Remend closes and lessons learned

The first company I worked as an interaction designer, Remend has officially closed it doors.  Here are some lessons learned for dot-coms building a product:

  1. A dollar spent in architecture is 10 dollars saved in maintenance.  Never hard code anything if you can help it.  The lesson is that things change.  Details change. You have to roll with the punches.  It is alot easier to change something flexible than something rigid.  Architecture is the key.  You don’t get alot of chances to start over.  This is exactly the opposite of what Extreme Programming states with YAGNI (You Aint Gonna Need It).  YAGNI works on a project, but not on a product.  A product is something that has to scale and flex and grow.  You will absolutely need it.  In a project with a start and end and defined scope, YAGNI works fine.
  2. 2 Great engineers are better than than 5 mediocre ones.  Great engineers have a can-do attitude.  They understand the spirit of what the business needs and can expand their minds to see the technical requirements of tomorrow’s needs, not just todays.  A mediocre developer will just see the exact specification.  This will come back to haunt you.
  3. Never let the short term business needs dictate the long term technical architecture.  This is so hard to do, but it is critical.  If a large customer wants you to build something custom, you have to figure out how to implement it in a generic way so that it can be productized.  You should never forget that you are building a product, not a project.  Building a product is a marathon, not a sprint.  I understand how hard this is to follow, but it turns out badly when ignored.
  4. Do not play revolving doors with your Architects.  When you let your technical architect or your user experience architect position become a revolving door, then you run the risk of the entire application falling apart.  A dot com has one, maybe two, chances to make it.  You have to get good people and keep them.  Leadership can not be a hot potato.  If you lose your main architects, you have to seriously consider your entire game plan.

I don’t want to bash any company.  Additionally, I think these lessons are very difficult to follow.  All I am saying is, I have seen these problems come up several times, and each time it has been a critical blow to the company.


Comments

10 responses to “Remend closes and lessons learned”

  1. I’m in a position where lessons 1 and 3 are painfully obvious, but I don’t know what I can do to address the issues. We have big clients that come to us with a budget and a list of wants, and the decision makers concede to both the budget and the list. So over the years these projects have been built upon, hacked-up, and held together with duct tape, but they are about to explode. It’s frustrating as hell to us developers because we know they need to be re-architected, but there is no time or budget for us to do that.

    At any rate, it’s aggravating, and we’re doing a much better job as we plan brand new projects (lessons have been learned), but can anyone comment on how to get throw to the client facing business analysts the magnitude of what it takes to “fix” an application? I’d love to hear how some other people deal with these demands on existing applications.

  2. Glen Lipka Avatar

    I feel your pain. The best thing to do is to understand the customer needs and PRODUCTIZE them. This is critical. You can not just “add” the features. You need to abstract them and turn them into configurable productized code, so that the next customer feature is a configuration and not new code.

    This requires that you LIE to your bosses and tell them, “Sure, we can do that in (padded for extra architecture time) hours.”

    Ultimately, these lessons are for bosses. They are the ones who need to learn them.

  3. Wow, I would have thought it was a good time for Remend with all the mortgage defaults/foreclosures these days.

    Your lessons learned are right to the point. The difficulty I see for startups is trying to ride the cash wave of one big client, basically building a one-off application to meet that clients business processes and trying to build a product at the same time. Especially if your architecture is to rigid.

