2018 Business Improvement Plans (2017 Recap Pt2)

If you missed the first recap of the business, make sure to catch that as well.
You can find it here.

This post is about some of the planned changes coming in 2018. Note that these are planned, but not guaranteed. Reality can, and often does, put a damper on things. That and un-planned opportunities and ideas can get in the way too. Great example? I never planned on doing LED adapters for the Bally/Stern equipment. The opportunity was there, I went with it, and it was the single biggest thing that ever hit this business.

So what IS planned for this year?

Software Upgrades]

Software upgrades are already in progress, but the change will be swift and dramatic as I start 2018. A lot of this goes back to the profit margin / efficiency improvements discussed last post. Having a healthier bottom line makes many things possible which weren’t affordable until now. The many freeware packages out there are decent, but professional software does have big perks.

Financial Software

All my internal accounting is moving away from crappy Excel spreadsheets designed by yours truly. They have been making my bookkeeper/wife cringe for a couple years now, and they’ve become cumbersome. We’re moving to Quickbooks. This will be a more streamlined, integrated, and powerful approach. It may not be something that makes money for me, but it makes handling the business a lot easier, which gives me more time to work on production and product development.

Design Software

Speaking of product development, I’m rapidly shifting away from KiCad for board design. It’s served me well for close to a decade, but I’m bumping against its limits now. Particularly the need for improved routing of schematics and traces. After trying the demo for a month, I’ve started moving up the tree with DipTrace. It may not be perfect but it’s definitely smoother and faster to use for me in most areas. Since I have to re-design everything already, it’s the perfect time to switch. The full-up package may be expensive, but DipTrace is also scaled. Meaning I can buy the more basic pack, with certain PCB size limits, and upgrade over time as I need to. It’s a good fit for where I am right now.

New Markets

Oh, and about the product lineup? Well, I’m going to be moving outside of doing exclusively pinball. Don’t worry, I’ll still be doing all the pinball stuff I’m currently handling, and continuing to expand my lineup. I have far too many good ideas, a rather full pipeline, just that I’ve identified another market that I want to start working concurrently. That is the Arduino makers market. Parts of my current lineup would be a good fit to get my toe in the door, and I’m well positioned to start adding some Arduino specific products as well. In fact, they’ll probably be easier as I can hopefully stick with 100% surface mount assembly there. Due to the nature of the Arduino customer base, it’s actually advantageous to NOT solder connector headers in place. I’ll just include the headers in the package.

I’m going to ramp this up a bit slowly at first, but once I get traction it could be a hell of a ride. I’m not going to make myself dependent on this, however, because it’s going to involve some new business issues I haven’t had to deal with before. One is having actual competition, there’s a LOT of AVR stuff out there. The other is Chinese knockoff’s, which would become a very real possibility if I have a good idea brought to market. These issues aren’t even close to scaring me off, just that I have to plan for how to deal with situations of this nature.

My Online Systems.

Then, lastly for now, and this is actually a top priority that I’m going to work on right away, my shopping cart software. I’ve been using Prestashop since I set up my shopping cart system. I’m having issues right now with it, and it’s causing havoc with international customers. I can’t ignore 30% of my customer base, so I need a solution. The more I did, the less happy I am with my current system and its quirks. So it’s time to evaluate different software packages for my shopping cart. I don’t need a complex system, but I do need it to look professional and be a self-hosted package.

Right now I’m leaning toward OsCommerce, so don’t be surprised if my shopping cart completely changes in a few days here.

In Closing

So those are the major changes going on for 2018, some of which are going to be visible to you, some of which will not. I also have a healthy plan for product development, but I’ve decided to stop making that public. I’ve had a few people jump onto design ideas, and beat me to market with them. I’ve also had some designs move down in priority and/or get cancelled. So I won’t be announcing new stuff until it’s just about ready to hit the market.

In a few days hopefully I’ll have my numbers put together for how 2017 ended, until then keep ‘flippin.

2017 Recap – State of the Business

You know, I really look forward to these posts every year. I hope that anybody at all reads them. Particularly on a year like this one. I haven’t calculated the numbers yet, so I don’t know how it stacks up financially. However, from a business development standpoint, it’s definitely my most successful. Here’s all of what went into me saying that.

Volume starting to justify real efficiency improvements.

Lets face it, in the past I was basically a ‘semi-custom’ shop. Everything was made by hand, with an iron, and that made thru-hole parts the better choice. It made me flexible and the cost of entry was low. It was inefficient but it let me get in the game.

Well, things are busier now. The cost pressures are somewhat relieved. However, I am really struggling to keep up. Parts costs are higher, labor costs are higher. Most importantly labor TIME has been badly holding me back. I can only solder so much. I don’t turn enough volume to really see a benefit of outsourcing, and don’t have the liquid assets to afford it.

