2015 was a year of big changes for us. Previously, we'd been a hybrid company doing a combination of web development and DevResults. At the start of 2015, we officially closed the consulting branch of the business to focus exclusively on DevResults. With this identity change, we experimented with changes to a lot of processes and workflows that we'd been itching to fix but hadn't had the time or humanpower to attack, testing out various approaches and trying to quickly apply lessons learned. We were able to generate a lot more momentum around the software and roll out more changes with this added focus.
And this has led to a small change in our own mindset about the software, too: we aren't just selling software. We're focusing on our top goal of supporting excellent data management practices in the international development community. You entrust us with your data and partner with us to try to solve some of the most difficult problems facing your organizations. This has prompted us to think a lot more about the software holistically: what it currently does, how you currently use it, what other problems you'd like us to solve, whether those are problems we can/should solve, and most importantly: if we can and should, how do we get there?
As part of those changes, Leslie, Josh, and I have laid out a high-level vision for new software features/functionality for the year. Once we have that yearly roadmap a little more fleshed out, we'll add it to our public website, so you can always have a sense of where we're going. This blog post is just a sign of things to come, but it goes beyond just software to tell you a bit about what we're expecting to change with people, processes, and product. (Sound familiar? It should if you read my 2015 Retrospective post!)
All the awesome changes we rolled out on the process side of the shop have made us feel confident that we can and should add to our team. This year, we'll be adding staff on both sides of the house:
1 Data Science
We are in the midst of interviewing for an additional data hire to help handle some of Leslie's and my workload. While many of our 2015 process changes have given us a bit more time, we're both stoked to have a third person to help tackle new client setups, trainings, and help material creation. We're both planning on semi-specializing some of our own time in the process: Leslie will be architecting the addition of analysis tools to the software, plus further systematizing new client setup. I am now the Director of Product Development, so a lot of the nitty gritty details about what the team works on each month and how we break up larger projects will fall on me. We have all kinds of plans for more training and help materials (see the Process section below), and we're excited to have a third person to help see those plans to fruition.
We are also looking to hire at least one more engineer so that we can get more of the changes you want faster. If you or someone you know might be interested, check out the job posting on our site and on Stack Overflow.
Several of the processes we put into place in 2015 have worked out very well. Here's what we plan to continue to change or add in 2016:
1 Customer Issue Reporting
There's not much we're changing here; we did most of the hard work in 2015 getting the ticketing platform live and working out internal workflows around who handles which types of tickets. An on-call engineer will be the first one looking at most tickets and will involve a data scientist for feature requests, complicated examples or use cases, etc. We, in turn, will stay out of the engineers' way so they can keep fixing your bugs quickly. If you have ideas or suggestions on things you'd like to see us do differently around bug/issue reporting, please let us know!
2 Client Communications
In the past, most often clients communicated directly with Leslie or me, and most of that communication only occurred when you initiated it. This is a pretty reactive way to handle communication, and it tends to keep us more in a software troubleshooter role rather than a solutions provider or partner role. The fact is, we want to be providing solutions. Leslie and I would much prefer to talk to you about more of the intangibles and the struggles you're facing while we let the engineers deal with a random error message you've gotten in the software. To keep us focused on this goal of helping you solve your larger problems, we're starting two new communications processes:
- Client Survey Interviews: Informally and anecdotally, we've always collected information about clients' experiences and opinions about us and our software, but we believe that being more systematic about it is useful. We debated doing client surveys, but in the end it seemed like interviews would be the best way to collect this data initially. And we wanted someone who'd had no previous relationship with you, so you'd be comfortable being as honest as you like. For those of you willing to participate, you'll be interviewed in the first half of 2016 by a short-term hire who will conduct the interviews and then aggregate that information for us, so we won't necessarily know which client said which bit of feedback. Our goal is to get a better sense of cradle to grave on DevResults adoptions, particularly the things that we'd never know by simply looking at site usage statistics: why folks picked us, what setup was like, how adoption has been, how we have or haven't solved the problems you initially had, what kinds of culture or process changes you've undergone as a result of the adoption, etc. If you have not yet been contacted but are interested in providing feedback, please email me. Right now we're seeing this as a one-time effort, but we'll see how it goes!
- Quarterly Check-ins: Leslie and I are also trying to schedule a detailed check-in call quarterly with each client to review what your top pain points/feature requests are, where we're at with our roadmap and plans, and any functionality or use case questions you have for us. You're certainly not limited to these quarterly check-ins, but our goal is to make sure that we're talking with you even when you aren't reaching out to report if something's wrong. This helps us keep a better sense of priority on the things that matter most to you that might not lend themselves well to help tickets, and it gives us time to deepen our knowledge of your organization, workflows, etc.
