As a creative agency, we're often approached by clients and prospects that haven't completed a cost-benefit analysis (CBA) for web development they'd like us to complete. That's often because they simply don't know how long their new feature will take to make. The problem is, often we don't know either. Read on for our web developer, Liam's take on this ever-present web development problem.
So Liam, what's the issue with making cool new web features?
More often than not, a request for some feature or function on a website will come along. I’ll listen to the various details and specifications of this great new thing, how it will work, what it will look like, and so on. Quite often, it’ll be an idea to engage people more on a website or improve the experience for visitors. The motivation and reasoning, I’m pretty sure, come from a genuine desire to improve and deliver a positive user experience.
And so, over the years, I have taken delight in helping to realise one cool feature or another, doing my part to help create a better, more usable world via the Internet.
Also over the years, I’ve become increasingly aware of the various types of client out there, with varying budgets and time constraints. Along with those budgets and time constraints comes estimates.
So what's the problem with estimating?
You'll be familiar with the most pressing question that a client wants the answer to...
“How long will it take to develop X and what will it cost?”
The challenge for me, as a developer, is that each new widget or cool feature, often seems simple to the client but can take hours of research just to figure out an estimate. I'm rarely asked to add something new, that I've already done before. There are always variables to contend with.
The first thing I'll do is try to find a ready made solution, that can solve the problem with minimal modification. Unfortunately, that means that my time in researching the estimated time for the solution will need to be factored in.
There are times, a good many of them in fact, where my guess is wrong. Sadly for me, it is rarely an overestimate! It must be my optimistic outlook but I always seem to think something is possible and will take less time than it turns out to. What I thought would take 20 hours, ends up taking 30, maybe even 40.
Sometimes, I've thought there was a ready made solution I could modify but it turns out it doesn't work how I thought. Other times, it would have been possible to come in on time, if we compromised the functionality to match a pre-existing widgets rather than trying to reinvent the wheel at a client's behest.
It's often the case that the time I guess it will take for me to develop that cool new feature, may not be justified by the benefit the feature will bring. For example, if it takes me 20 hours to build the widget but only 5 people use it over the course of a year, then that 20 hours could have been better invested in generating and sharing content to attract more visitors to the site.
I think website owners should consider the cost-benefit analysis before approaching an agency for an estimate. That way, if they think the added value of the feature would be worth investing £500 in, they can outline that as a ballpark budget at the beginning.
It is far easier for me to then decide whether this is worth my time to investigate a solution, if I have an idea of what the client believes the value of the cool feature would be.
How should a client evaluate the cost/benefit?
With time, cost, and estimates in mind, I find myself thinking about value. What is the value of what I'm being asked to develop? Now let’s throw in a term like return on investment (ROI). You’re not just looking at a cool new feature, but it’s something that takes time and money to develop; and, as with any investment, you’d want to see a return on that investment.
So, how is that thing we developed last year working out? Is anybody using it? Has it helped your bottom line?
As Jerry Maguire might say, “Show me the money!” (Although, some statistics or data will do just as well.)
Some things are easier to evaluate than others. How many files were uploaded via the new widget? How many submissions are in the log files? What do the Google Analytics say? Sometimes, it’s simply client or customer feedback — “Oh, I use that feature all the time. It makes things so much easier!” That’s what you want to hear. And, really, those can be the most satisfying to make.
But that's all with the benefit of hindsight. How can a client predict the ROI?
It's relatively easy to evaluate the ROI after the fact but not so simple to make a reasonable projection.
I’m a big fan of invention through necessity. You can immediately get a sense of how useful something will be and how often it will be used. Spending six hours developing a function or tool you know is going to save 50 hours of manual operation over a year is clearly worth the time and effort. But if it’s only going to save 15 minutes of one person’s time during that year, then we might be better spending that development time on other things.
That said, “cool is still cool”, and the value of cool is more difficult to measure. If you just want to make something that’s “cool”, and there is room in the budget for it, then let’s look into making something cool. If it’s cool and useful, well then, I'm all for it.
So cost-benefit analysis in web development comes down to statistics and guess work?
Pretty much, yes. We can only go with a best guess based on a median between an optimistic and pessimistic outlook on the expected value of the benefit brought by the cool new feature.
Can you show an example?
Sure, how about if we guess that the new feature might allow users to find a specific product 30 seconds faster. Let's say we sold 100 of them last year, we could guess that we might now sell 120 or maybe even 150. Let's go with 135 then! So, that's 35 more at £20 per item, making £700 in revenue more in a year; and if that continued over 3 years, then that's £2,800 in revenue. Now, let's say the gross profit on those extra sales is 50%. That means the extra sales would generate £1,400, so what would it be worth investing in the feature to generate that profit? £700? £1,000? I'll leave that to the individual business to work out.
How can the client know if their cool new feature is viable then?
Well, they can make an educated guess. If they know that it's worth between £700 and £1,000 to them, then they also know that that is approximately 1 to 2 days of an agency developer's time. Does it seem likely that the feature can be developed in that time?
If it does seem realistic, then at least when the client approaches me to ask if I can make it and how much it would cost, they can give me an indication of what they are prepared to spend on it.
If I know from the outset that the maximum budget is £700, I can pretty quickly assess whether I believe that is plausible. If it isn't, at least I've not wasted 3 hours on research, and, more importantly, our client hasn't wasted a couple of days waiting for my estimate.
That makes a lot of sense, Liam. I guess sometimes the best solution will be a functional compromise?
That certainly can be the case. If an existing widget has 90% of the functionality required, it often will not make sense to invest the time necessary to have the full functional requirement, as the added benefit will be only minimal next to the time needed to develop it.
Often, it makes more sense to modify a business procedure, rather than try to develop ideal solutions to fit that businesses specific need —particularly with smaller businesses, where the likely sales volumes are relatively low.
That's not always the case, though. I do quite like a challenge, and sometimes in web development, there can be surprisingly quick wins. The problem is, it's very hard to predict when!
Cost-benefit analysis in web development is great for clients and developers
You heard it from the developer's mouth. It's a big help to both the client and the web developer to have an idea of the likely benefit a solution to a problem will bring.
With a bit of research, some simple maths, and educated guess work, you can have a decent idea of what that benefit is worth in monetary terms, and therefore what it is worth to invest in creating the solution.
It may not be right but it's bound to be better than starting with no idea. Once you've got that ROI figure, you can use that to evaluate the real success of your cool new widget.
And hey, if it turns out not to be worth the hassle, you saved yourself (and us) a whole heap of time and heartache.
Got an idea for a cool new website feature you'd like us to quote for?
If you've got an idea for a great new feature, we'd be happy to discuss it with you. Especially if you've done your cost-benefit analysis and think it could be viable. Why not contact us about your website idea.