Blog homeI have been thinking recently about how far ahead companies should be thinking with website planning. What about next year? What about the year after that?
Planning vs return on investment
Both are important
Planning: Understanding what you are going to do when. In IT terms this is a “roadmap”. In larger companies everyone talks about “the 5 year plan” - but smaller companies tend to plan for the web in shorter cycles. For example, I plan developments on our software upto 4 months ahead. We are only planned upto end of Feb at a detailed level. I have ideas for post February - but nothing set yet. The advantage of these shorter planning cycles is that we can react much faster to changes in the marketplace and customer needs. Its not that we are doing nothing post February - just that we have no need to plan for it yet.
Return on investment: This is a principle of “When are you going to get your money back from the costs of undertaking the project in the first place?”. Return on investment (ROI) calculations form a key part to the long term planning process at larger travel companies (where every development decision is made like an investment decision) - but smaller travel companies don’t tend to undertake ROI calculations that much. They just say “we have 2 staff - lets keep them busy 100%” - rather than “If we spend £30,000 GBP on this (60,000 USD) we could make £100,000 GBP (200,000 USD) sales next season - so we had better get some temporary staff (or a web design agency) to help out.
Return on investment calculations for websites tend to be worked out over 3 time periods:
- Anything related to search engine optimisation (where return will be made through sales from customers finding the site through organic search engine results) - 1 year. Some may push this out further - but frankly SEO can change so much - you had better be expecting to cover your costs in the first year - and anything after that is a bonus
- Web developments (and design changes) - 3 years
- Back end changes (reservation system etc) - 5 years
[These are my numbers - not used by anyone I know incidentally]
Examples & questions
Long term planning
I was asked to comment about the URL structure of a new development that went live this week. I pointed out that the URL structure wasn’t ideally future proofed nor would it make it easy to enhance the functionality in the future with additional products. The person I was speaking to said that project cost was a factor - and they had made the decision to go this route - to “just get it working” - and were going to worry about the longer term consequences later. In this kind of situation I would prefer to spend more time getting the foundations right - and then build on top. It may make the upfront cost higher - but the long term return from that development will be higher. A difficult balance though.
User generated content
Say you are looking at adding some more “customer experience feedback” type functionality. You know the kind of thing - people putting up videos of their holidays or reviewing their trip etc. The problem with this functionality is that it could be an entire year (or 2) before it begins to make sense.
First you have to tell everyone you have this new functionality - then people have to, the next year, travel on their holiday, then they need to come back and review or upload videos / photos etc. This whole piece of work will not have immediate impact on sales as soon as it is launched - indeed it may take 18 months to even begin to pull its weight on the website. How many website marketing executives are prepared to wait 18 months for something they have spent a great deal of money on to find out if it works or not? Not many!!!
(This is often why people launch their user generated content ideas on new websites - so they can capture content without their main site looking so empty - then, once they have some good content, they can feature it back on their main site)
Alternative principles
Another principle that you need to be aware of is the concept of “Fail fast”. This is an agile development principle (Agile is a method to run development projects that some developers like to use). In essence “Fail fast” means learning as quickly as you can whether something will work or not. In the example I gave above about not knowing if your new “user generated content” feature will work for 18 months….. this is an example of failing slowly. By failing quickly you know when to stop spending more time or effort on something - and you can move onto the next thing.
Getting the right mix in your roadmap
When you are developing your roadmap of developments it is important that you include those that deliver a return now (or over the coming months) and also those that have a more long term outlook. If you just do short term developments you may be OK today - but over the long term your more forward looking competitors will have built their long term developments up - and taken the lead. Of course you may need your short term developments to succeed so you still have a business while your longer term plans are building up!
Incidentally this is why you are seeing a few acquisitions at the moment for web based travel content companies. It isn’t for today - but for tomorrow - when these tools and systems will deliver their return.
Why I don’t like web design agencies
I don’t like web design agencies. They are OK when used as what I call a “resource extension” i.e. they are providing more staff for your team on a short term basis - but they can become a real cost overhead when used instead of an internal web team.
One of the reasons why I don’t like agencies is that I don’t believe they really get into this long term planning or ROI discussion. They tend to be very good at delivering a project in 3 months - that looks great - and delivers for today. But once they have your website live they go and move onto their next project - and never have to live with the consequences of the decisions they have arbitrarily made during the design process.
To better explain what I mean, I once worked for a web based hotel distribution company. We had a really difficult to work with booking engine that we just struggled and struggled to make better. We knew it was a bit of a pig - but we could do very little about it. This was 7 years ago. Customers were not booking, we had invested a great deal of money into it - and there seemed very few options (except for start again). However, between us, we came up with all sorts of ideas for how to make a poorly designed booking engine work better. We were forced into thinking about minute percentages of conversion uplift - and what the costs of achieving them were. This took us into areas that web design agencies were not going to consider.
Likewise, you have to wonder whether agencies code their work to assist with future maintenance. If you are in a situation where an agency is doing the project and then it is being passed to the internal team to manage going forwards - you can pretty much guarantee that there are going to be these awful hacks in the code to “get it working during the project” that you will then spend the next 2 years worrying about as your internal teams struggle to make any real changes to the system. It is not that agencies want to deliver bad code - but they are forced to meet their deadlines (they want to retain a reputation for that). Also agency staff tend to get very good at working on “greenfield” developments - where projects are done “completely new” - rather than maintaining code over its lifetime (3-5 years) - so they just don’t understand how difficult making maintainable websites and code is. They just have no experience of it.
Summary
- You need a long term plan (for example a 2 year roadmap)
- Your roadmap needs to contain projects that have both a short and a long term return
- Always consider the long term consequences of that “I will just put this up on the site and see what happens” development
- Ensure that with any design or development you have undertaken by an agency - that you understand the kind of thinking they are going through while delivering it. They will work how you want them to work - you need to drive this!
If you want to be notified next time something is published sign up for email alerts or subscribe to the RSS feed. Thank you for reading!
