While developing software for programmable logic controllers (PLCs) in the mid-1980s, David Farajun had one of those bad experiences that can happen with software: He lost a lot of valuable data.

That led him to start Asigra Inc. in 1986 and begin developing software that could back up data to a remote server.

The idea was new and unfamiliar in those days. But Farajun never gave up, continuing to plow his time and money into Asigra even as friends urged him to invest in real estate instead.

By the late 1990s, online backup was finally an idea whose time had come, and Asigra was ready.

Since then, the Toronto-based company — still headed by David Farajun as president, with his son Eran now as EVP — has been chalking up annual growth in the triple digits.

And it's racked up a string of awards from magazines such as Byte and Switch, InfoWorld, and Windows IT Pro.

Realizing that few enterprises would trust critical data to an unknown startup, Asigra had to find channel partners that had already established relationships with end customers.

Among the technology partners using its Televaulting software to offer managed backup services are Bell Canada, HP, Level Platforms, and SYNNEX.

David and Eran lived through the lean times, and the "overnight success" that followed those 21 years of persistence.

Here are their 16 tips for software success through thick and thin.

Through thick and thin tip #1: Start with first-hand knowledge of the need.

Asigra grew out of a disaster when a hard drive failed and David lost a large amount of valuable data.

He wanted to make sure that never happened to him again, so he went looking for a foolproof way of doing backups. He didn't find what he wanted — but he did find that many other people had suffered the same sort of catastrophe.

At the time, the usual way of backing up PCs was to tape. But many people failed to back up often enough, if ever. What was needed, he thought, was a completely automated approach.

"This is where the bulb ignites," says David. "If I've got problems, and you've got problems... some solution must be done to help other people."

So, like quite a few other software entrepreneurs, David started with his own particular need, found that nobody was selling what he needed, and decided to develop it himself.

That meant his understanding of his customers' needs was more than theoretical — it was first-hand. In a sense, he was the customer.

And when you have experienced the need first hand, you really believe in it. That turned out to be important, because while the need was real, for quite a long time the market was not ready.

Through thick and thin tip #2: Stay the course. Eventually you will land clients and partners.

David started working on his remote backup and restore software in 1986, back when the internet was not yet a household word and a fast modem ran at 9,600 baud.

The idea of backing up data remotely may seem obvious today, but it didn't 21 years ago.

On top of that, Asigra was an obscure startup. Even if David could persuade prospects that backing up to a remote data vault made sense, he then had to persuade them that his unknown company could be trusted with their valuable data.

Overcoming those challenges took years.

Before the idea could take off, the price of bandwidth had to decline, the capacity of storage had to increase, and the pain associated with tape backup had to reach a level where companies were hungry for an alternative, explains Eran, David's son and Asigra's EVP.

And Asigra had to find a way around every prospect's nervousness about trusting an unknown company with their priceless business data.

The solution turned out to be working through channel partners — more about that later — but that realization took time. And even those partners had to be convinced that Asigra was the company to help them offer their customers this new service.

For more than a decade, Asigra kept developing its technology and searching for customers.

"I believed that this market would take off one day, and I just continued to develop," says David. "Yes, it was lean years, very lean years."

David financed Asigra through that development phase with the proceeds from the sale of his previous PLC business.

Throughout that time, he recalls, friends and associates kept telling him to forget software, and put his money into the booming real estate market in Toronto.

He resisted. Asigra landed its first major partner, a U.K. firm, in 1999. The lean years were over. And for the past few years, the company has been chalking up annual growth in the triple digits.

"If you believe in something and keep persisting," he says now, "you'll be successful."

Through thick and thin tip #3: If prospects won't trust you, find channel partners they do trust.

As we mentioned, the first major hurdle for Asigra was everyone's unwillingness to trust an unknown startup with their data.

"Quickly I realized that large companies will never give you the data," David says. "For security reasons, for quality reasons, and so on. I didn't have any status. You have to go through a channel or through somebody that knows the customer best."

