Welcome to Curious Business

Every Friday, I post a small insight into running Curio City and/or Blue Hills Editorial Services. My most recent posts are directly below. You can also start with the first post, or use the subject labels to the right to home in on particular topics. Feel free to comment on anything that interests you.
Add to Technorati Favorites

Friday, September 30, 2016

The Good-Bad, the Bad-Good, and the Pretty Ugly

This month's numbers look good because September 2015 was so weak. It's a clear win when income goes up and advertising goes down. Coming in as the 9th-best/3rd-worst September ever doesn't mean much because this September had only four full weeks, versus five last year. The QuickBooks numbers below are a better apples-to-apples comparison.


Total income: +26.8%
Total COGS: +31.4%
Payroll: +44.7%
Marketing: -9.7%
Net Income (Profit) vs LY: -1.9% (-$8)
Actual Profit/Loss: -$451


Total income: -6.7%
Total COGS: -5.9%
Payroll: -6.4%
Marketing: -4.9%
Net Income (Profit) vs LY: -19.3% (-$392)
Actual Profit/Loss: -$2,417

I haven't had a single sale since my developer switched my site from "http" to "https" and implemented the new AdWords script (see below) on Tuesday. I get nervous when a big drought follows technical changes to my site. Sure enough, last night I noticed that Firefox was showing my product pages as "partially secure" and displaying a yellow warning triangle. Uh-oh. It turned out that my reCaptcha plugin was broken, and the Reviews tab has a Captcha to thwart bots. Fixed that and the lock icon turned green. I still haven't had a sale, though, so I'm not sure that was *the* problem...but it was definitely *a* problem. I'll feel better if/when I get a real sale...the four-day drought makes this the worst week in, well, maybe ever.  
Something good-bad happened: I finally hired my developer to implement the AdWords value-tracking code that I couldn't do myself (and that Google wouldn't help me with). That's good because it will allow me to control one of my biggest expenses more effectively over the long haul; it's bad because I had to pay up-front for something intangible. As it turned out, implementation required several hours of reverse-engineering to discover some necessary Sunshop variables, so Google's support wouldn't have been able to fix it anyway. It would have been nice if they'd told me that and saved me six weeks of grief. 

Then something bad-good happened: A product that caught my eye led me to a new vendor with dozens of products that fit into my unusual/useful philosophy. That's bad because bringing in all of the items I want would set me back a cool $2,400 that I don't have; it's good because I'm confident that this stuff will sell, and until now I had nothing but the same old, same old for Christmas. Then it's bad again because I see most of these products on Amazon for a lot less than full price. One product, for example, costs me $19 apiece. Because it's sold in dozens and comes in three colors, I'd need $684 (plus inbound freight) to bring in the minimum assortment. I'd make that bet if I could sell them at $38. Since it's retailing for as little as $24 on the Amazon, that's one big investment I won't be making. In fact, competitors are selling many of these products at unacceptable prices (an $80 globe retailing at just $105, for example). A shark ice mold that costs me $5.50 (plus inbound freight) can be had for $6.49 with free shipping on the Amazon -- how can anyone make money on that? An umbrella that costs me $17 is going for $25 with free shipping. This got less bad when I realized that Amazon logs me in as a business account, so I might be seeing prices that aren't advertised to the general public. 

In the end, I decided to focus on just six of the 14 products that I wanted. That still sets me back $1,100. But I ain't ordering anything until I start getting some business again. 

Finally, something good-good is in the pipeline: I have a line on a part-time, work-at-home editing gig that pays such a handsome hourly rate that working just 12 hours per month would effectively double my income. If it pans out -- and it's not in the bag yet -- I'll cut my pay in half next year to pull Curio City mostly out of debt. As much as I hate to take it out of my own hide, that's the only way I can see it surviving. QuickBooks says my profit margin last year was -4% and I'm running at -7.5% so far this year. No way I can dig out while expenses exceed income.  

What's pretty ugly is that my debt actually crept up last month just as I'm on the verge of burying myself in Christmas purchases. The cautious debt ceiling that I had defined just two weeks ago has to fall unless I'm willing to paralyze the holidays, and a retailer who can't play the Christmas game is already dead.   

Friday, September 16, 2016

After Christmas, What?

Christmas is crowding out debt repayment as I begin bulking up for holiday sales. This year I'm going to pass up virtually all of the cute novelty items that I ordinarily buy from the same vendors and invest only in sure bets, with very few gambles. I am going to hold my debt to $8,000 (I'm at $7,000 already) and I am going to be stingy with advertising. That's going to mean lower sales.  

