Most of the systems of management and optimization of advertising campaigns aimed directly at professionals. For project managers, brand managers and marketing managers who definitely are interested in the results of the campaign, but not engaged in operational work, such a system is hard to be adapted. Not so long ago we wondered, to make a separate dashboard for managers where show only the data that they need. And in December 2014 we announced the launch of the updated AdTrack, in which this idea was implemented.
Now time to test the tool and to understand how successful the idea is, we want to tell have been through, and I learned how to visualize complex data. A few years ago in the industry of Internet advertising the concept of performance marketing. This approach is aimed at achieving a particular, useful for business, and most importantly, measurable results. For the online shop for the purpose of this marketing plan will be an increase in the number of orders is increasing brand awareness for financial institutions — increase the number of attracted clients instead of improving the perception of the Bank as a reliable.
Modern performance-marketing — is primarily a complex and technological industry. The achievement of the desired result to the client is impossible without daily and hourly measurement of results of advertising campaigns. Optimizing rates and other settings of advertising placements, we are improving performance, achieving maximum business performance client. Moreover, these campaigns often have a large amount of data — tens and hundreds of thousands of ads, millions of keywords and dozens of targeting combinations.
All this means that manual analysis and management of advertising impossible. These actions will be unacceptable to leave a lot of time. We had a choice. To use already existing solutions for optimizing and managing advertising campaigns or to create your own tools.
While we have tried to use different systems, but eventually came to the conclusion that it is necessary to develop a solution that could be possible to adapt to the business of our clients. AdTrack we created in 2010 and developed it for more than five years. A few words about what the management and optimization of advertising campaigns on the market. For simplicity, divide them into several groups.
First, it can be a Western solution (Adobe AdLens, Kenshoo, Marin, Acquisio, and so on), but you need to be prepared for the fact that Western systems do not support some Russian channels (in this case still have to use a local solution). Secondly, depending on their needs, advertisers can choose the system to work only with one channel (for example, with contextual advertising — eLama, Garpun, K50, social media — Plarin, Control) or a system that will allow you to manage multiple channels — Adobe AdLens, AdTrack, MarinSoftware, Marilyn. Such systems are typically aimed at large advertisers.
Third, an individual decision there are some agencies AdTrack, Adhands, “Context Motor”. As mentioned above, your own solution allows agencies to target the automation tools under those market segments in which they specialize. Creating AdTrack, we wanted to kill two birds with one stone.
Get the tool for internal use and provide customers access to the course advertising campaigns. At first everything was fine. System solved our challenges for management and automation of advertising campaigns and at the same time solved the problems of customers. As time went on, analysis, and management grew in response to the challenges of a constantly evolving advertising market.
Within the Agency no campaign was not conducted without the use of AdTrack, and among the clients activity, on the contrary, decreased. For them, the system was complicated, moreover, many were not interested in how we work with advertising campaigns, they needed results and dynamic is we have seen the number of requests for reports on the placements. In the end, we spent a lot of time on reporting, although clients can see for yourself the results. In addition, giving them access to advertising campaigns, we wanted to make our work as transparent as possible and thereby increase confidence of clients to our Agency.
But we realized that the task of providing information to clients AdTrack solves inefficiently. We tried to conduct a small master class, which told about the possibilities of the system and made instructions for use. After that began to track the entries under client login and password — they still had too little, and requests for records Manager continued to regularly receive. Master class didnt help.
Opportunities of Internet advertising is constantly evolving, the data became more and more, AdTrack successfully coped with new challenges, but it interfaces with all was more like the universe — and the customers needed a simple and handy tool that would enable monitoring of key parameters. For example, heres what they saw, by accessing your login at AdTrack.
How to solve this problem and save the interface, which is used by our experts. We realized that customers needed a separate dashboard (information panel). But to make a really good product, and this one is for customers, not for yourself, you need to think like customers, and clearly understand their needs. To identify these needs and visualize information in an accessible form for clients, we turned to design thinking — the approach that allows you to work with “implicit knowledge” (tacit knowledge) of the consumer.
At the heart of this approach is empathy — the ability to look at the world through the eyes of others, understand their needs, desires, and goals. We tried to put ourselves in our clients shoes and understand what dashboard will be useful to them. The development process itself, according to design thinking consists of five stages. Empathy, focus, ideas, prototype, test.
We had a problem but we decided to dig deeper and immerse yourself in the customer workflow. For this we conducted several interviews with advertisers on how they work with these campaigns and what features they lack. Received the following points (interestingly, the range of needs of our customers is quite wide).
When we moved on to the focus, we needed to piece together all heard and seen to understand the overall picture and to identify the most important parameters. To simplify this process, we tried to visualize the information. Wrote out all nine points, obtained in the initial interview, on the Board and tried to draw links between them. We realized that we have a looming three main problems.
