Coffee Cellar Postmortem

Coffee Cellar has officially been in the App Store for 1 month now, you can check it out here. Now is the perfect time to examine what went right and what went wrong. This is the final post in my App development series. If you want to read about the process of creating Coffee Cellar you can do so here.

What Went Right

1. The production process

I had a clear production process in place right from the start. I carefully created a step-by-step plan, determined not to fall into the same traps as before. As you may have read by now, this process proved extremely successful allowing me to remain in control throughout the project.

Not everything can be rammed into a neat and tidy process however, and I still had to make design decisions during development that significantly changed the workflow in Coffee Cellar. Even so, the process I used gave me degree of control that had been lacking in previous projects.

In addition to this, writing the posts detailing the process has served to keep me motivated throughout the project. It essentially creates an artificial deadline and, because its public, you feel a pressure that I personally found helpful.

2. Beta testing

This is the first App where I’ve used public beta testing (I’ve alway used friends and family in the past). The feedback provided by my beta testers proved invaluable and really enabled me to focus on areas of the App that people were most excited about.

Its always looked like a fairly daunting process to me but now that I’ve done it, I wish I had jumped into it sooner. I would certainly look to capitalise on it more in future (see what went wrong) but Coffee Cellar is a much more solid App as a result of the testers.

3. Artwork

I did the artwork for coffee cellar myself and I’m extremely pleased with it. I spent lots of time iterating the appearance as I developed the App, especially the cupboard which is the first screen you are greeted with. I found that switching between art and code worked great and gave me the creative breaks (from code) that I needed.

I know that in general people are advised to steer clear of attempting to do everything themselves – you can nearly always get the job done to a higher standard and in a fraction of the time by hiring a specialist. However, I get so much out of knowing that I produced the artwork myself and that alone makes it worthwhile – not on every project, but definitely on this one.

Producing a moodboard beforehand was also a great driver in creating cohesive artwork. Especially useful was the colour palette which I decided upon before coding and design began. Having a visual reference point makes a big difference.

4. The photographs

Coffee Cellar lets you take a photo (or select one from your library) and assign it to a coffee. It appears as a little polaroid image that you can tap to enlarge – this action removes the smaller polaroid and places a larger one in the center of the screen (you then tap again to dismiss). Feedback on this feature has been really positive. Its easily one of the most liked features. It’s not something that I focussed heavily on as a core feature, I just took care to implement it well.

Its an area of the App that I should expand upon in future (adding photos to the coffee sharing is fairly high on my list).

5. Reaching out to mainstream press

Coffee Cellar is fairly niche and I was unsure how many iPhone review sites would be willing to pick it up (and rightly so it seems, so far zero). I opted to target both food/coffee related publications along with the mainstream press. This was a good idea. The Guardian (a notable UK newspaper) responded and briefly mentioned Coffee Cellar in an online App round-up feature.

I think the key message is to understand that mainstream press are as much a viable PR route as the more niche, App specific sites.

What Went Wrong

1. The Marketing

One day this won’t be on my “What went wrong” list but for now it remains. My PR and Marketing was lousy. It was heaps more organised than my previous attempts but it just wasn’t focussed. A very small number of the contacts I reached out to responded and an even smaller number gave me any coverage. My website and twitter remain my biggest drivers of sales (for obvious reasons).

I need to spend some time learning and practising the basics of PR and Marketing. It certainly gives you a new found level of respect for what these teams do. It simply isn’t easy to create buzz from nothing.

2. Underestimating time required

This also features heavily in my “What went wrong” sections. This time it was down to migrating to Xcode 4 and learning to use Interface builder and UIKit. They weren’t terribly taxing but I did find myself having to learn and reference a lot of things – very time consuming. In future I’ll allow more time for getting to grips with new technology and tools.

3. Social Networking

Coffee Cellar lets you share coffees via email and twitter. I had originally planned to include Facebook too but removed this due to time constraints (It will appear in an update). I also ended up implementing twitter by passing details off to the twitter website, leaving it to deal with authentication and posting. This has created a fairly inferior social experience and as I think back, this is something that I should have thought harder about and allowed more time for.

I plan on addressing this in an update, along with solving the issues around sharing coffee photo’s.

