Conversion and PPC: Can you start small?


While some customers (and friends!) are ready to go out and spend gobs of money on their pay per click campaigns, I don’t usually hear that. More often, I hear, “I’d like to start very small, learn what works and what doesn’t, and then roll out in a large way.” It sounds like a great idea, but it doesn’t usually seem to work — in my opinion. Traci Scharf, our pay per click (PPC) specialist, disagrees, so after I write, she is going to do a rebuttal (and you’ll get to see both sides of the problem.)

Robbin’s opinion: It’s hard to start small with Pay Per Click:
So you want to start small with your pay per click campaigns and roll out after you know what works? Here has been my experience:

To make the numbers easy, let’s say that a click for the customer we are working with is $1.00, and his conversion rate for visits to the site is .5%. May be he wants to spend a million dollars eventually, but up front, he is starting with $100/day. He starts wtih 500 keywords and multiple ads.

If a click costs $1.00, and his budget is $100/day * 30 days, he has a monthly budget of $3000, i.e. 3000 clicks/month. With a .5% conversion rate, that’s 15 orders.

Those 15 orders will be spread over many keywords. There may be two keywords that got three conversion each, and three keyword that got two orders each, and another three keywords that got one each. Some of those may be branded keywords, too, like “LunaMetrics conversion rate.” When someone does a branded search, they are already looking for you (a topic for a different post.)

So what can we learn from this? My answer: just about nothing. We don’t have enough data to be able to say, “This keyword does well,” or “This keyword does poorly.” If everything is coming from the same AdGroup or campaign, we may be able to learn more there, but my experience has been that we generally learn what we already know — which products, or which areas of the site, draw the most visitors. Which products tend to sell the most. Whereas real learning would be, “When we use exact match on these five keywords, we have a higher quality score and get better conversion for less money, but on this AdGroup, we can’t get the kind of traffic we need with exact match, and so we need to use a different kind of match” (for example.) Or even, “These keywords suck!! We have to retool this whole campaign.” Now, that’s learning.

OK, Traci, your turn.
Traci’s Opinion: It’s Really Important to Start Small with PPC
I’ve always felt that limiting your initial spending in a PPC campaign is a smart move for most businesses. Let me give you a little analogy:

Say you want to lose weight, and someone says to you, “Hey, you can lose a lot of weight on the peanut butter diet.” You might be willing to give it a try, but probably don’t want to invest too much of your time and energy until you know whether it works for you or not. A rational decision, then, might be to try it just for three weeks, and then get on the scale and see whether your weight went up or down.

When client companies say they want to learn small, this is in effect what they’re proposing. They’re saying, “Let’s spend a set amount of money, and see if our investment is getting a positive ROI.” Because, just like a diet (“Are you losing weight or aren’t you?”), PPC is pass/fail (“Are you making money or aren’t you?”). Maybe you know that, given where your initial budget is set, you need 15 conversions to break even on your PPC campaigns over the course of three weeks. If you find that your campaigns are getting 30 conversions over the course of three weeks, you’ve learned one big important piece of information: “You’re making money – so go ahead and increase your budgets.”

But probably the most compelling reason I encourage companies to limit their initial ad spend is because they’ll want to have enough money to act on what they learn. Consider the company that runs PPC campaigns for three weeks and finds out they are getting, on average, three conversions a week, but need to be getting ten/week in order to break even. Well, assuming I’ve done my job in setting up their campaigns to drive qualified traffic to their site, we will want to look at what is going amiss with their landing pages. That will mean doing a best practices analysis on their landing pages, and then creating and testing alternate versions so that we can transform their conversion rate. However, if they’ve already blown their whole PPC budget, there is no place to go from there, except to cut losses and admit defeat – not the best strategy for getting ahead!

So the bottom line is, be leery when anyone tells you that you have to spend a lot of money in the beginning months of your PPC campaigns. Exercise the same caution you would with anything else, and remember that you can’t just throw money at PPC and expect success.

Our founder, Robbin Steif, started LunaMetrics in 2004. She is a graduate of Harvard College and the Harvard Business School, and has served on the Board of Directors for the Digital Analytics Association. Robbin is a winner of a BusinessWomen First award, as well as a Diamond Award for business leadership. In 2017, Robbin sold her company to HS2 Solutions and has since retired from LunaMetrics.

  • Guys, I totally agree. For me this is one of the most frustrating things about working with a new client. (And in Australia $50/day is a stretch for some smaller clients)
    Two points.

    One. Which keywords do you choose to start? Do you do a complete & full setup of the account. Multiple campaigns, hundreds of adgroups? Or just the ones that are likely to work quickly.
    And how much to charge for this? Do you charge for a full setup? Or do you ‘scale in’, charging a small amount for a cut-down account, then another setup fee if/when it works & the client sees the benefits?

