Blog > How long is a piece of string and other quantitative quandaries
Some recent posts winging their way around the blogosphere have been tackling the sensitive issue of how much web design should cost. Looks like the conversation has been mostly around blog template design, but since we're no strangers to barging into parties uninvited and hogging both the cheese dip and the conversation, I though it was high time to add the Radiant Core opinion.
Looks like it all got started when Chris Pearson got tired of being asked and threw down the gauntlet. He quoted some other folks from a SxSW panel, including Peter Flaschner of Toronto's own The Blog Studio (and no stranger to Radiant Core, having collaborated on a few projects), and mentions Javier Cabrera. Peter expanded on his SxSW discussion in a lengthy but excellent post on TBS' blog. Go read - I'll wait while you catch up.
We've been in this game as Radiant Core since August 2003, and individually as freelancers and members of other agencies since the web basically started. I've seen people pay almost nothing and get incredible websites and I've heard of a company paying $800,000 for a basic marketing site with a contact form. I'm sure it would be great for our clients if the industry standardized on pricing, but that would make us an exception rather than a rule - try getting your house renovated to understand what I mean. We often use a house building analogy when explaining our process to our clients because it helps them to understand why building a quality website takes as long as it does (except with Brookfield Homes, since they actually do build houses). You would never choose to live in an unattractive house where the builder cut corners, so why would you choose to live that way online? Granted, there's less risk of having the roof come crashing in on your head, but having tying your online reputation to someone else's design and development work can end up having some serious consequences if they don't know what they're doing.
It's harder for us to list our pricing because we build everything from so-called brochureware marketing sites (see Päaeez as an example) right through to complex web applications (see the Brookfield Customer Connection). We don't follow a fixed pricing model where we shoehorn customers into a specific bracket, so each job gets priced on its own merits and complications. Generally speaking, smaller sites built on our Foundation Website Management Platform start at about $3000 (including design and Foundation licenses), level off at about $5000 for a more complex site with Flash, and again at about $8000 for a site involving eCommerce. The sky's the limit after that, based on our clients' timelines, requirements, and imaginations. We've built entirely custom applications (though always based on our trusty Foundation) for anywhere between $10,000 and $50,000. Our competitive analyses have shown that we sit somewhere around the Lexus point of the industry, which is where we're quite comfortable and happy to be. What's the Lexus point you say? If we sold cars, we'd be in the Lexus range: luxurious and high performance with excellent quality and customer service.
This dovetails nicely into a conversation that started at the first TorCamp back in Novermber '05 and continued at the Cost vs. Value Roundtable dinner. Our industry has adopted the pricing models of the advertising and marketing worlds because we very closely match their services (and often work with them). We generally build things using the "Cost" model, where we figure out how much it will cost us to design and develop a project, add in a little profit margin, and that's the number we give our clients. The question raised by Jon Lax of teehan+lax: why don't we use the "Value" model? We should be pricing projects not based entirely on what they cost us to build, but rather on what their value is to our clients. In other words, if a major multinational approached us to build a website, we should adjust the price based on the impact it will have on their bottom line.
Let's use our friends over at Widget Emporium as an example (a funny little fictional company who we developed to demo Foundation). Let's say that they sell 1,000 widgets a year through their old website, with an average price tag of $100 each, which is $100,000 a year in revenue. Now let's say that they approach us for a redesign and expansion of their site, with the goal of selling 100,000 widgets a year, or $10,000,000 in revenue. If we charge them $10,000 for their site and they reach their goal with our strategic advice, we'll have enabled them to achieve $9,990,000 in additional revenue after the cost of our services. The Value approach to pricing suggests that, instead of $10,000, we should price the job at something like $2,500,000, or 25% of their first year gain. Although that sounds fantastic to us, the client is unlikely to go for it since it requires them to adopt a big risk - if they don't hit their targets, they'll still have paid us a big wad of cash. But what if we priced the site at, say, $25,000, and then signed on to receive 25% of their first year revenues? We've now got an incentive to turn their old site into an eCommerce powerhouse, and they've spread the risk out much more evenly. If you watched the Jim Coudal video that we linked to earlier this week, you'd have heard Jim mention the exact same train of thought about 1/3 of the way in.