4. Beta Testing

Beta testing provided me with some really great feedback but it was also an area of weakness for the project. I didn’t get enough people on board early enough and I didn’t have a good plan for the beta.

I think what’s required is a carefully considered, phased approach to beta tests. You need clear guidance as the what you are trying to test in each release and you need to be organised about responding and acting upon the feedback. This is a chance to grow the first seeds of your Apps community and thus communication is crucial.

5. The tasting notes

Coffee Cellar is all about recording your entire coffee tasting experiences. Whilst the tasting notes functionality is perfectly serviceable, I think its a little too hidden away. It also leaves new coffee drinkers in the dark a little as what to write.

My original intention was to provide some integrated help to assist those new to coffee tasting. Unfortunately, I was a little impatient and ended up stripping this back. As a result I’m sure that not everyone is using the tasting notes effectively.

Again this is something that I need to address in an update.

(Image courtesy of easop)

Part 1 – Doing it properly
Part 2 – The Design Document
Part 3 – The Wireframes
Part 4 – The Artwork
Part 5a -The Code (Xcode, GIT, and Core Data)
Part 5b – The Code (Coffee sharing and Camera functionality)
Part 5c – The Code (A change in workflow)
Part 6 – The testing
Part 7 – The Final Artwork
Part 8 – The Marketing
Part 9 – The Launch
Part 10 – The Numbers
Part 11 – The Postmortem (You are here)
(Bonus) Part 12 – A review of the process


  1. Josh Jones says:

    You did all the art yourself? Very impressive! Most coders aren’t artistically inclined. ;)

    I feel bad because I didn’t beta test it more. I think by the time I got around to it, you were already submitting it to the App Store. I’ll remember to do better next time!

  2. Chris says:


    Thank you. Actually I’ve been surprised at just how many coders are artistically inclined :) I was originally on a path to study Art at university but opted for Computer Science at the last minute.

    Don’t feel bad about the beta testing. I had enough responses for it to be valuable and I know from experience how difficult it can be to find the time to test.

  3. adam says:

    Your volume / focus of marketing isn’t the problem; the app itself is the problem. It sounds like you did “enough” marketing that the app should have done much better — if people had found it interesting enough.

    But looking at the video, two things stick out: this app is pretty, but appears to do almost nothing of marginal value. (i.e. compared to what your audience already has on their phone). Most people I know who care about coffee already record a LOT more detail than you showed using EverNote or similar (even using Apple’s built-in Notes app). If they adopted your app (which they’d have to pay for), it would be a HUGE step down in functionality. The only benefit I can see is that it would be easier on the eyes.

    So, I’d suggest you have a basic sales problem: people aren’t writing about the app, and aren’t buying the app, because you’re not giving them a reason to do so.

    It could be that your sales materials are too weak (for instance, the video has some mistakes, like white-on-white text, and a lack of focus – trying to show many small, unexciting features instead of just one or two big, exciting features).

    But personally I suspect it’s simply: there isn’t a market for an app that does this. The nearest app I can think of is a Tea equivalent, which has easily 5x the functionality if not more (again built by one person, over a long period of time). I can’t remember exact sales figures, but I think he made it to the low thousands.

  4. Chris says:


    Than you for taking the time to write that. I always value the opinions of others.

    You may not read this post again but it would have been great to know from you what the detail is that your friends record. Additionally it would be helpful to have the name of the equivalent tea app that has more than 5x the functionality.

    Coffee Cellar allows you to store your coffees, along with tasting notes, in a fashion that makes them easier to search. You can also share the Coffees with your friends.

    I built Coffee Cellar because I’m passionate about coffee and wanted an easy way to record the gourmet beans that I buy and record tasting notes for them. An advanced note taking application, like Evernote, whilst adequate is not designed specifically for this. I don’t understand your comment about Apples built-in notes App being a step up.

    Coffee Cellar contains a notes area against each coffee so it contains at least that functionality at a minimum.

    I am of course growing the feature set of Coffee Cellar. Some things I currently miss are the recoding of a price and the ability to store multiple sets of tasting notes against a coffee.

    I totally take on board your point about the video. I’ll listen to what you have said and perhaps attempt to create a new version that has more impact.

    My comment has perhaps come across as fiercely argumentative, which was not my intention.