    Two. So much depends on factors outside of our control, but because PPC can be measured to the dollar – the finger gets pointed there for anything less than stellar results.
    Offline media, website usability/copy/nav/call-to-action all have a massive impact… and if the client is smart enough to measure conversions, some education is needed on why there may be a problem.

    I’m interested to know how you deal with these issues internally. As we grow & move up the chain to larger, hopefully more web-savvy businesses, it seems a lot of internal processes need to change. Luckily it’s the best job/business in the world, so it’s all good fun!


  • He he 🙂
    100$, 50$ a day. The majority of our clients want to start with under 20$ a day! Some feed google just with 10 dollar.
    What do you think is the minimum $ to spend in 30 days?

  • I agree with Robbin, but for a different reason. The problem with starting small is that you quickly run into diminishing returns. Say you’re spending $10 / day and on PPC and getting $100 in sales. You show the boss a month’s worth of this and he says “ramp it up!”. So the next day you spend $1 billion dollars and get $10 billlion in sales, right? Obviously not, somewhere in there the efficiency of the program drops off. There’s no way to tell how it drops off without spending the money.

  • Mike-in-Australia, we recently ran into that issue (it was the site, not the campaign) and told the customer that we didn’t want to spend any more of their money until they fixed the site (and now, we are actually working on that aspect instead of the PPC.) So we didn’t have to fire the customer, but it might have ended up that way, and it would have been sad, because they are a great customer.

    Sebastian, I am of course biased toward spending more money, but I don’t think it is about dollars, it is about clicks and conversions. So if you are getting keywords for 25 cents (and sometimes I do, especially with heavy industry), that is a different story than keywords that cost $5 each. And even with keywords that cost $5/click, I have seen them convert beautifully, so unfortunately, the answer is that almighty catchall: when it comes to the minimum per day: it depends.

    MikeK, I so miss your blogpost, and I still depend on your WA detector, even though I have two others on my browser. Diminishing returns, that’s a good thought, wish I had thought to include that one…

  • If the client has the ability to spend a lot of money on PPC, then I definitely agree with Robbin. The greater your spend, the greater the amount of tests you’ll be able to run. As you said, Robbin, this will give you more credible data so you know what keywords work and which ones don’t. Limited budgets make it incredibly difficult to fully optimize your campaign.

  • Robbin,

    That’s a great point – it’s the quality of the click, not the quantity of them!
    One of the big players here (names deleted to protect the stupid) have started selling a guaranteed number of clicks per month for a set fee. They say that will get the client on all 4 (!) major search engines, but of course it’s a huge benefit to propping up their own ad space.
    The problem is with their salesforce & clout in the market, they’re training small businesses to expect a set number of prospects for a set budget. If only it was so easy! Who can tell in advance which of the 5000 keywords will have the best conversion rate & how much Google will actualy charge (Google is ~90-95% of the search game here in Oz.. according to them)
    Ah it’s a fun game we must play to get the best for our clients 🙂

  • Ravi Pathak

    I think both Robin and Traci has the right point.

    The Customer segment you guys are trying to build your case on are different.

    Traci’s argument is more focus towards the first timers who are just trying to experiment. While Robin’s argument (may still be applicable to first timers) is more towards those customers who have gotten flair of internet marketing or may be few campaigns already.

    however , I think Traci remains correct for those first timers as to decide whether investment drives enough ROI or not. Or is it even covers the cost or not ? If that doesnt make sense i don’t think I would want to get into PPC.Even if few KWs shows potential, i would want to test and re-test with smaller amount over period of time to see what works. May be time duration will be longer but then advertisers become comfortable over period of time.

    Robin’s argument is certainly right, but it may be for those who have tested PPC or may be tested different ways in which they can generate traffic. To get someone to believe without numbers to back it up may be real tough. However Robin, you are more than capable for that :).

    Very Enriching post , tempted even someone lazy like me to comment.

  • I think starting small is generally a good idea – and I typically ignore all of the pings from Google telling me that I’m missing out on clicks. It just feels like they want me to spend more of my client’s money when I haven’t had a chance to fully evaluate. I also turn off the option that allows Google to determine what the best ad is – they usually do it way too fast – like after a few click-throughs. The problem is that you get this hugely distorted amount of impressions on one ad, and nothing on the other ones…I’d rather determine which one performs better with equal impressions, thank you very much.

    Have you guys had any success with driving PPC visitors to a landing page with live chat?


  • @Ryan – I completely agree with this – when you’re starting out in PPC and Adwords, you have to remember that Google makes money when you actually buy ads so you have to watch out for their “suggestions” because this will result in you spending money on clicks that aren’t relevant!

Contact Us.

Follow Us



We'll get back to you
in ONE business day.
Our Locations
THE FOUNDRY [map] LunaMetrics

24 S. 18th Street
Suite 100

Pittsburgh, PA 15203


4115 N. Ravenswood
Suite 101
Chicago, IL 60613


2100 Manchester Rd.
Building C, Suite 1750
Wheaton, IL 60187