So I scrimped, and saved, and pinched pennies for a while. Not easy in a year with a big income gap in the middle. I finally brought in a surface placement machine. Last year I had started doing a few select, simple, assemblies in surface mount. These were just a warmup while I built up equipment. The equipment is now here.

What are the advantages? Lots of them. It cuts down labor time massively. What would take me days to do by hand, I can do in a few hours. I just ran 120 pieces of the new solenoid testers in about 2 hours, with surface mount resistors and LED’s. This would have taken me days to do in the past. And this is with a very basic, entry level, limited component machine. As time goes on, and I need to order new PCB’s, I’m going to be eyeing surface mount as extensively as possible. It’s the ONLY way I can continue to grow.

Surface Mount Technology Improving Margins.

Small secret. In the past, in the best of years, I barely turned any profit. Most years I’ve actually been losing money running this business. This is unbelievably common for new businesses, and I was mentally prepared for that situation. If you want to have a successful business it’s just the nature of getting started. Success takes time. I’m all out of generic platitudes now, so lets start talking about the future.

Going to surface mount tech has been a long term goal for me, because it will have a massive effect on profit margins. Not only does it significantly reduce time and labor expenses, it also has an equal effect on component prices. For some designs the PCB will cost less too, either due to smaller size, or because of less drilled holes.

As an example here, we’ll look at a generic 1/4w resistor. Absurdly common in everything electronic. Average price, if you order by the thousands, is going to be 3 to 6 cents each for thru-hole. A SMT resistor? About 0.7 cents each. That’s about 85% or so of a reduction in component cost of resistors. Which are much faster and easier to assemble too. Not EVERY component see’s that big of a price drop, logic chips can be about 10% to 40% less expensive, depending on the part. Still adds up though, that’s for sure. Oh, and the shipping costs go down too, since SMT resistors are much lighter in weight.

This means that some things I may actually turn profitable for me. The new bench displays are a great example. I was losing money on every piece I sold with the old version, after factoring in the labor time. The new version? I won’t give exact numbers but profits on them are pretty darn good now.
Solenoid/Special Solenoid testers are another one. I’ve sold a lot of them. They haven’t generated much of a net income in the past. Hoping that changes here.

Digging out of debt

I know, #1 question, will this mean reduced prices for customers? In some cases, maybe, in the long term. Right now it’s helped me hold the line to eliminate price increases. I have a fair amount of business debt that I’ve accrued in the past few years, which I need to pay down. The nice thing is that I actually AM paying it down now instead of increasing it. The rate of payoff should increase too, as profit margins continue to improve.

But I honestly cannot afford to lower prices right now. For the time being I’m just happy I don’t have to put parts on the credit card all the time. The more I can pay for expenses out of the bank account, and see the balance increase, the better shape I’m in. I’d love to make things cheaper for everybody but I haven’t reached that point yet. The truth is that as I continue to grow, my expenses are going to start increasing.

Within a couple of years I have a feeling the business will be moving out of the house and into a dedicated location. Then things will get really interesting. So I need to get my finances as solid as I can until then.

Going Forward

Now that 2017 is gone, and 2018 is here, there’s more things I am doing to improve the business. That’s for another post in a couple days. I’ve got boards to make.

Say goodbye to an old trusted vendor – a problem too far.

Well, good news, the pick and place machine is here and is running.
I’ve even made a few sets of aux adapter kits on it already.
Yep, this thing is going to be a bucket of awesome for me.

But man, what a chore it turned out to be, in regards to ancillary equipment.
Let’s back up a bit to explain, shall we?

To keep things in the realm of affordable, in providing a sub $5k pick and place with vision, they manufacturers made a sensible decision to offload most of the processing to an external PC. This external PC is NOT provided, you have to supply. Not a huge deal and makes sense. Doesn’t require anything fancy either. All the software fits, and runs from, the provided thumb drive. Really as long as there’s a single open USB port and a dedicated ethernet port, you’re in good shape.

So, I went to my trusted supplier for inexpensive PC stuff and ordered a factory refurbished mini-desktop and refurbished monitor. Well, things are really slipping at NewEgg these days. I know I ordered during the holiday rush but it was 5 business days before I got a shipping notice, and it was all coming via ground. Thank goodness for holiday shipping schedules or I’d probably be waiting another few days. But hey, I got a good deal, no worries.

