Following Through with Post-Launch Strategy

by Aaron Mentele

18 Reader Comments

Back to the Article
  1. Some really great points in here!! But even when you consciously plan to continue to help, post-launch work almost always winds up on the cutting room floor until they realize they need it. In my experience, this almost always results in a change order, or an entirely new engagement. Thinking about a retained service in the height of a major rebuild is tough for most clients. They can’t see that far out to know they’re going to need more help than they can manage in house. If we can figure out how to make this bundle work all the time, it will be a much nicer web for us all. ;-)
    Copy & paste the code below to embed this comment.
  2. It’s been my experience that any client who hasn’t learned the hard way embraces the launch-and-run strategy for reasons of upfront cost. Eventually, a stakeholder wonders why a competitor is drinking their milkshake, or (as suggested in the article) conversion targets are missed like the broad side of a barn. At that point, the lesson is learned… the hard way. As for SEO… well, I invested a small amount of energy into resenting SEO practitioners because… well, somehow, I don’t think I have to defend that position here. Several months ago, however, I realized that genuinely good SEO sets the scope of content strategy and, done right, is about as user-centered as a web discipline can get. The caveat to that is that a site getting the most from SEO requires ongoing care and feeding, not just new content shoveled on whenever the client tumbles to bounces on search terms they hadn’t expected. The challenge I see is not in the work, but in setting clients up to pay for it without slipping into a steaming bath of resentment…
    Copy & paste the code below to embed this comment.
  3. We’ve had pretty good success bundling the ongoing service at the time of the initial proposal. Typically, the goals of the project line up very closely with post-launch work, so we give specific rationale rather than add a blanket line-item (and only if it makes sense). If nothing else, it lets the client know we can help if they come to the realization that they need it down the road.
    Copy & paste the code below to embed this comment.
  4. The market I work in is light years behind all these kind of problems, so my questions is kind of off-topic. How much traffic do you need to be able to run a decent A/B test?
    Copy & paste the code below to embed this comment.
  5. As a digital strategist I’d just like to say, “hear, hear!”. Too often I find that companies, especially small businesses, refuse to look at the long run strategic digital approach because of a) smaller budgets and b) the need for immediate and measureable ROI. The problem is some of these programs take a little time, but have major positive outcomes as you illustrated above (and don’t get me started on the SEO thing; well, just read my blog post for details: http://www.deco-dig.com/dig-it-blog/2013/7/15/i-dont-do-seo-heres-why). ‘ Anyway, a hearty thanks from a digital strategist for hitting the nail on the head with this post.
    Copy & paste the code below to embed this comment.
  6. Excellent article and some good points made and strategies shared! We are struggling with the exact same scenario, and after feeling like we were leaving our clients hanging high and dry post-launch, to muddle through Google analytics by themselves wondering if the website was working, I decided to make similar changes. I have the same reaction to the letters SEO from my team as you described, but we are changing our mindset as we become more educated. And, I reached out to many of our larger and “want to keep the relationship alive” clients about their feelings regarding post-launch strategies and SEO in particular, and they were all, in unison, saying “yes please”. They were dying for a web agency to be willing to stick around, and provide that kind of support. It’s strange that web agencies have built up this animosity towards SEO, and sometimes I wonder if it is born out of a resistance to establish a long term relationship with clients. As web people we know, the longer the site is out there, the more things break. The longer the site is alive, the more we want to go back and do it “the new better way”. Developers HATE having to dig back into older projects, and designers want to move on to the new work, not mess around with projects that have seen their moment in the sun. However, as a business that is delivering client services, we are missing a huge part of the challenge if we don’t stay engaged. Even as we struggle to figure out how to manage post-launch work, and juggle an ever-growing list of clients (most good some not good), we know the real goal is to build better and better relationships, and you can’t do that if you hit the road after every launch. I’d also say that there is a resistance to “data” from a creative crowd, as nobody wants to see their new innovative idea get squashed because it doesn’t test well. It’s a tough lesson to learn, but data has to play a role. Not to mention, it’s fantastic information to share with the client, and base decisions on - when tested properly. I think one of the biggest benefits the web offers businesses is its ability to provide data and analytics. There is power in that data, and we are foolish if we disregard it or consider it an afterthought. Thanks again for a though provoking article, and for all the support as we go through some of the same growing pains. A parting thought, nothing lets me sleep easier at night than knowing I’ve got revenue coming in from annual contracts alongside the fixed bid, project by project business.
    Copy & paste the code below to embed this comment.
  7. In our agency we always say that the success of our customers is our success. It’s useless to create a spectacular website, if then the client will not know to use it. For that reason, you must train the clients,  teach them   to use programs like Google Keywords, Google Analytics, etc.. As the article says, this is not just for the customer to make more money, but that the project team can also see how your design works and how you can improve it. A question to the author, How long estimated to be providing this service post-launch?
    Copy & paste the code below to embed this comment.