So he turned to VARs and service providers who already had relationships with end customers.

For some of those, managed backup was a logical extension of their existing offerings. Eventually their customers started asking for it, and Asigra was ready to fill the gap.

That's what happened with the company's first big customer. A British web-hosting company (that the Asigra executives don't want to name) had a customer looking for something like what Asigra had to offer.

On a visit to the U.S., company reps happened to see an early article about Asigra in an American trade publication. They tore it out and took it home, and a couple days later David got a trans-Atlantic phone call that lasted several hours.

The British company flew people to Toronto to see the software, and soon Asigra had its first major customer. The initial deal was worth almost $1 million, and over the years the relationship has grown to be worth many times that.

For several years, Asigra sold this way, only through managed service providers (MSPs).

Eventually, those partners began to encounter prospects who wanted the service, but wanted to manage it themselves.

So two years ago, Asigra launched an offering that end customers could host on their own systems. Eran says today about 25 percent of Asigra's sales are direct to enterprise customers.

Through thick and thin tip #4: Avoid channel conflict, and focus on developing software.

But rather than sell direct to end customers, Asigra lets existing partners bundle its software with storage and servers, and sell the complete package to customers who want to host it themselves.

The reason is simple, David says.

"We avoid channel conflicts... and it allows us to stay focused on providing good technology and support, and allows the MSPs to own the relationship with the customer end users."

MSPs make the sales to enterprise customers, taking a markup on the software and making money on the hardware to run it on.

Enterprise sales tend to be bigger than MSP deals — averaging around $150,000 — and they take five to six months to close, versus around three months for an MSP sale, Eran says.

Through thick and thin tip #5: Evangelize your idea, even if it creates competition.

When Asigra was younger, David talked up his idea to anyone who would listen.

He says now he was naïve. He showed prototypes and business plans to other companies, and was surprised when a few of those companies came out with competing products a couple years later.

But in the long run, this turned out to be a good thing.

David had a good idea that he really believed in, but the market wasn't ready for it. Asigra could not succeed until the idea caught on, and other companies entering the market helped legitimize it.

"I don't think that one company can serve the entire market," he says, "so the more demand, the more awareness about the technology, the better for the customers, and for the vendors."

It takes at least two companies to make a market. So David's early disclosures helped to create the market he needed to catapult his software to success.

And as interest in online backup has continued to grow, Asigra worked at making itself known to the press, industry analysts, and of course, customers. Tactics included speaking at trade shows and conferences, press interviews, and producing podcasts.

Through thick and thin tip #6: Focus on what you do, and be the best at it.

Asigra has one product: its Televaulting software. That's it, that's all.

"We've got lots of temptation over here to develop different products," David says, "and with sizeable income to the company."

But he believes that would be the wrong direction to take.

"I decided a few years back. I said, 'I'm going to be a backup expert, and I would like to be 10 miles deep, one inch wide backup — period. And what we do, we like to do the best.’"

Another way of putting it, Eran says, is that you have to know when to say no.

This doesn't mean Asigra can't develop new products, just that any new products it does develop must stick close to its focus on backup. Right now, David sees opportunity in data archiving. More on that next.

Through thick and thin tip #7: Aim for where the market is going, not where it is.

Most hockey fans know Wayne Gretzky's axiom about skating not to where the puck is, but to where it's going.

Based not far from Gretzky's hometown in Brantford, Ontario, Asigra is a good example of how the same principle applies to software development.

Asigra started with an idea the market wasn't quite ready for at the time. Not every startup in such a position would have survived; David's determination and ability to finance the company through a long development period made the difference.

But when the time came, Asigra had a head start on what has now become a red-hot market.

Now Asigra is looking at what happens next.

Driven by the declining cost of storage, ballooning volumes of data, and tough new regulatory requirements, data archival is a huge growth area.