Except the deal wasn’t so good. The refurbished monitor may work, but it looks beaten to hell. The PC also looks used, but not as bad, at least it came with a brand new keyboard and mouse. What the PC didn’t do however, was function. More correctly it intermittently crashed and gave me BIOS beeps to tell me it was a RAM/Motherboard problem. After some diagnosis I narrowed it down to half the RAM being dead. Talking with customer service my options are to send the PC back for an exchange (about 10-15 days process), send it back for a refund (10-15 days process), or send it to the refurbishing company for repair (10-15 days MINIMUM). So, in summary, I received beat to hell and only partially functional equipment. ANY kind of resolution for me will be time consuming and frustrating. All this on top of, by modern standards, abysmally slow shipping for in-stock items.

So what did I end up doing? Found a better deal at Amazon, which will be here much faster, and once it all arrives I’m shipping back the NewEgg system via the RMA’s that I’m sitting on. I’m also going to cross NewEgg off my list of companies to deal with. Their shipping used to be pretty quick and the deals were great. Now the deals are average (on new items), their refurbs can’t be trusted, and service is abysmal. I’m done. No regrets.

When it rains, it pours

Ok, lots of things going on here, it’s been a very hectic month.

1:   Busy filling orders from the ‘Black Friday’ sale.   Every time I think I’m catching up I get another pile of fresh orders. I can’t seem to get under 5-6 pending orders that I have to build product for. Moving as fast as I can.

2: The new bench displays will be delayed a week in shipping. I found out that I accidentally started building with an 8mhz crystal instead of the correct 16mhz crystal. This causes it to run at half speed and may be related to some goofy problems that customers are reporting with the pre-production units. The correct part is on order, will be here in a few days, and then I’ll get shipping.

3: I’m almost out of stock on solenoid testers, special solenoid testers, and switch matrix testers. These are all due for a redesign. The switch tester I did have a small run of surface mounts style ones that I already did, and a fresh run of 100pcs is on order with revised dimensions. The solenoid and special solenoid testers are moving to surface mount parts now too, which necessitates them NOT using the same basic PCB any longer. The solenoid testers were ordered at the same time as the switch matrix boards. The special solenoid boards I’ll order next week, so there may an availability gap as I’m down to the very last one right now. It just worked out good for me timing wise due to the new pick and place being ordered. Great opportunity to move more stuff to the surface mount line due to timing, worst case I can go with tweezers for a bit too until I get the machine up to speed. You’ll like the new designs I hope. More info when they start rolling in.

4: The new pick and place machine will be here Tuesday, but I’m also waiting on a new PC to run it with, which should ship on Monday or Tuesday and be here late next week. Exciting stuff there. Heck it’s even almost paid off already too, bonus!

5: Ooops, I broke the shopping car for international customers. For the time being, if you live outside the USA, you may get some errors during checkout. If that happens just e-mail me and I can do a paypal invoice to work around the problem until its fixed.

Some Basic Housekeeping on the site

Well, I finally figured out WHERE and WHY I’ve always had problems with shipping to Australia via my shopping cart.  The default section for entering Australian addresses is completely screwed up.   None of the localization data is complete,  none of the States or Territories were input at all,  and all sorts of issues.  So I’ve been working this morning to get this all fixed,  as I ship to Australia a surprising amount of the time. So, my Australian friends,  I hope the process has become easier for you now.  If any problems persist please let me know.

Also had an issue today with shipping to New Zealand,  which of course completely unrelated both politically and also in my website.  This was an odd error a customer received about a package being too heavy to ship.  This one has/had me confused but I believe it to be fixed now.

Another thing becoming apparent to me right now is how out of date a lot of product info is on the site.  That will be a bigger project but when (if) I have time, I’ll have to go through every product page to update descriptions and photos.   That will take a bit longer.

HUGE day for the business, in a good way.

Ok, one big piece of news, and one massively huge piece of news.

First big news piece,  the regular production of the new bench displays has begun.  No more prototypes, early production, or any of that.  We’re talking full-blown ready to go units in production numbers.    They’ll go live again on the site shortly.

Second news, and this is huge.   We’re talking massive shift for the company here,  biggest news I’ve ever had.   I just ordered a brand new pick and place machine.   This has been the gate preventing me from doing more complex boards,  and I’ve been working up to this moment for a couple years now with the gradual shift to surface mount testers.

This machine is going to let me make MUCH more complicated boards, with more speed and precision, than I could ever reliably build by hand.  This un-corks a lot of the dream designs that I’ve wanted to look into but couldn’t due to the cost and labor intensive build methods available to me.   An example is the gameplan LDU-1/2 replacement.    I couldn’t do that before, in a production manner, but now I’ll be able to.

2018 is going to be a good year.

Business Status Update

I know, half of you just checked out, another business status update.   But hey, it’s been a crazy two months here, and I have the feeling things are going to get even nuttier.   Short version,  I’m doing the best I can to kick orders out the door, but don’t be surprised if it takes a few days before you get a shipping notice.