To deal effectively with these problems, we decided to dig to the target from two sides, hence two focus groups. One technological leader, the other with marketing. Before that, we prepared several variants of draft picks prototypes, based on customer needs and problems, and showed them to each group. During the demonstration, draft dashboards we have given his explanation, answered questions of survey participants (and recorded for themselves, what exactly they cannot understand or dont like).
We tried to understand how each group sees the solution outlined in the beginning needs. We had a few brainstorms. We listened to comments and tried to understand what I can do, solved a problem and came back with new ideas. In the end, we came to understand that the dashboard should display the following data.
Heres what else you must realize. So, the ideas we had, and we had to test them on clients. Because originally the needs were very different, we didnt just one prototype.
There were several stages in which we showed the prototype to customers, asked their opinions, made changes in accordance with comments and showed a new variant prototypes. Such visits were. We have prepared the first version (page “General stats” and “Statistics by segment”) and discussed them with those who have already seen AdTrack, and those who didnt work with it.
Each participant of the process it was necessary to choose blocks that he likes and doesnt like, and justify your choice. Option 1. We have shown clients the first option of dashboards.
All data that customers wanted to track were presented on the prototypes. Intentionally dwell on what was not liked by the customers because for the rest of the blocks they, as a rule, said that “all is well” and “nothing.”. General statistics. Segments.
Option 2. We tried to consider and correct all customer comments. We showed customers a second version of prototypes, and heres what they didnt like it this time.
Interestingly, many blocks we have left unchanged, and only in the second viewing clients understand that it is inconvenient.
Option 3. This time was made a lot of work. Because customers didnt like the jumble of data and symbols, you had to figure out how to present these indicators otherwise. In addition, we have a few of their edits — and, as usual, showed customers what happened.
This was the last approach to the prototype development, because of requests or comments from customers we have not heard. So, what and how we altered. We had to completely rework the bottom of the block of data was really very much clear and try to present them in a single small window is impossible.
Now all the data were presented in tables. So far have done a great job. We found out the preferences of our clients know why they dont like certain points to design.
Although we were almost at the finish line and we already had an understanding of how it should look, in practice all has appeared not so simply. First looked at what others are doing, chose a few successful examples (good graphics, button icons, status), have discussed these options with clients to understand what they like and why.
And here are some unsuccessful, in our opinion, examples. We knew that the design should be modern, stylish and lightweight. And if one blocks everything was clear, the other had to Tinker.
For example, I had to try several versions of the representation graphs.
Long worked with tables. Realized that aggregate statistical metrics and performance indicators these are large figures, and to make them easily readable, you need to post in different tables. In this case, the information should not be lost. Again there were several calls in which customers give their comments.
What do you like and what not. For example, we have divided the page into two blocks, the data to be easily seen. For quick navigation on the dynamics of the auxiliary die tables made in color.
It was the most controversial blocks. After we have taken into account all the shortcomings, has launched dashboards in this form. We got two interfaces.
For managers and professionals. It remains to test new dashboards to clients. We deliberately decided not to give access immediately to all customers, and limited the number. Now that we got the results of the test.
At this point, we figured what tasks dashboard does not solve. For example, some customers do inside the campaign statistics according to various marketing channels, including offline. Managers still have to provide statistics on such placements as an Excel file or table specified by the client format. We have been developing and promoting AdTrack.
We cannot say that the implementation of new functionality always passed us easily and smoothly — but, honestly, the creation of dashboards for managers we spent more time and effort. We are happy with the result. We have tested the effect of design thinking, learn to listen and speak with customers clearly identified their needs (even if it was a long and not always easy), have discovered exactly what functions and in what form it to them. In other words, we have obtained a tool which is already used by our clients and not only we.
And continue to consider their wishes. Now we check the two ideas that are at the stage of creating dashboards were postponed for later. On the one hand, we see the interest in this function because we all love to talk about good results.
On the other hand, the results of campaigns can be business secret, and therefore share success will be impossible. Now we study this question in more detail to find a balance between the needs of. Actually, this idea has already found application in practice. For clients who are driving Analytics for all advertising activities within own system, desired data from AdTrack.
For them we made a special API, through which they can get campaign results and use them in your reporting systems. The devil always lies in the details. You can have a brilliant idea (and everyone will think its brilliant), but the implementation will become an insurmountable barrier. And after this start all around, including developers, will be puzzled.
Why this brilliant idea didnt “take off”. In the project of reorganization of the interface AdTrack team managed to beat “the devil”, because it is at the level of reorganization of parts of the project has been the turning point in obtaining a positive user experience. Such a victory could provide design thinking — an approach that not afraid to look into the eyes of the user problems, which quickly reveals all the schools of development, may at first sight insignificant.
The case shows very well how the design thinking and empathy in practice mode improvements “here and now”.