"Customers do not delete anything," David says. So today Asigra is focusing R&D on creating data archival tools to address that expanding market.

Through thick and thin tip #8: Balance R&D headcount with QA and support.

Asigra has about 70 employees. More than a third of those work in software development, and roughly the same number work in QA and support.

It's a company principle, David says: "If we have one person in development, we have one person in QA."

Putting significant resources into testing and support costs a bit of money, he admits, but it's a vital part of keeping customers satisfied.

Through thick and thin tip #9: To weed out fakers, test new hires before you interview them.

One of the secrets of Asigra's success, David says, has been hiring the best people. Sure, every software CEO says that; the question is, how?

Asigra's starts by testing candidates before interviewing them.

Anyone who applies to Asigra is asked to complete a set of tests. Prospective programmers write seven tests; those applying for jobs in QA or support get five. The tests cover math, database knowledge, and C++ programming. There are also general intelligence questions. And every applicant is asked to do a piece of descriptive writing, Eran says, "to see how well they can communicate."

David, Eran, and Asigra's senior developers created the tests over 10 years or more, and they're still evolving. "Every year we re-assess them," Eran says.

Because applicants occasionally pass on tests to their friends — and headhunters have even secured copies to show to applicants — Asigra rotates through several versions of each test.

The tests are a response to over-inflated resumes. Too many applicants "think they're walking into a company where they'll sit in a cubicle amongst 800 other people, and nobody will know... what they don't know," David says.

Asigra's tests are designed to screen out those people, and they do. Between 80 and 85 percent of applicants are eliminated at this stage, David says. Some just clear the 70-percent passing grade. And many spend half an hour with the tests, then get up and admit they don't have the qualifications they claimed.

Through thick and thin tip #10: Put new hires on probation. And test them again before you hire them permanently.

Those who pass the initial tests go on to interviews. Those who get hired go on 90-day probation, during which they're trained on Asigra's products while managers evaluate how well they work with others.

At the end of this period, each new hire faces another 100-question test to assess their knowledge of Asigra's product line. Only after someone passes all these hurdles are they offered permanent jobs.

Asigra's best source of new employees is recommendations from its own staff, Eran says. To encourage that, the company periodically offers cash rewards of $500 or $1,000 for recommendations, which it pays out if a recommended person is hired and lasts more than three months on the job.

With Asigra's growth, hiring is on the increase.

Eran says the company will probably hire 20 or more people in the coming year. And most of those will stick around; he estimates the company's retention rate is 80 percent or better.

Through thick and thin tip #11: Touch your customers' systems as little as possible.

Asigra's software is agentless, so it resides on the server without installing anything on the client desktop.

There are two reasons, David says. One is that installing the software on fewer machines makes it easier to update. The other is that Asigra wants to touch its customers' PCs as little as possible.

That, David says, avoids the notorious finger-pointing problem.

"We had some experience in the past," David says. "Customers came to us and said, 'Listen, you installed your client on the LAN, now the printer does not work, this doesn't work, that is not working.' We said, 'No problem, we'll just disconnect the client.' And they said, 'Oh yeah, the problem's still there.'"

That would prove it wasn't Asigra's fault. But today, the company doesn't even have to go through this song-and-dance to troubleshoot an individual PC, since they never touched it.

Through thick and thin tip #12: Design features that help your channel partners.

Televaulting is online backup and restore software, but one of its key pieces is a billing module.

That's because Asigra sells the software mainly to MSPs, who use it to provide backup services to their customers, the ultimate users of the software. So the MSPs used the billing feature.

But now that Asigra has begun licensing its software to enterprise customers, those billing features enable some of those enterprises to manage and charge back services to different parts of their organizations.

"It's backup and restore software at its core," Eran says, "and it's agentless and it's clever in the way it minimizes the data movement across a LAN.

"But it's got a lot of service-oriented or service-provisioning-oriented functionality that can be used to deliver a managed service — whether you do it internally for chargeback, or whether you do it as a service provider to deliver a service to your customers."

