Am I doing this right - PPC calculations

Need help with PPC maths

[pavlovapete]

I'd like to try PPC to sell our magnificent widget at the low price of $9.99. We have allocated a marketing budget of $2.00 per widget.

CPM looks no good - maybe PPC would work. Am I doing these calculations correctly?

CPM
Assuming .1% CTR
- per 1000 impressions, 1 click

Assuming 5% conversion rate
- per 20 clicks 1 sale

20,000 impressions, 20 clicks, 1 sale

Assuming CPM of $10.00
- I would need to spend $200 to get one sale
- Obviously not profitable

PPC
Assuming 5% CTR
- per 1000 impressions 50 clicks

Assuming 5% conversion rate
- per 20 clicks 1 sale

Can only afford a CPC of $0.10

Throw in a currency conversion rate from AU to US and the USD CPC has to be around $0.06.

This seems like an unrealistic hope.

Have I done these calculations right? Thanks for your time.

[Rehan]

The calculations are right.

The actual CPC and conversion rate will depend on lots of factors, but it's certainly not impossible to have a successful campaign with the marketing budget you mentioned.

[RhinoFish]

Assuming CPM of $10.00
- I would need to spend $200 to get one sale

Your CPM calcs all look correct to me. $10 is probably on the medium to high side of available CPM rates, but of course, that depends on your niche and aggressiveness and other factors. Typically though, CPM ad rates aren't ideally suited for ROI centric advertisers (I avoid them like the plague). Branding, name building, reach, etc, can all work well in a CPM environment. Sometimes, ROI can be decent, but it's not commonly the case. In addition, impression fraud is much harder to detect / avoid than click fraud, making CPM generally a lesser opportunity for ROI-centric folks. This is just my opinion and a very generalized one at that.

PPC
Assuming 5% CTR
- per 1000 impressions 50 clicks

Assuming 5% conversion rate
- per 20 clicks 1 sale

Can only afford a CPC of $0.10

In your CPM calcs, you finished by deriving the ad cost per sale, I didn't see you clearly take that last step for PPC... perhaps you worked backwards from your profit margin, I wasn't sure from your post.

For ROI calculation purposes, impressions and CTR are irrelevant to the PPC math.

If you set your bids so that your CPC costs are $0.10 per click, those 20 clicks (to get one sale) would cost you $2.

So, with the numbers you've assumed, your CPM cost/sale was $200.
And, with your numbers you've assumed, your PPC cost/sale was $2.

Would a $2 PPC cost per sale leave you with a profit?

[edited by: RhinoFish at 3:25 pm (utc) on Feb. 5, 2009]

[eWhisper]

Just want to toss in a few other thoughts.

The price is $9.99.
Or could it be different in your calculations:

Do people buy only 1 or multiples?
If one, can you show them why they need 2 or 10?

Do people buy multiples throughout a year?
If no, can you remarket to that audience via a loyalty program?

Is it an intro product that they can add-on to (cross-sell, cross-promotion, or upsell)?
If no, are there complimentary products right now that you could offer to increase average sale amount?

While your initial sale price is$9.99; you might want to think about ways where the lifetime value of that customer can be increased (which will help out with the math).

[Green_Grass]

Assumption of 5% conversion rate has to be tested.

You may get 10% ;-)

or closer to 1% :-(

[pavlovapete]

Thankyou all for your considered responses.

$10 is probably on the medium to high side

Some of the properties I've researched want 2-4 times that cost. In AUD. They are being sold as branding vehicles (which I cannot comprehend as the ROI seems elusive and immesuarable)

perhaps you worked backwards from your profit margin

Yep

Would a $2 PPC cost per sale leave you with a profit?

Yep

For ROI calculation purposes, impressions and CTR are irrelevant to the PPC math.

I don't understand this. If the CTR is high but 0 conversions then I've lost money. Have I missed something?

lifetime value of that customer can be increased

Admittedly I did not consider the LTV. In isolation, it is a single SKU once a lifetime product. Unfortunately our catalog does not include many relevant, associated products. However, you have provided the catalyst for an idea - a list. Thanks for that.

It seems that LTV bidding is riskier unless you have some firm data/ experience behind you. I can see that it would be easy to fall into the trap of spending big on the assumption that the LTV ROI would be positive. Any way of fast-tracking this type of research to get to a (semi) conclusive answer without waiting a period of months/ years?

Once again thankyou all for your input. It is a great relief to confirm that I have grasped the basics.

[Yoshimi]

you only pay when you get clicks, so whether yur CTR is 1% or 10% if you are getting 10 clicks you pay for 10 clicks.

I'm doing this calculation for a campaign this morning, The product generally sells for £300, we have a set cost of £40 CPA (that is we can spend £40 to aquire the sale)

I know from previous campaigns that our conversion is around 2%

Which means that for every 50 clicks we will get one sale, meaning that I can spend £40/50 clicks = max cpc £0.80

Once I have the max CPC I can work out what terms I can bid on and where in the rankings I should aim to sit. That then helps me work out how mant impressions I will get (based on the positions I will be bidding on) and from there I can predict sales.

Does that help you?

[pavlovapete]

Hi Yoshimi,

thanks that does help.

Cheers