In days of yore, November and December combined brought in $25,000; for the past two years, that's been closer to $16,000. Let's say that this year's extra caution knocks it down to just $13,000. $6,500 of that pays for merchandise (I'm pre-spending some of that already in the form of debt, so I won't need to spend an additional $6,500, but let's pretend that I do as a worst-case scenario). $2,600 goes to payroll (I'll live on that until June) and another $300 to payroll taxes. I refuse to spend more than 20% on advertising, or another $2,600. PayPal's cut is about $400. The $600 "profit" that's left is diddlysquat. So I anticipate being $7,500 in debt when the 0-interest Amex promotion expires in January. I'll need another $1,000 to pay my CPA and taxes, and this year I have to come up with $400 for the USPTO if I want to keep my trademark. $9,000 would get me free and clear after Christmas.

That's an obscene amount of money. Where could it come from? PayPal Working Capital allows one to borrow up to 18% of one's annual PayPal sales; if my sales this year come in at $50,000, that just happens to be the $9,000 that I need. PayPal would then take a percentage of my sales plus a one-time fee based on the percentage that I choose to give them. For example: If I let them take 30% of my revenue, their fee is just $762 and I have to repay $9,762 for an 11.8% interest rate; at the other extreme, they skim just 10% of my income and tack on a $3,005 fee, making the repayment amount $12,005 (33% interest). One must pay at least 10% of the total debt every 90 days for the first 540 days of the loan. So on a debt of $12,005 I would need to pay them $1,200 every 90 days, or $400 per month. A typical month's sales are $2,500, so if they're taking 10% I'm going to come up short by $150 per month. I'd have to make up that difference or go into default. And it would take me 30 months to get out of debt. 

At the low extreme my debt would "only" be $9,762, so I would owe $976 every 90 days or $325 per month. 30% of my sales would be $750 out of my monthly $2,500, easily covering the minimum requirement, and it would only take me 13 months to get out of debt. Of course, that assumes that I can function on 70% of my usual income without going further into debt, which would defeat the whole purpose.  

Amex "only" gets 12.49%, which isn't too bad for a credit card. Their minimum payment requirement is trivial. However, the interest isn't finite, as it would be under the PayPal scenario, and would siphon off $100 per month indefinitely. Because it's revolving credit, I'm constantly adding to the same balance that I'm trying to pay down, so no monthly decrease is guaranteed. The PayPal scenario has the advantage of separating past debt from current operating expenses, while the Amex scenario can drag on forever. 

I'm probably going to look for a part-time job come February. My Curio City salary this year happens to also be $9,000 -- again, equal to the capital that I need. If I stop taking a salary and survive on my outside job I can bail Curio City out in a year. But then, what's the point of owning a business that doesn't pay me? 

I need to think on all of this a lot harder. I won't be able to take any action until I see how Christmas goes, but I need to have some options defined before then.


Google finally wore me down. When I gave the whole sordid history to an account rep and begged her for help she just said "That sounds like a Support issue" and gave me the 800 number for the same people who failed me last time. I'm absolutely not going to waste any more time with Kriti in India, so I'm stuck...for now, anyway. Implementing conversion value tracking is beyond my paltry skills and I can't get any help from Google. Maybe I'll pay my developer to do it.

Friday, September 09, 2016

Through the Looking Glass

Grappling with Google (Cont.): For a third time, I dutifully let a full week slip by without capturing any AdWords conversion data, took a couple of screenshots to prove that my new file doesn't work, and emailed them to Google's support kid. But this time, I didn't revert to the old, working file after proving my point. "Kriti" won't do anything if the file is working when s/he checks it, so s/he had to catch it in the non-act of not receiving data. Three days later, I got this response: 

Thanks for writing in.

I have checked your account and see that currently the tag is inactive. Hence please follow the instructions attached with this email to place the tag correctly and start getting conversions.

If you have any other query, remember I´m just an email away!

Kriti attached the original, old script that I've been trying to replace. I'll pause a moment to let that sink in. Yes, folks, I am literally right back where I started. 

Maybe Kriti understands only rudimentary English, or maybe s/he is deliberately screwing with me, or maybe it's a bot, or maybe (horrors!) s/he is as breathtakingly stupid as s/he seems. No matter which of those is true, there is no point to interacting with the Kriti-bot any more. I would get farther arguing with an AI.  

Instead I'm going to compose this whole sordid tale into a coherent narrative and find a human in Google who can understand it and will pretend to care. This morning I got this email with the subject line "Directly from Google": 

My name is Holly, and I am the strategist assigned to your AdWords account directly from Google. I did take a look through your campaigns, and I noticed a few issues within the account. What is a good day to go over that?

Oh dear. Poor Holly is about to get an earful (actually, I'm just going to ask her to refer me to a competent tech support person). Tune in again next week for another exciting episode. 