  8. @Manuel - I’m no CRO expert, but I do know 100 visits to the page your testing is super bare minimum. Ideally you’d aim for that volume in as short of a time period as possible, say 1-2 weeks, as other variables could change which could skew the test. If volume isn’t there, you could always try to buy traffic too, but the quality of paid traffic vs. non-paid will vary as well. I’d venture visits 500+ and beyond are much more robust though. @Aaron - Thanks so much for writing about this. Although I’m an SEO consultant, I can speak fairly objectively - since I both am relatively new to the SEO industry (3 years) and started as a self taught DIY (little bit of everything - dev, design, etc - and importantly business owner first). SEO gets a hugely bad reputation - and I’ve seen this from as much the view of an outsider as an SEO myself. I wish MORE developers and designers would speak their thoughts and have a constructive conversation with us, rather than just writing it off because of a few (ok… a LOT) of bad apples. But the industry is maturing very quickly. I myself try to keep learning as much about dev as possible - just last week I attended a MeetUp about Twitter Bootstrap and LOVED it. I also have a huge passion for really nice design - one example being the Electric Pulp site - it’s stunning in my opinion. And I think we can do more all working together than not. Despite a lot of bad SEO companies, scam artists, etc - there’s a lot of good ones!
    Copy & paste the code below to embed this comment.
  9. @Manuel and Dan
    To get a statistically significant A/B test, or a true result, it involves a lot of math. Luckily there are apps like Optimizely that do all the complex math for you, so you don’t have to pull out your old college Stats book.  But, it basically boils down to making sure you have enough people taking the A/B test to ensure that your percentage of improvement is greater than your margin of error. How many page views you need really depends on the test and the magnitude of how successful the test is to be able to say with statistically significant confidence that A version is better than B version. For example, when using the conversion rate as the measurement of success, it usually takes 25 - 50 conversions, or completed checkouts, which is usually around 5000 people taking the A/B test, to decide a winner.
    Copy & paste the code below to embed this comment.
  10. ^Christopher Hamze is the Christopher in the article, by the way.
    Copy & paste the code below to embed this comment.
  11. Thanks everyone for the fantastic comments. @Manuel, I expect Chris got you what you were looking for. He’s smarter than I am anyway. @Phil, I was excited when ALA expressed an interest in the topic. I think it’s a huge stage of necessary evolution, and I think we’re already behind the curve on it. As far as budgets go, most clients are already spending much more through traditional marketing channels. We’re simply a better buy. @Tracey *hug* It’s good to know fastspot is going through so many of the same things we are. Your comments are great. You hit a few points I missed. ALA needs to put you on deck for the next edition. @Manya, everything depends on the specific case, but most of our relationships are ongoing rather than just setting up 3 month kickstart contracts. The latter probably have their place, though. @Dan, my pleasure. You’re right. SEO gets a bad rap. The space is so flooded with so many bad (not just bad, really bad) seeds that it just kind of permeates the air. We all know there are good consultants out there, though, and the more we integrate post-launch work, the more obvious it becomes. @Chris, hello.
    Copy & paste the code below to embed this comment.
  12. Thank you Aaron for the interesting article. Do you have any other results to share from your usability testing? I’m curious if there’s much that runs counter to current ideas/best practises.
    Copy & paste the code below to embed this comment.
  13. Hey, Jylan. We don’t have specific counter-intuitive discoveries that are widely applicable. Most of our general testing has been related to mobile optimizations (e.g.: http://electricpulp.com/notes/you-like-apples/) and they all seem to follow intuition. The one counter-intuitive generalization that we could make is that certain audiences prefer clutter if it comes in the form of endorsement. I.e., adding guarantees and ‘secure’ badges and testimonials and accreditations seem to help even though they can really screw with design integrity.
    Copy & paste the code below to embed this comment.
  14. Just want to say how much I love this article.  We (Newfangled) came to the same realization around the middle of last year.  While we’ve always had some post-launch consulting done by our project managers, we did not have a dedicated resource to lead this initiative. Once we started offering marketing automation training and consulting, that gave us the boost we needed to create a new department at Newfangled for post-launch consulting, where we work with clients on the things you have mentioned, as well as their marketing automation strategy (where applicable). So far, things have been going quite well, and we now have TWO strategists consulting clients after sites launch.  So obviously we think this was a great decision and I’m glad to hear that it is working out for you all!
    Copy & paste the code below to embed this comment.
  15. This is the problem I have argued for years now. As a UX practitioner, how can you justify your worth if you do not measure your success? Furthering this thought, we have been collectively pushing the beliefs that you launch, learn, launch again in an infinite loop. My feeling has been because agencies are looking for work on the whole and don’t always have the right people in place who are selling in the understanding that this isn’t a one night stand or a summer fling it is a long term relationship - we want to go steady and if all you want is wham bam thank you man you need to find another team.
    Copy & paste the code below to embed this comment.
  16. @Chris thanks for the comments. Love hearing that others are having success in this area. Post-launch is such a neglected topic, which seems at least a little strange considering how much we all love to extend our discovery, design, content, build and QA processes. @Andy yeah, having tangible data to feed design intuition is powerful stuff. We do a lot of work with ecommerce and financial applications, so it was an obvious conclusion - the data was in our face. That’s not to say we handled it as well or as soon as we could have. We probably have some making up to do.
    Copy & paste the code below to embed this comment.
  17. Great insights, thanks. We’ve just started down this path and expecting great things!
    Copy & paste the code below to embed this comment.
  18. It is great article!
    Thank you for sharing.
    自動車下取り
    Copy & paste the code below to embed this comment.