(Brief aside that might have been better as another post but isn't in another post so you're stuck with it here: This gets more complicated with sites that don't sell things directly (like our recent Bacardi Breezer project, or our work for Mozilla), but it really just means getting more creative with your metrics. If the conversion goal of the site is to get people to download the client's software and they have a 10% conversion rate of demo downloads to sales and sales are worth $250 each, then each download is worth about $25. If your new work increases the amount of downloads to 100,000 a year, it's worth $2,500,000 in downloads.)
It all sounds great and there are definitely some clients out there who will go for it (and a lot of clients who will suggest the 25% without the initial $25k, thereby asking us to shoulder all the risk). The problem is one that both Chris and Peter touched on: if there's someone out there willing to build the site for $5k and do a halfway decent job, then we can never progress to a more - well - progressive pricing model. If everyone in the industry made the jump together, perhaps led by a few big players who explained why it made sense and tried it successfully with some larger clients, then everyone would benefit. We haven't played this game yet because we're waiting for the right client to come along (maybe that's you - get in touch). I promise to write it up on here as soon as we do.
Looks like it all got started when Chris Pearson got tired of being asked and threw down the gauntlet. He quoted some other folks from a SxSW panel, including Peter Flaschner of Toronto's own The Blog Studio (and no stranger to Radiant Core, having collaborated on a few projects), and mentions Javier Cabrera. Peter expanded on his SxSW discussion in a lengthy but excellent post on TBS' blog. Go read - I'll wait while you catch up.
We've been in this game as Radiant Core since August 2003, and individually as freelancers and members of other agencies since the web basically started. I've seen people pay almost nothing and get incredible websites and I've heard of a company paying $800,000 for a basic marketing site with a contact form. I'm sure it would be great for our clients if the industry standardized on pricing, but that would make us an exception rather than a rule - try getting your house renovated to understand what I mean. We often use a house building analogy when explaining our process to our clients because it helps them to understand why building a quality website takes as long as it does (except with Brookfield Homes, since they actually do build houses). You would never choose to live in an unattractive house where the builder cut corners, so why would you choose to live that way online? Granted, there's less risk of having the roof come crashing in on your head, but having tying your online reputation to someone else's design and development work can end up having some serious consequences if they don't know what they're doing.
It's harder for us to list our pricing because we build everything from so-called brochureware marketing sites (see Päaeez as an example) right through to complex web applications (see the Brookfield Customer Connection). We don't follow a fixed pricing model where we shoehorn customers into a specific bracket, so each job gets priced on its own merits and complications. Generally speaking, smaller sites built on our Foundation Website Management Platform start at about $3000 (including design and Foundation licenses), level off at about $5000 for a more complex site with Flash, and again at about $8000 for a site involving eCommerce. The sky's the limit after that, based on our clients' timelines, requirements, and imaginations. We've built entirely custom applications (though always based on our trusty Foundation) for anywhere between $10,000 and $50,000. Our competitive analyses have shown that we sit somewhere around the Lexus point of the industry, which is where we're quite comfortable and happy to be. What's the Lexus point you say? If we sold cars, we'd be in the Lexus range: luxurious and high performance with excellent quality and customer service.
This dovetails nicely into a conversation that started at the first TorCamp back in Novermber '05 and continued at the Cost vs. Value Roundtable dinner. Our industry has adopted the pricing models of the advertising and marketing worlds because we very closely match their services (and often work with them). We generally build things using the "Cost" model, where we figure out how much it will cost us to design and develop a project, add in a little profit margin, and that's the number we give our clients. The question raised by Jon Lax of teehan+lax: why don't we use the "Value" model? We should be pricing projects not based entirely on what they cost us to build, but rather on what their value is to our clients. In other words, if a major multinational approached us to build a website, we should adjust the price based on the impact it will have on their bottom line.