  4. Glen,
    I found your post very interesting. I am the CEO of [COMPANY] (a competitor of Remend) and your four points read like a blueprint of how I try to run our company. I would add to your list “Manage your growth”. In a small industry like ours (default mortgage banking) a bad reputation is worse than no reputation. If you are unable to manage your growth and you don’t deliver for a client, everyone in the industry knows. I think Remend may have bit off more than they could chew with [CUSTOMER]. The problem was that [CUSTOMER] does not do business like other mortgage servicers (none of them due). What Remend lacked was the experience to know which parts of the application had to be 100% generic. Had they been in the default mortgage banking software business a little longer they would have been able to identify the functionality in the system that was going to be different for the next client and make sure they made those modules generic. The other item I would add to your list is KISS (Keep It Simple Stupid). As a software company you are much better off doing a few things very well than doing a lot of things poorly.
    I would definitely put item 2 (2 Great engineers are better than than 5 mediocre ones) and item 4 (Do not play revolving doors with your Architects) at the top of the list. If you have quality developers that know they are in for the long haul they will make sure that your architecture is solid. I have been very lucky to be able to re-assemble the core of my team from [OTHER COMPANY]. All of the developers have 5+ years of mortgage banking experience which is the key. You have to know what questions to ask the client and you have to be able to understand the client’s business needs and not just the project scope. In three plus years we have had no turnover and they all know that I will do everything I can to make sure they always have a job here. We are located in Orange County (near the Irvine Spectrum), if you know of any developers from Remend that share your philosophy and are willing to relocated please have them contact me. They can research us at [URL]. I have an amazing team of developers and we are always looking for people that will fit in.
    You might like to review the [PRODUCT] section, I think you will find it interesting.

    I know Sal and some of the guys at Remend and I have a lot of respect for them and I wish them well.

  5. Glen Lipka Avatar
    Glen Lipka

    Sorry Scott, I had to edit your post to remove the names of the companies. Although I always appreciate people commenting, I want to draw the line between commenting and self-promotion. Commadot isn’t for job board or SEO Link sharing, it’s for…hmm..it’s for something. But not advertising.

    Some free consulting. I reviewed your website.
    Your site is old school. Definetely Web 1.0. That menu is positively last century. What’s up with all the table cells. Your SEO is hurting because of the non-semantic markup. Your pageRank is ZERO. Wow, that’s as low as it gets. A high ranking is really useful to get Google qualified search traffic and it’s largely free to do. Use Google Analytics too. You need to know what is happening and where people are coming. You might want to invest some in Pay-Per-Click too. I did a search for Mortgage Banking Industry Solutions and you weren’t there. It’s not expensive.

    Introduce your web developer to jQuery. It will change his/her life. And improve your conversion rates.

    This is the rare test of management. You are the CEO. My bet is, if I look at your site in 6 weeks, it will still be a zero pageRank with no JS library, and no analytics and no PPC.

    Prove me wrong. My advice is free, but ignoring it will cost you some.

  6. David Kavanaugh Avatar
    David Kavanaugh

    Hi Glen,

    I knew this was going to happen also. I managed to find a new job here in Sacramento 4 months ago before this closure happened. All I could see was that the CEO kept pushing without fixing the problems the application had. Something had to give. Anyway, I’m now back in Java development at Accenture and learning/using the Spring framework.

  7. I heard this product was really crappy…from businesses that used it. No wonder they went out of business. In a housing market like this…don’t you think it is strange they did so bad? I am glad this turd got drained down the toilet.

  8. Hey, Glen, I was first UI designer in Remend in the very beginning, back in SF. Care to tell me what happened?

  9. I was researching on software packages in Mortgage Field Services catering towards vendor/ contractor management (managing vendors under property preservation). From my research I ended up going to remend.com and researched the product which was pretty cool. I dont understand why the closed – with the current scenario, they could make some waves in the market and make some good sales. Could someone like Glen guide me where I can find a similar product in market? which has following features for vendor management, order management in mortgage field services.

    The primary modules we are looking for are as follows:
    • Forecasting orders and source of orders
    • Recruiting and looking for more vendors
    • Credentialing vendors
    • Performance monitoring and Quality Check of vendors
    • Zone modeling and contractor / vendor mapping by office, status , services etc.
    • Reporting based on performance

    Thank you,

  10. “A dollar spent in architecture is 10 dollars saved in maintenance.” A smarter thing was never said!

Leave a Reply to Glen LipkaCancel reply