Comments on: Friday Q&A: Considering hiring an agency to build your product? Read this first.

By: Nicholas Ng

Nicholas Ng — Sun, 11 Sep 2016 06:18:49 +0000

I’m a dev agency owner from SEA and a technical guy. We normally build product for startups. We always have to clean shlt from previous developer or agency. One issue we noticed is sometime the owner don’t really clear what they want to build or the exact project scope. I would advise if you want to appoint agency or freelancer developers, make sure:
1. Always know what you want to build, what kind of feature, even know how it should work.
2. Communication is key to the project success, be upfront and tell this project potentially will be took over by internal team. So, scalability and maintainability is very important, tell them what’s your plan in the future, so we can prepare for it even from start. Some might not aware, ‘simple’ feature can make a different architecture in the system. Besides, always work together with dev team to follow up the progress.
3. Find dev agency and developer that is not over promise and under deliver. Development need time and planning, so both parties have to agree with what can be done and what not within certain period.
4. Budget is always important, we always have to consider between these 3 criterias( cost, time, quality)

By the way, even though I’m working in this field, but I think if you have your own budget and have a friend of technical guy that can assist on project planning, then that will be great. Lastly, i believe finding a right technical person for your internal team is always best choice for startup.

By: Laszlo Alt

Laszlo Alt — Fri, 09 Sep 2016 18:02:33 +0000

We use contracted work instead of an agency where we manage the people instead of just handing off work. This way you can be in control of the project. It does require some technical know how to properly scope a project before building. Outsourcing is okay if you plan ahead of time and put in the work before you build anything. The biggest issue I’ve seen in projects of my lifetime is that the owners of the projects don’t realize the amount of time and work needed at the beginning of a project/solution/product directly affects the outcome and success of it. If this means market fit, workflow fit, doesn’t matter. You need to validate and run through the idea ahead of time before you spend precious resources on building something capable.

By: Audra Carpenter

Audra Carpenter — Fri, 09 Sep 2016 17:48:54 +0000

So many awesome points here so I won’t expand on them. But one I’ve ran into many times owning a marketing agency for years I’d like to throw in the mix.

Company hires outside reputable company to build them a new website/product. Since the owners are sometimes only good at their product the agency sells them pie in the sky results, then the hand off happens and the internal team is left with a product/project they can’t get to work right … now having all the responsibility to make it work without any of the original authority to prevent it from ending up that way. Big issues!

Now… I do know it can work, but there needs to be a reliable experienced liaison in the middle to protect all parties through to the end; otherwise you better hold on for a bumpy ride:)

By: Burton Kent

Burton Kent — Fri, 09 Sep 2016 17:21:38 +0000

My company (the Nerdery) is often brought in to clean up other people’s work. I was on a Drupal project gone bad for Sports Illustrated. This week we talked to a future client about their WordPress site – it’s too inflexible and they want their front end design people to be able to put their designs in WordPress. It’s an ongoing problem.

My suggestion is to find out about coding paradigms used to make a platform extensible. WordPress has hooks, child themes, shortcodes and custom post types. Drupal has their own way of doing hooks, adding custom types, etc. Symfony/Laravel/etc. has its bundles/libraries so you don’t have to reinvent the wheel, and dependency injection is a huge help. And so on.

I’ve worked with code where the original developer was actually pretty smart. I could see intelligent decisions being made, or interesting, but wrong hacks. It was painfully obvious when the developer didn’t understand the concepts built into the platform to allow flexibility.

It’s also very important to separate code from presentation as much as possible. You want all calculations done before rendering the page if possible – that makes it much more simple to READ the freaking code and not have to go hunting for how content is being built/retrieved.

So those two things – knowing the paradigms any given CMS/platform uses to allow flexibility, and separating code from presentation makes a huge difference in how easy the code base is to maintain and extend.

You WILL find highly intelligent people who are good coders, but don’t understand the above two concepts. They’ll produce brilliant code to get around their deficiencies. You don’t want that. As Brian Kernighan said: “Debugging is twice as hard as writing the code in the first place. Therefore, if you write the code as cleverly as possible, you are, by definition, not smart enough to debug it.”

By: Karl Sakas

Karl Sakas — Fri, 09 Sep 2016 16:14:58 +0000

The biggest risk—investing too much before you confirm market fit. As an agency consultant and previously a digital agency PM, I’ve seen this repeatedly—clients would pay the agency lots of money to build a product that (it turned out) the market didn’t want.

Minimize your initial spend so you can test fit. If you’re so desperate that the product *must* be virally successful in v1, that’s not a recipe for being satisfied as an agency client. I had a client who expected a [niche] clone of Facebook for $30,000.

Don’t forget to budget time and money for marketing. “If you build it, they will come” is not a reliable marketing strategy. The “$30K Facebook clone” client? They hadn’t reserved budget for marketing, and the product never took off.

Agreed on checking references, having a tech expert confirm their work, and pre-negotiating code ownership. Ideally have a “work for hire” clause (you own it 100% once you’ve paid for it) and be sure they’re using your code repository, not theirs.

By: Rally Stano

Rally Stano — Fri, 09 Sep 2016 16:05:14 +0000

I work at Soshal, we’re a design agency, and from our perspective there’s similar pros and cons to hiring an agency for your product design. The pros: cohesive brand design from product to marketing, getting high quality work done fast, a large pool of expertise that you don’t need to hire and retain after the work is done. The cons are pretty similar to what you mentioned, most importantly the hand off. Be sure that you’re working with an agency that doesn’t tie you down forever, meaning that you’re not forced to work with them every time there’s a design need. To avoid this situation, ask to receive detailed design guidelines that you and your future marketing/dev team can reference. It’s a win-win for both the agency and client.

By: jamie

jamie — Fri, 09 Sep 2016 15:42:10 +0000

Having your program built for you is doable, but I would suggest only doing so once you’ve confirmed your product-market fit. There’s no substitute for having full control over iterations to learn efficiently from your MVP, not to mention how quickly this tweaking adds up in cost.

By: Black Thorne

Black Thorne — Fri, 09 Sep 2016 15:42:09 +0000

What an excellent question. Since we’re on the topic, I’d love to hear about the agency you contracted to implement your product vision, and how the experience went (and I’m not expecting all sunshine and roses here – even if they were the most amazingly talented and well set up agency imaginable, just the fact that you had to ask someone else to create the product was SURE to create significant disparity between your initial vision and the end product).
I’m also curious what you guys plan to do when you hit 500k/month. Any plans for that? (Suggestion: make the new goal 500K in profit/month)