One of the more significant decisions a business owner makes is choosing the technology partner and it's not always scrutinized with the due care it deserves. A mid-level hire will likely get more attention from most companies than the agency that they will spend the next three years building and managing their digital front door. That's backwards. A poor agency relationship has an impact that is not only forepay, but accumulates. Things slip up due to the lack of locking scope. Budgets keep growing and growing with the addition of "one more feature" with no change order. Then the platform that finally ships is such a bloated piece of crap that you need to hire a developer, a ticket and wait 2 weeks to change the headline. It's not always the businesses that have a bad agency that are the ones that are being burned. They're the ones using an agency that was more than adequate for the job and that they need to fix. If you really understand how much money is going where — and where it shouldn't — then you're on your way to a “paying for itself” website and not a “line item on the P&L” website.
The Core Vendor Selection Trap: Incurring Additional Custom Complexity Costs.
The common scenario for all these agency horror stories is a company that needs a corporate website, a blog or a marketing site with several landing pages. Nothing exotic. Which is why the agency proposes to take a proposal for a completely individualized construction: bespoke content management system, bespoke architecture, bespoke component library. The quote is between $30K and $80K, and includes a “discovery phase,” which is prior to any actual work being done.
All that complexity doesn't help the client achieve what he or she wants to accomplish, but it's not often sold that way. It sounds great that it's custom code. It's like the client is having something made for him as opposed to getting something "off the shelf. All a basic content website needs to do is to deliver two things well: it is built too slowly, and the client is locked in to that agency (or a very similar one) anytime there is a change.
Technical debt is where it begins to accumulate even before the site is even launched. With a proprietary CMS, the client won't be able to simply hire any freelance developer or in-house marketer to make updates — they will need someone who has a wealth of understanding in that particular custom build. There is sometimes scarce evidence of documentation. The institutional knowledge is in the mind(s) of 2 or 3 of the agency's developers and if they leave the client's timeline for making even the smallest of changes extends. It is not by accident that they are dependent on you. It's a business model. The client is admitted with the build fee and the agency earns the big bucks with the maintenance retainer, which typically ranges from $2,000 to $6,000 per month and is based on the services of "hosting, updates, and support.
The issue is exacerbated by the fact that code is owned. There are many contracts that leave the agency with rights to the underlying framework, custom components or proprietary CMS which the site is on. Client is the owner of content and the agency is the owner of engine. Before the new agency is able to do anything, try and get out of the system and go to a less expensive vendor eighteen months in. It's not some theoretical. A system that only the original builder can decipher is the top one reason businesses feel like they can't get out of a company that they don't like — not loyalty, but sunk cost.
Ask, before signing on the dotted line, if the build that is proposed is really required to get the job the site needs done or if it's actually required for the agency's bottom line. Almost never do you need a custom backend for a corporate blog, lead-gen site or content hub. It should have a good load time, good SEO basics, good information architecture and must be able to publish content without anyone having to file a ticket to do so.
That's why it is important to avoid these critical pitfalls in the agency selection process.
The most expensive pitfalls that can affect a business typically occur when evaluating, before spending a dollar on development. Watch for these:- - Not paying heed to mobile first speed benchmarks prior to signing a contract. Most traffic to most content sites is now mobile – but most businesses endorse designs based on a mockup of the desktop version. Make sure that they actually tell you in writing what their commitment is on the load time – not "it'll be fast.
- - Not asking for the request of real Core Web Vitals or PageSpeed Insights numbers. If a proposal is presented that contains only pretty pictures and no performance numbers, then there's no idea of how the completed proposal will perform. Using a lightweight, pre-optimized SEO blogger template like Piki Templates, you can check out demo sites and get a true idea of how much performance you can expect before paying for an agency to provide a custom template.
- - Paying too much for what is simply regular lay out and design, and a proven template can do. The typical cost of a simple magazine or blog layout UI/UX design and build out by agencies is $8000 – $20000. A magazine style layout, ad-ready placement and mobile-first responsiveness are already built into premium templates from Piki Templates, like Grid Mag, and have been tested on thousands of live sites and paid for once; and they cost a fraction of the price of having a custom design created.