Let's use our friends over at Widget Emporium as an example (a funny little fictional company who we developed to demo Foundation). Let's say that they sell 1,000 widgets a year through their old website, with an average price tag of $100 each, which is $100,000 a year in revenue. Now let's say that they approach us for a redesign and expansion of their site, with the goal of selling 100,000 widgets a year, or $10,000,000 in revenue. If we charge them $10,000 for their site and they reach their goal with our strategic advice, we'll have enabled them to achieve $9,990,000 in additional revenue after the cost of our services. The Value approach to pricing suggests that, instead of $10,000, we should price the job at something like $2,500,000, or 25% of their first year gain. Although that sounds fantastic to us, the client is unlikely to go for it since it requires them to adopt a big risk - if they don't hit their targets, they'll still have paid us a big wad of cash. But what if we priced the site at, say, $25,000, and then signed on to receive 25% of their first year revenues? We've now got an incentive to turn their old site into an eCommerce powerhouse, and they've spread the risk out much more evenly. If you watched the Jim Coudal video that we linked to earlier this week, you'd have heard Jim mention the exact same train of thought about 1/3 of the way in.
(Brief aside that might have been better as another post but isn't in another post so you're stuck with it here: This gets more complicated with sites that don't sell things directly (like our recent Bacardi Breezer project, or our work for Mozilla), but it really just means getting more creative with your metrics. If the conversion goal of the site is to get people to download the client's software and they have a 10% conversion rate of demo downloads to sales and sales are worth $250 each, then each download is worth about $25. If your new work increases the amount of downloads to 100,000 a year, it's worth $2,500,000 in downloads.)
It all sounds great and there are definitely some clients out there who will go for it (and a lot of clients who will suggest the 25% without the initial $25k, thereby asking us to shoulder all the risk). The problem is one that both Chris and Peter touched on: if there's someone out there willing to build the site for $5k and do a halfway decent job, then we can never progress to a more - well - progressive pricing model. If everyone in the industry made the jump together, perhaps led by a few big players who explained why it made sense and tried it successfully with some larger clients, then everyone would benefit. We haven't played this game yet because we're waiting for the right client to come along (maybe that's you - get in touch). I promise to write it up on here as soon as we do.
Posted by Jay Goldman on Saturday, June 10, 2006 at 10:53 AM in Taking Care of Business with tags pricing, lexus, foundation • Permalink • 3 comments
Comments
Peter Flaschner
says:
Jay, the concept of value pricing is a very interesting one - and one I didn't consider in my rant. We recently completed a project for the VeraSage Institute (www.verasage.com) - a think tank whose sole purpose is to forward the value pricing model. I'll admit that until I read your post, I didn't really "get it". In light of this conversation though, value pricing makes a ton of sense. Ultimately, it's up to us to deliver work and service of sufficiently high caliber that our clients will accept a hybrid of cost and hybrid value. It's true that there's always another shop out there, but if I want a Lexus, I'm going to shop at a Lexus dealer. Great post!
Posted by Peter Flaschner on Saturday June 10, 2006 at 2:42 PM
Scott Sehlhorst
says:
Value-based pricing is a great idea. The challenge is that we are almost always competing with someone who is doing cost-plus pricing. And that can mean a 2x (or 10x) price differential. For us to succeed when we're price-uncompetitive, we have to win the sale based on beliefs, not features. This is a challenge with software and services sales in general, and definitely applies to design contracts. When we are pursuing sales to existing customers, or referrals from existing customers, we are much more likely to close a value-based deal. In those pursuits, we are selling our reputation. If our reputation allows us to anecdotally demonstrate the value, we will probably be able to get our desired price. When we are looking at a first-contract with someone, it can be much harder to sell the value. The problem is that we can't sell the value (in spite of what I just said) - we are selling the perception of future value. We have to create a perception that our services or our products are simply better for this customer. Using a percentage-of-benefit approach is an exciting way to approach risk-sharing and revenue-sharing. I spent 8 years with a company that transformed its business model from fixed-fee to T&E and was trying to change again a few years later from T&E to a gain-sharing model. Every single initial conversation with clients about gain-sharing was very positive. The idea of "we don't pay if we don't benefit" is very exciting. It became much harder when the deals reached accounting/purchasing. Purchasing managers generally aren't geared to deal with gain-sharing deals. They don't like them, because they don't trust them. Purchasing folks are wired or trained to get the best deal they can. They are also predisposed to not trust suppliers. "Free if it doesn't