3 Communicating about Software Changes
In 2015, we began to write blog posts when new features or changes went out. In 2016, we are increasing that communication even more: for truly huge overhauls or new features, we would like to make these available in advance in a mirror copy of your site so that you can test them out, update your own documentation and workflows, and give us feedback about what you like or don't like. We see this as two-fold: it keeps you more up-to-date and able to adjust to changes, and it gets us a lot more feedback on new features from a group of beta testers before we go live, so we can catch hidden bugs or make minor cosmetic tweaks.
To start with, we've added all site owners to an email group to get notified of these changes so you can test them out. We'll be emailing that group with more details once we're ready to try this for our next major feature.
4 Ongoing Training Needs
Our training model when I started was that we'd get a site built and then go on-site with a client for a set amount of time to train a group of people on how to administer the site. This works well, as far as it goes, but we've noticed a high amount of churn in key points of contact as awards end, management changes, etc. And we also know that paying for an entirely separate round of training can be cost- and time-prohibitive. So we're working on some new offerings to address those needs:
- Expanded Help Materials: Our vision for the help materials is to slowly provide more carefully curated and guided curriculum so that anyone you hire anywhere in the world could do a Teach-Yourself-DevResults course. Initially this will probably be a sequence of suggested help pages and exercises, but we're looking to greatly expand the conceptual pages and videos we began last year to address these needs more holistically.
- Standing Trainings: To help address cost issues, we're also planning to offer open-seat DC trainings about once a quarter. You'll be able to reserve seats at these trainings and can send your staff as needed, either as brand-new trainees or for refresher training. We are hoping this more on-demand offering gives you the flexibility to get new staff quickly up to speed with in-person, hands-on training if they want it. We'll be posting full details once we have the logistics of this worked out and can begin offering it!
5 Additional Consulting
Though we spent 2015 trying to focus exclusively on DevResults, we continue to see a need for focused data management-related consulting in this space. We are now offering additional implementation services (i.e. business process consulting, data management consulting, etc.) to help meet this need. We're still in the early stages and we'll keep you updated as we solidify our offerings.
We have several large features in the works or planned for 2016. This year, we are focusing on features that more seamlessly fit with your workflows, provide more visualization options, provide more granular data management, or help make DevResults perform faster and look a little better. Here are the biggest efforts.
1 Pito & Visualization Overhaul, Dashboard Overhaul
- Status: In Development (Pito & Visualization Overhaul); Planned (Dashboard Overhaul)
- ETA: Q2/Q3 of 2016 Calendar Year
One of the most frequent requests we get from clients about new functionality has been visualizations. Many of you want more visualizations throughout the site and you want to be able to better configure/filter what you see in those visualizations. We will likely roll these changes out in several waves, to try to get you new functionality as quickly as possible:
- People like looking at multiple indicators side-by-side for easier comparison. We plan to enable additional visualization options in Pito so that you can see visualizations for multiple indicators across activities. (A bit like the Activity Performance tab, but with multiple activities!)
- Speaking of the Activity Performance tab, many of you have asked for more filters/sorting/searching and more visualization options here. We have plans for both of these requests, though they are dependent on the initial Pito & Visualization Overhaul.
- For indicator visualization, we plan to have a version of the Indicator Targets vs. Actuals chart that is much more configurable. This should include the ability to view the actuals split out by disaggregation, show or hide targets, select whether you want to see program vs. activity targets, and so on. This change should be included with the Pito & Visualization Overhaul.
- We plan to add filters to as many widgets on the dashboard as we can so that you can better customize what you're seeing to specific activities, indicators, etc. This will be part of the second effort here, the Dashboard Overhaul.
- One of our longstanding dreams has been to enable more than one dashboard per user and make it easy to share those dashboards with other users. So, for example, as a site owner I might create a dashboard for the executive committee and then share that dashboard with those users.
2 Disaggregated Targets
- Status: Planned
- ETA: Q2/Q3 of 2016 Calendar Year
DevResults, as I like to tell so many of you, does a good job of forcing you to compare apples to apples by enforcing consistent indicator definitions. This effort will work to extend that consistent granularity to indicator targets. At the same time, we know that sometimes targets are less granular than actuals, so we plan to have an interface that lets you configure how granular you want your targets in relation to your actuals and enter them accordingly.