- - Not checking the efficiency of themes/frameworks before deciding on a stack. Directly inquire about if the offered system has idle JavaScript libraries, unoptimized picture pipelines, or render obstructing scripts. A lightweight architecture – the type of architecture that has been put into the creation of templates designed to be quick loading blogs – will invariably beat one that is heavier, and was not built with page speed as a prime objective from day one.
- - Giving into a showy portfolio rather than seeking performance numbers. A pretty case study with nice screenshots, doesn't necessarily indicate what load time, bounce rate or mobile usability is like at scale. Don't just take visuals, get actual results from these past clients, or from the agency itself, such as page speed score, bounce rate, and organic traffic trend.
- - Not requiring in the contract that the code be 100% owned and that the IP be transferred. It should clearly outline in the agreement that the client holds 100% of the source code, the content-architecture and any custom built components, without any proprietary lock-in. If an agency is reluctant to agree on this clause, then that is the answer.
- - Not checking to see if the agency is willing to use a tried and true structure rather than to start from scratch. A good agency will be happy to tweak and customize a pre-optimized base (foundation) site such as a Piki Templates site like Grid Mag for a magazine style content hub, or Quick Spot for a leaner tech and news site, instead of creating a basic site for content from the ground up. If an agency says that they need to do a completely custom development for a simple blog or content hub, then they're thinking about their invoice rather than your budget.
- - Ignoring whether the design is compatible with Google Adsense or ad-networks from the get-go. If the site will be using display ads, check to make sure that the design includes clean pre-approved areas for displaying advertising. A good adsense friendly blogger template will natively take care of this, as an agency would charge their clients for doing this afterwards.
Increasing your control of your Web architecture and TCO
TCO is seldom mentioned in the first proposal, but will be the ultimate number 3 years in. Sticker price of a build is one line item and the bill for each and every additional update, each and every hosting renewal, each and every hourly bill to change a homepage banner is the real bill. Those that manage to keep that number down have one thing in common: They don't let an agency cross the line from what really requires their custom engineering and what doesn't.
The solution on the content side of the business – the blog, the resource hub, the news section, marketing pages that need weekly changes but no developer to get involved – is a low-maintenance, ultra-stable framework. A theme such as Wind Spot or any other free or premium blogger templates that a company can obtain from say Piki Templates will provide the company a solid, tried and tested starting point which will be handled by the marketing coordinator in-house. Colors, fonts, layout sections, ad placements – all of which can be adjusted in the template settings – without having to write any code or send a developer invoice.
That's not as dry as it is when you read it. Any time a business spends upwards of $150 per hour to swap an image on their homepage, is an hour of agency budget that could be spent on something that actually needs to be customized, such as a checkout flow, an internal tool or an API integration with a CRM, a client portal with real logic behind it. Smart operators keep tech spend in line with the value they're receiving by using expensive/specialist tech resources only when they are needed.
It's a control factor here as well aside from the price. By using a mature, broadly used framework, a business is not held ransom to one vendor when using content architecture. If the agency doesn't do a very good job, the relationship could come to an end without the site falling down – the site doesn't need any proprietary template and isn't undocumented and not known by only three people at one agency. Whether you want to change hosts, or you want to take design in-house, or hire a different freelancer to do the redesign, it's all still very feasible. It's an optionality that is more valuable than most business understand – until they are the one having to deal with a vendor who knows that they do not have the luxury of being able to leave.
Ensuring that Accountability and Performance First is the way to go.
A tech agency should speed up a business, not slow it down; speed up the time it takes to get a business online, to update it, and to load it onto a buyer's phone. When it does the other, it's hardly ever because of the lack of talent on the agency's part. In a nutshell, it is a case of not checking the client's side - but still, not just when the bills start to arrive, but rather before the agreement is signed. It is the companies that have been to the initial meeting with questions about page speed score, code ownership, and whether the proposed build is more or less complex than the problem itself, and not the size of the agency's day rate, who see the greatest return from their agency relationships. Clean code, speedy load times, and being cost-effective on a per-need basis are no frills. They're the baseline. If someone is offering less than that, in a good pitch deck, then they're pitching the wrong product.