work" sets off a red-flag for them, because it sounds too good to be true. Makes great sense to the stakeholders, who minimize their risk, but purchasing folks aren't stakeholders. They just know there's no such thing as a free lunch, so they become suspicious. Doesn't kill the deal, but doesn't make it any easier either. Purchasing folks trust the no-upside deal, because they won't be on the hook for something that ends up being unpredictably expensive. True or not, it is a perception. Accounting people really struggle with gain-sharing deals because they have to budget for them. As a freelancer, its easy to think of things in incremental terms - % of sales, % of sales relative to forecast, incremental units, $ per download, etc. There are two accounting problems, one financial and one managerial, with this approach. Financial accounting people generally want to budget for the expected cost. That means the internal company needs a realistic forecast of the potential benefit. This helps us, if the customer believes the software will really work. But it also hurts us, because when they do the math, 25% of a million dollars in incremental sales starts to look like an exhorbitant price. And while the customer may only pay $15,000 at the end of the day, they still need to budget for 'crazy money' if they really believe its worth it. If they don't believe our solution is worth crazy money, then we've lost some serious differentiation. It becomes a lot easier to go with the other guy. 80% of the benefits at half the price with no open-ended liabilites. True or not, that is the kind of perception we end up getting. The accounting folks also hate the open ended liability. In your post, you mention 25% of the first year, which is much easier for accountants than 10% in perpetuity. Management accounting folks will struggle to nitpick the details of 'incremental' revenue. First, they are on the hook for defining the formula that determines how much our software costs. Much more pressure than normal management accounting. These folks are used to allocating costs and driving decision making, but in a less-directly-relevant way. The extra pressure on them makes these discussions harder, but not impossible. The challenge is that there will be other reasons that revenues go up for our customer. Sure, our awesome website is helping. But so is that marketing campaign. Who gets what percentage of the credit for the incremental revenue? Imagine the hoops you'll have to jump through to get this nailed down. Now double them. I don't mean to tarnish the idea, it is a good one from the high level view. Organizations just aren't equipped to deal with it at the lower levels. My former employer closed maybe 10% of the gain-sharing deals it pursued as gain-sharing deals. Most of the deals were ultimately structured traditionally - always at value-pricing, though. Really interesting discussion topic, and I enjoyed this post very much. You just added another subscriber.
Posted by Scott Sehlhorst on Sunday June 11, 2006 at 9:36 AM
Jay Goldman
says:
Scott makes some excellent points (and a great decision to subscribe to our RSS feed - you there - reading this - you should totally do the same thing). He's identified a few parts to the problem:
- If the Value model makes sense, we need to move en masse away from Cost-based projects or someone will always be there to undercut us. We're not sure it makes sense for all clients (and we're pretty sure it doesn't for all projects), but if there's a consistent set of criteria where it does makes sense then it needs to be a widely adopted set.
- The Cost based model has been around as long as there have been Professional Services to offer and there is a huge and deep infrastructure around it which would have to be evolved. Convincing Purchasing Managers and Accountants is only part of the problem. Scott touches on budgets and he's right, but it goes deeper still. Value pricing suggests a more permanent partnership between vendor and client as it remains in the best interest of both members to stay in the team and to help each other. Our goal is to develop long term partnerships with our clients now, based on nothing more than our enjoyment and interest in working together. Value pricing adds a level of commitment to that which is generally not part of the negotiation currently.
- A successful website is part of a larger campaign whose value is usually greater than the sum of its parts. As Scott points out, how do you value the contribution of the website? If a client's metric is a 10x increase in sales in a twelve month period, we can do a lot to track how many of those sales were touched by the site, but if they are ultimately closed by an in-house sales team or by third party vendors, who gets the credit? It seems to me that some smart people (maybe even the VeraSage institute who Peter mention) could take a stab at coming up with some standardized contracts with pre-defined formulas which would help. I hereby invoke the LazyWeb and commission it to create us some contracts.
Posted by Jay Goldman on Sunday June 11, 2006 at 11:28 PM
Post a New Comment
Please note that comments are moderated to prevent abuse of the comment system and will only appear after they have been approved by the author of this post. Use the "View Source" mode to edit HTML directly if you're into that kind of thing. We might not approve posts that are offensive or really off-topic, and we will definitely delete spam (and place nasty, nasty curses on the spammers).