So the big new is that I released the first batch of the bench displays up for sale.  I only had 10 pieces,  but the first style took probably two months to sell that many units, so I thought I was safe.   This time around it took about 18 hours to sell all 10.   Um, whoops.  The good news is that all 10 units are built, tested, and ready to package.  I’m just waiting on anti-static baggies and some connectors to get them all finished up.  Some are shipping tomorrow morning, the rest by Wednesday.

PCB’s for another hundred pieces are on order,  I’m working on a batch of OLED’s via a couple of sources to get good pricing, and the rest of the parts are all standard stuff that I can get in 3 business days.   So while I’m temporarily out of stock I can say these are a regular product now.   When they’re available for sale again all depends on when the OLED’s can get here.

I’ve also worked on my infrastructure a bit over the past two months.   Behind the scenes I added about $400 in software, and a few hundred on a new printer.  Part of that is financial software, and the other part is a new PCB design suite that so far is a big improvement on the freeware system I had been using previously.  Business keeps growing so I need to keep evolving it to keep up from an efficiency standpoint.  The printer will be nice for you guys, since the old one only printed in medium grey,  this one actually prints black.   Plus color, and double sided.  Saves me a LOT of time.

There’s a bunch of PCB’s that I’m running low on and will have to re-order soon.   Nothing critical right now, just don’t be surprised if a few testers are ‘out of stock’ here and there.   This summer caused a big hiccup in things, and with how crazy it’s been these past two months it’s getting trickier to maintain parts inventory.

 

 

Pricing announced on the new bench displays

Short version,   price on the new bench display will be the same as the old bench display,  at $45.   I HOPE to have these ready to ship by mid-November and test units are being sent out this week.

Long version….

I really wanted these done already, as I’m sure you did too, but I had a few issues to take care of with my PCB.  These delayed things a bit, but in the long term let me trim things down further and the result is going to be cleaner and easier for everybody.

The biggest issue is that I put the wrong type of programming header onto it.  I put an FTDI header, but needed an SPI header.  I’m still new to handling micro-controllers, and uploading code into a blank Atmel processor on a custom board isn’t the same as uploading to an Arduino board.   I have that sorted out now thankfully and the next board revision will be a lot easier for me to program than this one,  plus I also ordered a proper programmer to take care of this issue.  Bootloading the beta units was a wee bit convoluted.

During this process I came to the conclusion that I can remove about half the components on the PCB.  The ‘reset’ button isn’t needed, and I also don’t see a purpose for the indicator LED’s.  So all of those are going to be removed.  I’ll be increasing the size of all the remaining capacitors for my own sanity sake but keeping the surface mounted Atmel processor.

Realistically,  the per-unit cost of parts, material, and labor is significantly less on this version than the old discreet logic version.  So you’d think I can drop the price a lot, but it doesn’t really work that way unfortuantely.   Here’s how that all pans out….

I was barely breaking even on the old ones, due to all that labor time.  Lots of soldering, a lot of time at the crimping press.  Originally they had been priced at $60 but the market didn’t support that price point.   So I marked them down to $45 to get them to sell.   All the ones I sent to Marco Specialties I probably LOST money on.

The new units are a LOT less parts, a LOT less labor, a LOT less wire.  So what’s different?  Code.   I didn’t write it myself, I hired somebody to write it.  I also need to buy a programmer to streamline loading the bootloader and code.  This is all non-recurring expenses but still expenses to get things ready for production.   Then there’s the planned code upgrades to add Bally/Stern and other generations of machines to the tester.   These are additional expenses to pay somebody to write the code.    It’ll probably add up to about $1,000 or so when all the revisions are done and everything.   If I sell 100 pieces, that’s $10 per unit just for the code.   So I have to factor that in.

But the good news is I’ve got a great product here, nearly ready to ship, which is going to make a lot of people happy with the additional capabilities and future upgrades.  It’s a LOT better unit than the old one,  and it won’t come with a cost increase.    I’m extremely happy with how this product is moving along, and what it portends for my product line in the future.

Good times, friends, good times.

Teaser on the new bench displays

A bit of a teaser photo here of the new bench displays.  Actual PCB’s will be here Monday and I can start testing the dedicated PCB version.   Notice what’s missing?  7-segment displays, and a lot of circuitry.

Yep, I’m moving to a micro-controller based version, and the code developer I worked with really came up with a slick method to do it.  This will an OLED display, giving you the full set of score displays, in a very easy to read package.

Initial support will be for Williams 6 and 7 digit displays.   Easy to select, no more dip-switches and moving wire headers, just a single pushbutton to select between modes.  Code updates are planned for other manufacturers too,  and I’ll roll out an upgrade plan too so you don’t miss anything by buying early.

In fact, I think I’ll start up a page now for the shopping cart as a ‘coming soon’