Chargebacks are my least favorite thing in the business world. I'm sorry that somebody might or might not have used your credit card fraudulently, but I'm more sorry that I'm going to take the loss for it (the bank sure won't; you can take that to the bank). Somebody complained to her credit card issuer that she "did not authorize this purchase" of $76 on July 18. It's a perfectly ordinary Switchables order that doesn't set off any fraud flags at all; in fact, the customer even created a store account, which thieves never do. I'm guessing that she saw "Curio City Online" on her credit card statement and did not mentally connect my store name to her Switchables order. She probably did get her Switchables, since she's not complaining about non-delivery (although since she didn't contact me, I don't know exactly what her complaint is). PayPal will undoubtedly snatch away $76 and fine me $25...and, of course, the merchandise is long gone. 

This is not getting a "reasons to hate PayPal" tag, btw, because the whole payment industry works like this. In fact, one of my previous processors hit me with $25 fines at each step of the dispute process, taking $75 for a single chargeback...plus the lost payment, plus the lost merchandise. 


Facebook decided to display last week's routine numbers post to more viewers (99) than any post I've ever posted (15 is more typical). I wish I knew why Facebook's algorithm deemed it suitable for so many news feeds so that I could replicate it with a post that I actually want to be seen far and wide. Blogger says that only 35 people actually read it, though, and that's only slightly better than normal. Apparently all that unexpected Facebook exposure didn't lead to many actual clicks.  

Friday, September 02, 2016

August Numbers

Despite the grim numbers you're about to see, August was pretty average, finishing as the 5th best (or 6th worst) in history. It would have been fine if the lady who wanted $1,100 worth of golf balls hadn't expected to get them for $600.


Total income: -27.9%
Total COGS: -28.8%
Payroll: -25.8%
Marketing: -16.7%
Net Income (Profit) vs LY: -96.9% (-$368)
Actual Profit/Loss: +$12


Total income: -9.0%
Total COGS: -8.3%
Payroll: -9.9%
Marketing: -4%
Net Income (Profit) vs LY: -29% (-$458)
Actual Profit/Loss: -$2,041

Told you it looks bad, but the Cost of Goods Sold actually fell slightly more than income did, advertising costs were down, I turned a tiny profit for the month...and I set an achievable target for next year.  

September historically vies with February for the title of Weakest Month of the Year, and next month's target looks especially easy because I shut down for a few days to attend my niece's wedding last year. 


In the Getting Things Done Department: Sunshop is upgraded to the latest version. My email signup dialog is finally re-implemented after far more hassle than I expected (and I've already had one signup). And my whole site is now behind my SSL certificate. There is no rational reason for the non-checkout pages to be encrypted except Google's preference; for the past year or so, they have been punishing "http" sites in search results. When Google says "Jump" retailers ask "Which ledge?" so now everything's "https." Like the TSA, this is all about perception rather than actual security...but if it improves my search results a little bit and prevents a few skittish shoppers from bouncing, it's worth the hassle of replacing all of my old non-secure links and embedded content. 

I am still wrestling with Google on the conversion-tracking failure that I've been writing about for the past several weeks. In our last installment, I had shown for the second time that the code in my new file doesn't work, and the Support monkey supposedly invoked his technical team. Then, instead of delivering the fixed file that I was hoping for, he came back with this:

Thank you for you continued patience.

I have checked the conversions for the month of Aug and see that the conversions have been recording.

I was ripshit. It takes a lot to make me lose my temper, and this kid succeeded. I fired off this reply:

Oh crap, not this again. We’re right back where we started.

Conversions record properly when I use the OLD file. They do not record when I use the NEW file. I keep reverting to the OLD file after proving that the NEW one is not working, and you keep checking the status while the OLD file is active. The attached screenshot is from the most recent week when I was using the NEW file. It clearly shows no conversions recorded for that week. 

Please give me an email address to contact your supervisor. We have spent the better part of three weeks repeating ourselves and this is going nowhere. Nothing personal, but I need an engineer to tell me what is wrong with the code in the NEW file. You just keep telling me that the OLD file (which is currently active) is working. I already know that and do now want to spend any more time arguing with you. 

I can't convince this idiot that he's wrong (which is a hallmark of idiots, come to think of it). He refused to give me his supervisor's contact and asked again for a phone call -- even leaving aside my phone phobia, what's the point in talking if he won't even acknowledge my complaint? 

I can't dumb down my request any further. I'm making one last slow, plodding effort to prove that their code isn't working. I might lose most of September's conversion data in the process, but one of us is going to prove the other one wrong before I drop it. His last communication reaffirmed his mistaken belief that the file is recording conversions. I should be able to prove otherwise next week.  

Google Search