3 Faster Performance & Cleaner Pages
- Status: In Development/Ongoing
- ETA: By end of Calendar Year 2016
Before I started at DevResults in October 2014, we made a decision to rewrite every page in the site to make them perform faster and be more user friendly. We have had some nice wins with pages that have been rewritten in the new style (like the new Index pages), but we still have a lot to do. We will continue to rewrite individual pages around the larger work outlined here and our usual on-call support. It is not fast or particularly glamorous, but you can bet I'll be touting page improvements in upcoming blog posts.
4 Enterprise 2.0 & a New Data Publishing Model
- Status: In Development
- ETA: Q2/Q3 of 2016 Calendar Year
One of the key trends in our industry is integrating different systems with each other. We have been working to get all of DevResults covered by our Application Program Interface (API) so that clients with more technical resources could do more to integrate DevResults with other systems and automatically update/transfer data. The goal here is to make submitting data to DevResults from any outside system a straighforward process (and to streamline the process of submitting data between DevResults sites, too).
At the same time, we are overhauling the functionality for our enterprise customers to automatically report data from child sites to their parent/enterprise site. We'll be using our own API behind the scenes to handle most of this (hence why I'm talking about them in the same section). In addition, enterprise users can expect a lot more diagnosis and helpers for linking child indicators to parents. I've spoken with most of our enterprise clients directly on this, and I'll be blogging about it more once we get past the API work and start making visible site changes. The goal here is to improve our enterprise functionality's user-friendliness and helpfulness so you can set up linkages faster while having to think less.
5 Reporting Cycles
- Status: In Development
- ETA: Q2 of 2016 Calendar Year
As DevResults currently exists, if you want to vary which indicators an activity reports on by reporting period, you can only do this at the Results Framework level. But we know that, often, you'll have indicators that should be reporting annually in the same section of a Results Framework with those reported quarterly. This makes total sense, and we want the software to better support this arrangement while keeping things as simple as possible for your end-users. Our solution is currently under development: reporting cycles. Here's a quick summary of what the change will look like:
- Define one or more reporting cycles for your site and assign reporting periods to the appropriate cycle.
- Define a default reporting cycle for each indicator as part of its definition.
- When you assign an indicator to an individual activity, you can keep that default reporting cycle or manually override it for that activity.
When the system generates the reporting grid and template for that activity's reporting periods, it will check the activity-indicator assignment to see which reporting cycle it should use and will only display indicators whose cycle matches that of the reporting period. You'll still be able to customize reporting periods based on results framework, too, but this should eliminate a lot of the need for that work.
What you can expect from us: we intend for this to be the first major feature released in the beta sites for testing, so you'll have time to kick the wheels and give us feedback. We'll also be automatically doing the initial setup and configuration for you based on information already in your DevResults site. Our goal is to have you largely set up to use this without having to do any work on your end beyond a few data quality checks for accuracy.
Note: These settings will only be enabled once you have more than one reporting cycle in your site.
6 Data Table Expansion
- Status: Planned
- ETA: By end of Calendar Year 2016
One of the biggest trends we're noticing with our clients is that more and more organizations want to be collecting very granular data. We believe granular data is where the industry is going, and we have every intention of providing you the tools you need to stick with that trend. The planned changes are a two-pronged approach:
- Improve the User Interface: we'd like to expose search and filtering in the data tables so you could do more with them online in real-time. We'd also like to give you more ways to interact with data table indicator data once it's been calculated in Pito, visualizations, etc.
- Basic Analysis Tools: This year we would like to start to make the shift from being just a project/performance management system to offering some analysis of data table data. We're still working out the details of this part of the roadmap, but the data science side of the shop is extremely excited about this!
As you can see, there's a lot we hope to accomplish this year. This is our first time making our development plan publicly available, so there may be some hiccups along the way as we have to reprioritize due to customer needs, industry changes, and the like. Some might ask: why make it public at all if you can't 100% guarantee delivery by each date? My answer: isn't this what all of you do when you set targets for a year and try to work toward them? Sometimes unforeseen circumstances prevent you from getting there; sometimes you just barely meet those targets; and sometimes you knock it out of the park. We want you to know what our top priorities are for the year and to let us know how these do (or don't) align with your goals and requests.
We know it isn't easy working in this industry, with its ever-changing set of standards and best practices, where terminology isn't always consistent from one organization to the next, let alone process and workflows. They aren't easy problems to solve; making the world a better place never is. But, like you, we are constantly working to try to meet those fluctuating needs. Thank you for your belief and trust in us to solve the problems that most impact your organizations' data. Thank you for continuing to believe in us by continuing to remain our clients and partners, and here's hoping 2016 is a wildly productive year for all of us!