceptable to YOU, as a seller, looking at YOUR margins and YOUR sales velocity needs. That's why I recommend thinking about an overall strategy for each campaign. Just as an example, I personally might choose 4% to get a boost, even though it probably won't be a huge boost. But if I really need cash flow, or I really want to basically liquidate some old inventory, I might go higher (or I might just lower prices, run a sale, etc.) I might also go higher if I just want an overall increase in sales, but there are probably very few circumstances where I would go to double digits....but that's me, other sellers are dealing with other circumstances.

To me, PLS is just a tool, and I can use it or not, but if I use it, I want to use it wisely.