    Coffee Cellar is a niche App that, as a coffee enthusiast, fulfils my needs. If it fills my needs, chances are it fills the needs of others (and according to my sales, this is true for at least some people). I’m not trying to compete with the general flexibility of a product like Evernote.

    Thank you again for the comments.

  5. adam says:

    I haven’t bought your app, so I’m going by the video, the diary, and the screenshots. As I said: the video could be more polished, but IMHO that’s very unlikely to be the cause of your low sales – your video is nice enough, and I’ve seen MUCH worse videos achieve much higher apps sales.

    The video in particular shows you typing into a small field, and the notes showing in a tiny percentage of the screen.

    Even the most basic alternative – Notes – gives the user considerably more screen space to edit and view notes, and a very fast, simple – and FREE – indexing setup. Notes also syncs directly to Gmail, and to other iOS devices (which your video suggests your app doesn’t – unless I missed that part, in which case sorry). So, if I wrote notes in your app on my iPhone – would they show up on my iPad?

    Things like Evernote go further, and have a decent tagging / search / document retrieval system. I’ve seen people using Evernote (and rivals) to store loads of rich info – essentially treating it like a user-friendly database.

    The tea app – I can’t remember the name, or I’d have stated it. Googling should find it quickly, I don’t think there’s that many hardcore tea apps. It has stuff like a brew-timer that lets you measure and record tasting notes for a single tea but for each different number of seconds / temperature of brewing. That’s a bit of a gimick – but at least it starts to differentiate from Evernote et al. On its own, I doubt even that is enough to move people away from the thing they know and is working fine for htem already, but it seems to be more obvious a differentiator than anything you’ve got here.

  6. Chris says:


    Thanks for responding :)

    I wonder if my video is confusing so I’ll clarify it a little. You add a coffee and are able to enter a name, country of origin, region, farm, and merchant. You can also choose an icon.

    Once the coffee has been created, you can add tasting notes – as individual words or multiples. You can also add any other notes you want – this is a full screen view. It may look small on the video because the keyboard has popped up over it.

    I don’t have synching but now that iCloud is out, it’s something that i’ll be investigating.

    Recording settings for temp, extraction time, etc is a good idea. I had considered it earlier on in the design but opted for simple notes to begin with.

    I think that for people who have gone to the trouble of recording notes etc through an App such as Evernote and are happy with their process, Coffee Cellar may not be enough to entice them.

    However, for people who are just “getting into” coffee, it offers a lot.

    Its certainly interesting hearing an opposite opinion. I don’t agree with everything you’ve said but I’ll certainly take on board some of the suggestions.

    Thanks again for taking the time to read this article and to comment.

    (Its always difficult to fight back the protection you feel over something you’ve poured so many hours into)

  7. Sam says:

    I think this is a great idea. In a former life I ran a micro coffee roasting company & I could see how this app would be very useful to those in the industry as well as those who are just enthusiastic about good coffee.

    There is a large network of successful independently owned coffee shops that actually care about the taste of their coffee. If you could find a way to team up with them to get your app in front of their customers that could be beneficial. These are the customers who would love your app. Many of these shops hold cupping events, this would be the perfect time to introduce your app to customers.

    Maybe you could offer the app for free and do some deal offers based on Geo-location. Or the app could be a way for these independent shops to notify customers of new offerings. Would the idea of a foursquare like service specifically for coffee shops make any sense? I don’t know the answer, but it seems plausible because people who are enthusiastic about their coffee are usually very dedicated to good coffee.

    It would certainly be a lot more work for an app that it appears you no longer support, but I can see the potential in it.

  8. Chris says:


    Its always hugely motivating when someone writes a comment like this, so thank you for the kind words.

    You’re right about getting the App in front of the right people. Working with a coffee house is something I’ve considered. I need to read up on the legal ramifications of handing out free copies in coffee shops. Apple have supprisingly tight rules on what you can and can’t use promotional codes for.

    Anyway, I love the ideas you’ve suggested. It’s certainly got me thinking again :)

    On a final note: I haven’t given up on Coffee Cellar, I promise. I’m fully supporting it for the foreseeable future and would like to create some more free updates for it. I have to pick my battles wisely though (I only have so much spare time) and in the short term I’m focussing my energy on my new game.

    Once again, thank you for the feedback – you have no idea how much it means to me :)

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.