David says one service provider who previously offered a managed backup service using a competitor's software told him that switching to Asigra made his operation three times more profitable.

It wasn't that Televaulting was cheaper; the software is better adapted to the service provider's operations, so that type of business can run more efficiently.

Through thick and thin tip #13: Design your licensing model to match your channel partners' business model.

Most online backup software uses a billing model based on the number of machines backed up.

"You have 1,000 machines, you buy 1,000 agents, and you pay per agent, and that's how much it costs you," Eran says. Sometimes the cost of the agents varies, depending on the operating system, the number of processors in the machine, and other factors.

Asigra's model is simpler: The customer pays for capacity in one-terabyte blocks. "You buy Asigra software the way you buy disk," says Eran — "one terabyte, two terabytes, 10 terabytes, whatever."

Why do it that way? Because that's what MSPs are used to.

"A service provider's business model is he charges his customer dollars per gigabyte or terabyte per month," Eran says. "If we are aligned — our licensing and price model is aligned with the way he goes to market — then we only make money when he makes money."

A typical Asigra sale is worth around $100,000 initially. Customers can add more capacity as they need it. Salespeople, who are paid a combination of salary and commission, earn commission on that added capacity, as well as on the initial sale.

Through thick and thin tip #14: Qualify your partners on all three aspects of technology, sales, and finance.

It usually takes Asigra three or four months to close a deal with a new service provider.

The company takes its time to make sure that the deal is a good fit for both parties. Asigra targets MSPs with $15 million or more in annual revenues. They don't have to have existing storage expertise, Eran says, but they should know network administration.

Eran says Asigra wants its channel partners to go through three kinds of due diligence before buying the software: technical, sales and marketing, and financial.

The technical due diligence is simply verifying that Televaulting works as advertised.

The sales and marketing aspect means making sure the prospective partner has customers interested in what Asigra has to offer, and sales people who can sell it to those customers.

And the financial part means assessing how much money the partner can make with the Asigra software.

To help with all this, Asigra developed templates for prospective partners to use in assessing and pursuing market opportunities. These include templates for customer presentations, service level agreements, proposals, and so on, with instructions on how to customize them to suit the individual partner.

Asigra also offers its partners a tool for calculating the internal rate of return they can earn with its software.

The end result is that Asigra signs up the right partners, and nobody is unhappy later. In the seven years or so since the business took off, David says Asigra has lost only a couple of its channel partners — and only because they failed to pay.

Through thick and thin tip #15: Train your channel partners extensively.

Picking the right partners is only the first part of ensuring successful relationships. The second part, for Asigra, is training them thoroughly.

Every Asigra technology partner starts with a week of basic training at the company's HQ in Toronto, where there's a full-time trainer, a training room, and a "pretty serious lab" for hands-on experience, Eran says.

They leave this basic training with a stack of printed documentation about three inches thick, he adds, plus access to online materials.

Later, after spending three to six months practicing with the software, they can — and most do — enroll in more advanced training.

"We train them to death," says David, "because we want them to support their customers thereafter." That not only helps the partners succeed, but allows Asigra to focus primarily on software development.

Partners are the first line of support for all customers, Eran says, but Asigra provides a backstop for any issues that partners can't deal with.

Asigra also provides additional training when it issues software updates, typically a major release once a year, with one or two minor releases in between.

Through thick and thin tip #16: When the wave comes, ride it.

David started Asigra with an idea that was ahead of its time, and kept going through many lean years before the market caught up.

But when online backup's time came, Asigra was in an enviable position.

"Having been operating within that market segment for so many years, we just know a whole hell of a lot about it. We have a lot of deep subject matter expertise associated with it, so we're able to ride the trend very well, and also to influence it," Eran says.

He says that anyone who asks about online backup at trade shows and industry events is generally pointed their way.

It looks like all that persistence through all those lean years has finally paid off, more than 20 years later.